"Now, it's time for the happy recap." - Bob Murphy
Poll Analysis Archives
January 23, 2017
POLITICS: Do Democrats Really Want Their Own Tea Party? Be Careful What You Wish For
January 17, 2017
POLITICS: The Real Reason Trump Won: Part 4 of 4
NRO: The Real Reason Trump Won: Part 4 of 4, looking at where Trump underperformed other Republican candidates.
January 12, 2017
POLITICS: The Real Reason Trump Won: Part 3 of 4
NRO: The Real Reason Trump Won: Part 3 of 4, looking at each side's state-by-state turnout as a percentage of the Voting-Eligible Population.
January 11, 2017
POLITICS: The Real Reason Trump Won: Part 2 of 4
NRO: The Real Reason Trump Won: Part 2 of 4, looking at how Trump performed in the battleground states by historical standards for post-incumbent challengers.
Posted by Baseball Crank at 4:00 AM | History | Politics 2016 | Poll Analysis | Writings Elsewhere
January 10, 2017
POLITICS: The Real Reason Trump Won: Part 1 of 4
NRO: The Real Reason Trump Won: Part 1 of 4, looking at how Trump performed nationally by historical standards for post-incumbent challengers.
Posted by Baseball Crank at 4:00 AM | History | Politics 2016 | Poll Analysis | Writings Elsewhere
December 21, 2016
POLITICS: Hillary Clinton: President of California
POLITICS: Hillary Clinton: President of California
Posted by Baseball Crank at 2:47 PM | History | Politics 2016 | Poll Analysis | Writings Elsewhere
December 8, 2016
POLITICS: New York's Vote Was A Microcosm of America in 2016
December 3, 2016
POLITICS: The Electoral College Is Not Like Slavery
November 21, 2016
POLITICS: Hillary Voters: The Minority in 37 States
November 16, 2016
POLITICS: Does Donald Trump Have a Mandate?
November 9, 2016
POLITICS: A Republican Win, Not Just A Trump Win
November 8, 2016
POLITICS: How Will The Senate Break?
POLITICS: My Final Prediction: Hillary 303, Trump 235
May 25, 2016
POLITICS: The Dog That Didn't Bark: Trump Voters in Down-Ballot Primaries
May 11, 2016
POLITICS: A Very Different Republican Coalition: Can It Fly?
May 4, 2016
POLITICS: Trump's Next Victim: Pollsters
April 19, 2016
POLITICS: What To Watch For In Tonight's NY Republican Primary
April 8, 2016
POLITICS: After Wisconsin, By The Numbers
March 24, 2016
POLITICS: Can Donald Trump or Ted Cruz Beat Hillary Clinton? A New National Poll May Surprise You
March 17, 2016
POLITICS: Ted Cruz or Bust: Armageddon Tuesday By The Numbers
March 15, 2016
POLITICS: This Is Not 1980, And Donald Trump Is Not Ronald Reagan
Posted by Baseball Crank at 10:20 PM | History | Politics 2016 | Poll Analysis | Writings Elsewhere
March 9, 2016
POLITICS: It Is Time For Marco Rubio To Join Ted Cruz For The Benefit Of Marco Rubio
March 3, 2016
POLITICS: Marco Rubio's Path To Victory After Super Tuesday, By The Numbers
March 1, 2016
POLITICS: New CNN Poll: Trump Loses To Hillary, Rubio & Cruz Would Win
POLITICS: A Vote For Trump is a Vote For Hillary Clinton: Why Trump Is A Sure Loser
February 27, 2016
POLITICS: Marco Rubio Gives Democratic Strategists The Vapors
February 25, 2016
POLITICS: Second Florida Poll Of The Day Shows Rubio Closer To Trump
POLITICS: New National Poll of Hispanics: Rubio Popular, Trump Hated
February 22, 2016
POLITICS: South Carolina By The Numbers
February 20, 2016
POLITICS: Exit Polls From 2012, 2008 and 2000 Tell Us A Few Things About South Carolina
February 19, 2016
POLITICS: New South Carolina Polls Show A BIG Surprise
February 16, 2016
POLITICS: New South Carolina Poll From CNN: The Race For 2-3-4
February 14, 2016
POLITICS/LAW: Scalia and South Carolina
Posted by Baseball Crank at 12:31 PM | Law 2009-Present | Politics 2016 | Poll Analysis | Writings Elsewhere
February 9, 2016
POLITICS: New Hampshire Primary 2016
February 2, 2016
POLITICS: Now With More National Review
As a reminder, you can follow me on Twitter @baseballcrank or bookmark these links to catch up on my latest work:
and now at National Review
Latest since my last post here:
Posted by Baseball Crank at 5:26 PM | Blog 2006-Present | History | Politics 2016 | Poll Analysis | Writings Elsewhere
December 18, 2015
POLITICS: Bomb Aladdin!
July 10, 2015
BLOG: Welcome Back, Blog!
I've been neglecting this blog rather badly for altogether too long - the archives say I haven't posted here since September 21, 2014. I've been busy in the interim on Twitter, of course, and publishing elsewhere. I probably need to post archived versions of some of those posts here. For now: links.
I will start with The Weekly Standard, where I have this issue's cover story, just posted today: Giving Thomas His Due, on Justice Thomas' opinions over the past year and what they tell us about his philosophy.
Then there's The Federalist, where I tend to post my longer essays these days. I ran a lengthy 5-part essay prior to the Obergefell decision, "Can Gays And Christians Coexist In America?". Part I looked at the Biblical reasons why Christians believe in one-man-one-woman-for-life marriage. Part II looked at the history of Catholicism and other Biblical Christianity in the battles over slavery and Jim Crow. Part III looked at the Christian concept of scandal and the battle between liberty-based and equality-based views of "LGBT rights." Part IV looked at the legal arguments over the rational basis for distinguishing between opposite-sex and same-sex marriage. And Part V traced possible ways forward for coexistence post-Obergefell, which admittedly are not looking especially promising at the moment.
The First Principle Of U.S. Foreign Policy looked at various approaches to our foreign policy.
Others from the fall, including some of my poll-analysis posts:
Polling Postmortem: The Best And Worst Senate Polls Of 2014 (I keep meaning to run the companion piece on the Governors races before 2016 polling heats up).
Do Democrats Always Win Close Statewide Elections? (covers the 1998-2013 elections; I should update this with 2014 results).
And of course, if you missed it last time, my essay on how History Is Not On The Democrats' Side In 2016 is still an important read on the coming election, undoubtedly the most significant piece I will write on the 2016 election.
The Rise & Fall of the Confederate Flag in South Carolina - I wrote this a few weeks back, but it's very relevant to today's news.
Reading Tea Leaves on the 2015 Supreme Court Term - Basically just some educated speculation on who would write what and when, which ended up having mixed results.
Democratic Party Now Literally Selling Hate - a Father's Day gift post!
Bernie Sanders, Deodorant and Diversity - a meditation on central planning and markets.
Marco Rubio Recounts The History of Obama’s Treatment of Israel - quick hit on a great Rubio floor speech. Rubio isn't my first choice in 2016, but he's done nothing but impress this year.
From the fall:
2014 and Republican Morale - a GOP victory lap and a reflection on what it meant.
The Breakers Broke: A Look Back At The Fall 2014 Polls - A personal victory lap on my 2014 poll analysis and how it relates to the polling controversies of 2012.
The 2014 Polls And The 2012 Exit Polls - An earlier look at the same topic and at some specific issues with exit polling and poll methodology.
BREAKING: Supreme Court Takes Obamacare Subsidies Case (on King v Burwell).
First Cut: 7 Polling and Elections Lessons From 2014 (Immediate 2014 election aftermath)
Why I Voted Yes On Question 1 (NY) (Election Day post on a NY ballot initiative)
A Sad and Desperate Attack on Chris Christie - Actually a fairly deep dive on voter fraud controversies.
Introducing The Senate Breakers Report - September 26, 2014, the start of my Fall 2014 stretch drive when I started getting too busy to cross-post here.
Posted by Baseball Crank at 9:22 PM | Blog 2006-Present | In Print | Law 2009-Present | Politics 2014 | Politics 2015 | Politics 2016 | Poll Analysis | Writings Elsewhere
September 19, 2014
BLOG: RedState and Federalist Roundup
I owe longtime readers here some explanation and apology - my work at both RedState and The Federalist is now exclusive, at least when first published, to those sites, and while I post links on Twitter and Facebook, I tend to forget sometimes to post links back here at the old stomping grounds. (I may well close the comments section here too soon, since the lack of activity means a high spam-to-real-comments ratio, and since most regular commenters by now know how to find me elsewhere).
Here's my most recent posts over the past month, all of them on matters of politics and/or history:
Where I Was On September 11 (a repost of the annual remembrance)
Posted by Baseball Crank at 3:28 PM | Blog 2006-Present | History | Politics 2014 | Politics 2016 | Poll Analysis | Comments (1)
September 15, 2014
POLITICS: Mid-September Polls Are Not The Last Word On Senate Races
A Snapshot, Not A Verdict: Will A Wave Still Swamp More Democrats?
The perennial question about election polls is back again, if ever it left: how far can we trust them? Should we disregard all other evidence but what the current polling of individual Senate races tells us - which is, at this writing, that if the election was held today, Republicans would gain 6 seats in the Senate to hold a narrow 51-48 majority? As usual, a little historical perspective is in order. It is mid-September, with just over seven weeks to Election Day, and as discussed below, all the fundamental signs show that this is at least a mild Republican "wave" year. A review of the mid-September polls over the last six Senate election cycles, all of which ended in at least a mild "wave" for one party, shows that it is common for the "wave party" to win a few races in which it trailed in mid-September - sometimes more than a few races, and sometimes races in which there appeared to be substantial leads, and most frequently against the other party's incumbents. Whereas it is very uncommon for the wave party to lose a polling lead, even a slim one, after mid-September - it has happened only three times, one of those was a tied race rather than a lead, and another involved the non-wave party replacing its candidate on the ballot with a better candidate. If these historical patterns hold in 2014, we would therefore expect Republicans to win all the races in which they currently lead plus two to four races in which they are currently behind, netting a gain of 8 to 10 Senate seats.
Read More »
The Mid-September Polls: Still Waiting For The Wave
When I last reviewed the Senate races in late June, the picture we saw was that Republicans had largely locked up three races for open seats currently held by Democrats - in Montana, South Dakota, and West Virginia - and locked down the one GOP Senate seat in deep-blue territory, in Maine. That left nine heavily contested seats, mostly in red or purple states (plus blue Michigan, the toughest race) and four other purple or blue state races in which the GOP had not yet become competitive, but retained hopes of bringing the race into its sights:
As I noted at the time, this left Republicans "waiting for the wave" - hoping that fundamental factors like low Obama job approval ratings and the shift from registered-voter polls to likely-voter polls would show a general, across-the-board movement that would tilt the field just enough in these races to pull Republicans ahead.
The battleground at present, as determined from this morning's RealClearPolitics polling averages, is only slightly different - the Democrats are still in a good deal of trouble, but the lack of major movement would seem to suggest that a "wave" has yet to surface:
As before, I'm looking at head-to-head polling between Bill Cassidy and Mary Landrieu in Louisiana, since the main third candidate, Rob Maness, is likely to draw votes away from Cassidy but won't help Landrieu avoid a runoff, so the more interesting question is who would win the runoff. The three locked-in pickups still look locked in, and Maine is still off the table. The two longest shots, in Oregon and Virginia, are not competitive - in fact, Republicans are actually closer now to making a race of Illinois and New Jersey than Oregon and Virginia (although both Dick Durbin and Cory Booker still have double-digit leads in traditionally blue states, so we would need to see further poll movement before declaring either of those states to actually be competitive). New Hampshire and Minnesota have tightened, but are not yet in the "hotly contested" bracket, while the GOP has fallen further behind in Michigan. In the Republican-held seats, David Perdue has opened a 3-point lead in Georgia and Mitch McConnell has gradually pulled to about 5 points ahead in Kentucky - a good lead, but no better than what Gary Peters in Michigan and Jeanne Shaheen in New Hampshire have on the Democratic side. Thad Cochran is well below 50, often dangerous territory for an incumbent, but he still has a 14 point lead and it's still Mississippi, so he is even less likely to end up in a tight race at the end than Durbin or Booker. The wild card for now is Kansas, where the Democrats have thrown their support behind "independent" Greg Orman against the suddenly vulnerable Pat Roberts but have failed in their bid to get their own nominee, Chad Taylor, thrown off the ballot. It will be at least a week, maybe two, before we have a critical mass of polling in that race that post-dates Taylor's attempted withdrawal. All of the remaining races feature leads of less than 4 points, four of them leads of less than two points, and in only three of the contested but not locked-up races is the leader at 47% or better (McConnell, Shaheen, and Al Franken). On the surface, therefore, what we have is a dogfight with a whole bunch of races that could still go either way.
The Fundamentals: The Tide Still Favors Republicans
But the recent history of Senate races says that a whole bunch of races will not go either way - it is more likely that a whole bunch of races will all go the same way, in the Republican direction. There's been a lot of discussion about the meaning of the term "wave" election. Political scientists use their own somewhat precise definition of a "wave" election, and Sean Trende has reviewed the arguments under that rubric. If you define a wave in terms of political spin - a large enough election outcome to draw conclusions about some sort of voter mandate - then you need to set the bar and measure the final results against them. For those purposes, I would say the GOP has had a successful year if it picks up at least 7 Senate seats, net, and can claim a real "wave" if it picks up 8 or more, which would require it to not only run the table in Romney 2012 states (including the narrowly won North Carolina) but also break through in some of the Obama 2012 states like Iowa, Colorado, New Hampshire, Michigan, and/or Minnesota.
But for my purposes, I will use a simpler functional definition of a "wave," one that covers each of the last six Senate election cycles and seems, from all the available evidence, to fit this one as well: a party has a "wave" election when the President's approval ratings and generic Congressional ballot both favor it, it gains seats on net in both the Senate and House races, and it loses no more than two Senate seats it held going into Election Day. As we can see, by that definition, we had a Republican wave in 2002, 2004 and 2010, and a Democratic wave in 2006, 2008 and 2012, and should expect a Republican wave again in 2014:
The only slight exception to the across-the-board wave dynamic in those elections is that the GOP trailed in the generic House ballot in mid-September 2004 and ended up tied on Election Day. And we see the same trend now: President Obama's approval rating is double digits under water, and Republicans have opened a growing lead in the generic ballot.
If you project the election results just from these national fundamentals, you could take the view explained by Trende that the Senate races should mostly track the President's approval rating and that Democrats' current poll performance in the generic ballot and the Senate races may be close to its ceiling because Democratic candidates have locked up basically all the voters who approve of the President's job performance, leaving a pool of undecideds who overwhelmingly disapprove of Obama and are thus likely to break against his party.
But what about the polls themselves? It would seem dangerous to disregard them. Josh Katz at the New York Times' Upshot blog looks at some recent history and concludes that poll-based forecasts, while more accurate than fundamentals-based forecasts, are less accurate than forecasting models that incorporate both polls and fundamentals. Nate Silver looks at the individual polls in Governor, Senate and House races, and finds a significant amount of polling error even in the last three weeks of a campaign - with October polls in Senate races consistently off by an average of 4 or more points. 4 points is not a big deal in an uncompetitive race, but when we're talking about races where the lead is 4 points or less, it's a very big deal. Silver, however, confines his review to individual polls, and not averages:
It’s important to note that the accuracy of the average poll — what these figures describe — is not the same thing as the accuracy of the polling average. The polling average will cancel out some of the errors from individual polls provided that the misses come in opposite directions.
Let's take a look, then, and see specifically how the final results compared to the polling averages as of mid-September for the last six cycles. For all of these, I will use the RCP average; I noted with an asterisk where I had to do my own estimating of an average (this was too pervasive to bother marking off in 2004 and 2002, as the averages were less sophisticated and available). The sample size is still not huge, but it's more than just a couple of elections: I was able to track down polling in 87 at least moderately competitive races followed by RCP, out of more than 200 Senate elections in that period. Those 87 races include all the races that were seriously contested except for the three races won by a candidate running as at least a nominal independent in a three-way race (Angus King in 2012, Lisa Murkowski in 2010, and Joe Lieberman in 2006), which I excluded because those races tend to defy easy D vs R classification. We have a good cross-section of Senate cycles to work with here - every Senate seat came up at least twice, we have three general and three midterm elections, three Republican waves and three Democratic waves, three elections when the President was popular and three when the President was unpopular. (What may be harder to quantify with the data we have is the extent to which the predictive value of polls has been affected by the expansion of early voting in many states over this period to allow votes to be cast well before Election Day, but I do not believe any state allows early voting in September).
2002 to 2012: A Senate Odyssey
Working backward, let's start with the 2012 election. I include here both the mid-September (where available, September 15) RCP average and the final RCP polling average, so you can compare how much of the accuracy or inaccuracy of the mid-September polls may be due to polling errors as opposed to subsequent movement in the electorate (the "poll error" column shows how far off the final poll average was from the result). The "2-Party Wave" column shows the swing in the 2-party vote from mid-September to the final election result. Thus, for example, a candidate who led 46-39 in mid-September is getting 54% of the two-party vote; if that candidate wins 51-49, she would be down 3% in the two-party vote. Another way of putting it is that a 2.5 point swing in the 2-party vote is, more or less, enough to wipe out a 5-point polling lead (I explained in my recent essay on presidential election history why the 2-party vote is a useful metric). I highlight movement towards the wave party in yellow, towards the non-wave party in light orange.
2012: Comparing September 15 Polls To November 6 Election Results
Three things jump out from this chart. One, Republican Senate candidates led in seven races at this point two years ago that they went on to lose, even though mid-September was right after the Democratic convention and thus while President Obama was still enjoying something of a post-convention "bounce." Tommy Thompson had an 8-point lead and lost by 6, a 14-point swing (thus, a 7 point swing in the 2-party vote in the direction of the wave party). The Democrats didn't lose a single race they led, and the biggest swing to the GOP was in the Pennsylvania Senate race, which went from a blowout to a 9-point loss. Of the seven states that flipped, four were carried by Obama in November, and two others were probably the result of bad polling - Republicans lost races in North Dakota and Montana, in which they led in the RCP poll average on Election Day. The other was Richard Mourdock in Indiana.
Two, there was an across-the-board 2-party swing of 2.6 points towards the Democrats, the largest in any of the races we'll examine. That testifies to the strength of the Obama ground game, and suggests that the state-by-state polls, which did better than the national polls in the great polling debate of 2012 (more in my 3-part postmortem on 2012 polling here, here and here), were still playing catch-up in the Senate races in estimating what the final electorate would look like.
And three, if you look at the "Wave Party" column and the "Non-Wave Party" column, which subtract the September poll averages from the final results, you can see that most of the Democrats' gains came from their candidates improving in the polls, rather than from the Republicans sagging. In other words, the late deciders broke for the wave party. That is the recurring trend in most of these elections.
2010: Comparing September 15 Polls To November 2 Election Results
In 2010, unlike 2012, the wave was mostly baked into the polling cake by mid-September; most of the poll movement was Republican leaders pulling away, albeit even in races like Illinois where the GOP lead in mid-September was microscopic and the number of undecideds enormous. In Wisconsin, as in 2012, the race flipped with the wave after mid-September, with Ron Johnson pulling ahead of Russ Feingold (Wisconsin generally holds its primaries late, and turning out college students who arrive on campus in September is a big factor). Republicans did blow one race they led in mid-September (Ken Buck in Colorado, where there had been a nasty primary and the Governor's race was imploding), and one where they trailed in mid-September but led on Election Day (Sharron Angle's race against Harry Reid in Nevada).
2008: Comparing September 15 Polls To November 4 Election Results
In 2008, we see a 1.6 point overall shift to the Democrats, and three races (all GOP incumbents) flipped in the Democrats' direction after mid-September, all in states carried by Obama in the general election.
2006: Comparing September 20 Polls To November 7 Election Results
I used different dates than September 15 for 2002, 2004 and 2006 because those were the closest dates on which I could find the RCP poll page on the Wayback Machine. Again, we have three races flipping in the direction of the wave party, the Democrats, two of them Republican incumbents in Missouri and Virginia (the latter due in large part to George Allen's "macaca" flap helping cost him his 5-point lead), the other the New Jersey Senate race where Tom Kean had led Democratic incumbent Bob Menendez by 4 but was stalled at 45. The one race that "flipped" in the GOP direction against the wave was in Tennessee, where Harold Ford had been tied in the polls in mid-September, but overall there was a 1.7 point movement in the 2-party vote towards the Democrats.
2004: Comparing September 16 Polls To November 2 Election Results
It's hard to believe now that in 2004, when Republicans still held two Senate seats in Pennsylvania, Democrats were defending Senate seats in North Carolina, South Carolina, Georgia, Florida, Arkansas and Louisiana. Flips after mid-September rescued a Republican incumbent, Lisa Murkowski, in Alaska, buried the Senate Minority Leader, Tom Daschle, in South Dakota, and flipped open Democratic seats in North Carolina and Florida. Every Republican lead was safe.
2002: Comparing September 30 Polls To November 5 Election Results
We only have September 30 polls for 2002, but it was the wildest last month of a Senate election in memory, as two races flipped after the Democratic incumbent left the ballot - scandal-plagued Bob Torricelli in New Jersey dropped out and was replaced by Frank Lautenberg, rescuing a race the Democrats were losing badly, while Paul Wellstone - who had a small lead over Norm Coleman in Minnesota - died in a plane crash, and was replaced by Walter Mondale, who completed his collection of losing statewide in all 50 states. Two other Democratic incumbents, Max Cleland in Georgia and Jean Carnahan in Missouri, lost after leading in the polls entering October; no Republican lost a lead. In Louisiana, I measured the polls and the 2-party vote with regard to Repubicans' chances of forcing Mary Landrieu into a runoff (which they did, but then lost), as that was their goal that year. However, the 2002 Louisiana Senate race had the worst polling of any of the races I examined, and the "average" here is of one partisan poll from each side, with no public polling I could locate.
Note that, as in 2004 and 2010, and unlike 2012, 2008 and 2006, there was no overall move to Republicans - but the tight races broke consistently in the GOP's direction.
What Does History Tell Us?
If we sum up the overall bottom line from these six races, a pattern becomes reasonably clear:
Which gives us this net result:
Out of 87 races, the poll leader in mid-September (counting Ford in 2006, since he was tied with a wave at his back) lost 24 of them, and 21 of those flips broke in the direction of the wave - an average of 3.5 gains and 0.5 losses per year for the wave party from where they stood in mid-September. 31 of 87 races moved 2.5 or more points (i.e., a 5-point or greater total swing) in the direction of the wave party, compared to 15 that moved 2.5 or more points against the wave party.
But recall that a lot of the big movement we saw was in races that were effectively decided by mid-September. We don't care about a 5 point swing in this year's Montana Senate race, and for the same reason we don't care about races in the past where the poll leader in mid-September had at least an 8 point lead with the wave at his or her back, or in a very safe state. If we remove those races - and I admit that doing so introduces a bit of hindsight bias - we get 46 races. In 12 of those races, the wave party was ahead or tied in mid-September, and won 9 out of 12:
(I count the GOP as "leading" in 2002 against Landrieu because they were simply shooting to force a runoff and she was under 50; her actual poll lead over her nearest opponent was enormous, but she did end up in a runoff). Several of these races did tighten, while others, such as Texas in 2002, turned from a narrow lead into a rout - the wave party gained in 5 races, dropped in 6, and lost the one tied race and two of the leads, one of which (Doug Forrester leading Torricelli in 2002) was the result of a last minute change of candidate. The only actual blown lead was by Ken Buck. The other big dropoffs were voters in Montana "coming home" to the GOP in 2006, but not enough to save Conrad Burns, and Jim Bunning limping badly home in 2004 after having a big lead in a GOP wave year. Of the seven incumbents in these races, only two (Bennet in Colorado in 2010 and Burns in 2006) got noticeably more votes than their poll average in September.
By contrast, we have a sample of 34 races where the wave party trailing in mid-September:
The wave party won 21 out of these 34 races, 10 of them against incumbents of the non-wave party. It gained in the 2-party vote in 25 out of 34, gained by 2 points or more (enough to wipe out a 4-point lead) in 18 races, and gained by 2.5 or more in 16 races. The wave party won nine races where it trailed by at least 3.5 points, eight where it trailed by at least four, five where it trailed by at least 6.5, and three races where it trailed by 9 points at this juncture. And of the ten races where the wave party overcame at least a 2.5 point poll deficit, six were against incumbents - not reassuring news for Kay Hagan or Mark Udall.
The overall trend here is not as overpowering as the historical trend disfavoring the Democrats in the Presidential race in 2016, and as I cautioned in that essay as well, there is never any guarantee that history will repeat itself. Strange things can and do happen in individual races after mid-September, as we have seen in those last six cycles. Republicans have proven themselves quite adept in recent years at coming up with novel ways to lose winnable elections. And even without big blunders, waves do not simply happen: capitalizing on them requires a lot of hard work from candidates, citizen activists, political professionals, and donors. But if 2014 follows the path of 2002, 2004, 2006, 2008, 2010, and 2012, we would expect to see between 53 and 55 Republicans in the Senate in 2015.
« Close It
June 26, 2014
POLITICS: Waiting For The Wave: The 2014 Senate Map
The polling tells us that the bulk of 2014's contested Senate races are basically dogfights. So why are so many Republicans optimistic? Because it's still June, and some of the elements of the dynamics of 2014 may not be fully baked into the polling yet. How good a year this is for the GOP will depend on those factors.
If you look at the chart at the top of this post, what you pretty clearly can see from the data is that the Senate races right now seem to be sorted into three general groups (although in each group I'm including one race that is less favorable for the GOP than the rest).
Group One, three currently Democrat-held seats in deep-red territory without real incumbents, is the likely GOP blowouts. Montana and South Dakota are both looking locked up, and the South Dakota polling may get even uglier for the Democrats if the third-party support for Larry Pressler (a former Republican Senator running as an independent) fades. West Virginia is closer, close enough that a giant gaffe or scandal or something could put it back on the table, and in a different year or state a 10-point lead would not look insurmountable. But it's hard to see where that support comes from, in a 2014 midterm in West Virginia.
Group Two is the tossups, nine states that are really too close to call right now. Seven of the nine are Democrat-held seats, five with incumbents (Alaska, Arkansas, Colorado, Louisiana and North Carolina) and two open seats (Iowa and Michgigan). One of the two GOP-held seats has an incumbent (Kentucky), the other is open (Georgia). The Democrats have settled on candidates in all nine, Republicans still have a primary in Alaska (the poll average here is the matchup of frontrunner Dan Sullivan against incumbent Mark Begich), a runoff in Georgia (the poll average here is the matchup of frontrunner Jack Kingston against Democrat nominee Michelle Nunn), and a "jungle primary" that will probably result in a December runoff in Louisiana (the poll average here is the runoff matchup of frontrunner Bill Cassidy against incumbent Mary Landrieu). In only one of these races, in Michigan, does the current leader have a 5-point lead; in five of the nine races the frontrunner is below 45%, and in eight of the nine (all but Cassidy in Louisiana) below 46%. While a 2 or 3-point lead in the polls in October may be meaningful, a race with a lead that size in June and 10-20 percent undecided is functionally a tossup, at least until you take into consideration the various factors (national environment, state electorate) that are likely to pull the race in one direction or another as we enter the fall.
Why do Republican analysts feel so optimistic? Because polls, as we recall from 2010 and 2012, are only as good as their ability to project who will turn out and vote, and we are probably still a few months from pollsters being able to really make accurate assessments of what the fall electorate will look like. As Sam Wang, Ph.D., has noted, the various models for predicting how the Senate races will go are predicting different things depending on the extent to which they look beyond the polls to incorporate predictive elements like the economy, the effect of incumbency, the President's approval rating, and the like. Sean Trende, here and here, offered a model based mainly on Obama's approval rating, and found even after some tweaks to incorporate a few other variables, that Democrats could be projected to face double-digit Senate seat losses if the President's approval rating was 43% or lower on Election Day.
That's just one way of skinning this cat, but right now, Obama's approval sits at 41.5 approval/53.9 disapproval, and has been trending rather sharply downward for the past month, with his approval on the economy, foreign policy and healthcare all consistently worse than his overall approval rating. (Via Ace, it's even worse in the battleground states). In that national environment, with midterm elections in general tending to produce Republican-leaning electorates, and with the historic poor performance of second-term presidents in sixth-year midterms, you really have to feel pretty good about GOP chances of winning most of those nine races. That may seem improbable, but there were basically seven Senate races that went to the wire or involved potentially big Democratic upsets in 2012 - Pennsylvania, Ohio, Wisconsin, Virginia, Massachusetts, Indiana, and Missouri - and I didn't think at the time they would run the table and win all seven. They did. In a few of those, like Virginia and Wisconsin, the Senate races tracked almost precisely the outcome in the Presidential race, meaning turnout from the top of the ticket was decisive. If the national environment really does show as sour across the board for Democrats in November as it looks from today, eight-for-nine or nine-for-nine could be a possibility. If the environment (including the parties' turnout operations) swings back to a more neutral one, I'd be looking more at the GOP winning five of the nine, which would net a six-seat overall gain in the Senate, enough for control of the chamber but by a very narrow margin that might not last beyond 2016.
For now, that's still a big if, not reflected in polls showing voters not really ready to commit to either side in most of those races. It's why Republicans are waiting for the wave. But it's also a reminder that those races won't win themselves - Democrats ran the table in 2012 by fighting all the way to the whistle in every race with every resource they had. One thing helping the GOP may be the Governor's races: for example, Rick Snyder is now comfortably ahead in the polls in Michigan, and the Colorado GOP dodged a repeat of the 2010 trainwreck by picking Bob Beauprez over Tom Tancredo; Beauprez may not beat John Hickenlooper, but he'll give him a tough race without Tancredo's divisiveness.
Finally, there's Group Three, the races in which the polling shows the Democrats safe for now - but, depending on the national environment, maybe not safe enough just yet to declare those races over. Incumbents Jeff Merkley in Oregon, Al Franken in Minnesota, Jeanne Shaheen in New Hampshire all have leads around 10 points, and Mark Warner in Virginia has a sixteen-point lead on Ed Gillespie. (It's also always possible some other races could come on the board; there hasn't been much in the way of general election polling in Mississippi or New Mexico, for example. But we'll have to wait and see). But none of them are regularly polling above 50%, the usual rule of thumb for a safe incumbent.
Realistically, those are "reach" races that only go on the board if things really get ugly for the Democrats. Oregon is, I would guess, the best hope for the GOP relative to its present polling given the Cover Oregon fiasco, New Hampshire the toughest of the OR-MN-NH trio due to Shaheen's personal popularity and the likelihood of a landslide win for the Democrats in the Governor's race (the other two will have tight GOV races). Also, Al Franken has a huge warchest, so his race with self-funder Mike McFadden could get ugly and expensive. Virginia, of course, is the longest reach, but Gillespie should be sufficiently well-funded and anodyne to take advantage if Warner slides into the neighborhood of actually being vulnerable.
Predictions? Anybody who's predicting the fall elections in June with too much certainty is nuts. But right now, Republicans have a lot of opportunities in the Senate. If Obama's approval rating keeps tanking, the GOP avoids any major campaign-killing gaffes, and the Democrats don't come up with a magic turnout bullet, the swing in the Senate could be bigger than anyone is realistically talking about right now. Don't count your chickens; this is just the optimistic scenario. But it is not, from the vantage point of late June, an unrealistic one.
March 7, 2014
POLITICS: The Democrats' 2014 Whitewash
Barack Obama's electoral success has shown the Democratic Party the value of a non-white candidate in driving turnout and enthusiasm among the non-white voters that are vital to the party's success. So why are nearly all the statewide Democratic candidates this year white?
If there is one central theme to the political strategy of the Democrats and the electoral analysis and optimism of liberal pundits in the Obama era, it is race. To say that they are obsessed with these topics is to vastly understate the case. Virtually every analysis of "the Republicans' demographic problems" and the long-term case for Democratic/progressive dominance is premised upon the rising share of non-white voters in the electorate and their identification with the Democrats. To be sure, these are not Republicans' only challenges - even with younger white voters there are a few issues (mainly same-sex marriage and marijuana) on which the GOP is out of step with generational trends, and there is legitimate concern that younger voters of all races are less likely to be religious or get married, two traditional markers of conservatism. But even looking at the 2012 election returns, we see that Barack Obama lost white women by the largest margin of any candidate of either party since Walter Mondale, suffered a huge reversal among white voters under 30 (who he lost by 7 points after winning them by double digits in 2008), and even narrowly lost white women under 30. So, all of the Democrats' advantages along gender and age lines are still really just symptoms of a racially polarized electorate.
And turning out that electorate has been a challenge for Democrats. The big turnout wave of African-Americans for Obama exceeded anything John Kerry or Al Gore was ever able to muster, and the midterm elections in 2009, 2010, and 2013 (with the arguable exception of the 2013 Virginia Governor's race) yielded electorates that were older, whiter and more conservative than the 2008 or 2012 electorates (this was even true in 2006, although depressed GOP turnout and heavy independent support for Democrats made that a big year for the Democrats anyway). There has been much open fretting by Democrats that the turnout will look the same this year - which threatens to make this a serious wave year for Republicans, given the public mood. That's even before you get to the fact that Democrats' rising success with non-white voters has coincided with hemorrhaging support among white voters and the very real possibility that the Democrats haven't yet found their floor among white voters. To say nothing of the possibility that the natural long-term arc of Hispanic voter preferences may move back in the direction of the GOP. In the immediate term, we have already seen polling showing that Hispanics are the most disillusioned of Obama's 2012 supporters. Few things in a two-party political system are forever.
And there is a very real sense in which the big turnout of 2008 and especially 2012 was a show of racial solidarity with Obama (and his wife) personally, as much as it was a traditionally political phenomenon. There were all sorts of signs of this in the 2012 exit polls. Only 23% of voters in the exit polls said that the economy was in good or excellent shape, for example, but 90% of these voted for Obama. Who are these voters? A July 2013 Quinnipiac poll - somewhat typical of the breakdowns these days - found that 47% of black voters, but only 25% of white voters, described the for state of the economy as good or excellent. By contrast, an October 2007 CNN poll found 69% of black voters describing the economy as in recession, compared to 42% of white voters. This, despite the fact that the objective evidence shows unemployment significantly higher among African-Americans in 2013 than 2007.
But forget the data; listen to liberal African-American pundits. Here is The Atlantic's Ta-Nehisi Coates, laying it out in the purplest of prose:
Barack Obama was not prophecy. Whatever had been laid before him, it takes gifted hands to operate, repeatedly, on a country scarred by white supremacy. The significance of the moment comes across, not simply in policy, by in the power of symbolism. I don't expect, in my lifetime, to again see a black family with the sheer beauty of Obama's on such a prominent stage. (In the private spaces of black America, I see them all the time.) I don't expect to see a black woman exuding the kind of humanity you see here on such a prominent stage ever again....I don't ever expect to see a black man of such agile intelligence as the current president put before the American public ever again.
Or here is the Daily Beast's Jamelle Bouie, writing in the midst of that election:
The upside of making the race of the candidate an existential issue for African-American voters is, it's a tremendous motivator to turn out to keep the symbolic leader in office. The downside is, it's not easily transferable to other candidates - not to other non-white candidates for lower offices, and certainly not to a bunch of white politicians who look pretty much just like the people they are running against.
And yet, bafflingly, that is exactly what the Democrats are running in 2014. At this writing, the Democrats are running a candidate in 62 Senate and Governors' races this fall (nobody has really stepped forward yet in the Nevada, Tennessee or Wyoming Governors' races). And depending how you count the frontrunners, anywhere from 57 to 60 of those 62 candidates will be white (92-96%), and 47 to 49 of them will be white males (more than 75%). Let's take a look at that roster of candidates, ranked by a very rough ranking of the competitiveness of the races ("1" being hotly contested races, "2" being races that will be contested but with a clear favorite, "3" being races that look lopsided and may end up being de facto cakewalks - this is giving the benefit of the doubt that a lot more races will be competitive than polling may suggest, but races like the New York and Texas governorships will be big-time battles even if the outcome seems pretty clear in advance). I also rated as at least a 2 every race with a Republican Senate incumbent who has a non-obscure Tea Party challenger. I marked with an asterisk the races in which the Democrats have a significant chance of ending up with a different candidate - for example, the one black female candidate here, Richland County Councilwoman Joyce Dickerson in South Carolina, is an obscure candidate with a white male opponent in a race so unlikely to be contested that there's been no polling (I rate her as the frontrunner because she at least holds elective office, but with a primary electorate that ran Alvin Greene in 2010, you never know). One white male Democratic Senator, Brian Shatz, faces an Asian female primary opponent, Congresswoman Colleen Hanabusa, who may well defeat him, and David Alameel in Texas is in a runoff with Kesha Rogers, a black female LaRouchie who wants to impeach Obama. On the flip side, the two non-white Democratic frontrunners for Governor, Providence Mayor Angel Taveras in Rhode Island and Lieutenant Governor Anthony Brown in Maryland, still face significant white primary opponents - Rhode Island State Treasurer Gina Raimondo and Maryland Attorney General Doug Gansler, respectively. So the number of non-white candidates could easily go down rather than up.
As you can see here, beyond Cory Booker (who faced a real race in October but as of now has no real opponent), not only are the Democrats running a virtually all-white slate of candidates in the marquee statewide races, just about every Democrat in a hotly contested race this year is white. (Protip to activists: somebody with the time to put together a graphic of all these candidates could have some fun with it).
Should that matter? Of course not. Does it? Look at the primary results from this week's Democratic gubernatorial primary in Texas - and you can see that white female abortion zealot Wendy Davis lost most of the Southwest Texas border counties - the places where Barack Obama did best in 2012 - to a primary opponent who has basically no campaign, but who had a Hispanic surname:
The GOP candidate, Greg Abbott, will not hesitate to send his Hispanic wife, Cecilia, to campaign there.
For a party so focused on "diversity" as a slogan and the turnout of non-white voting blocs as a lifeline, it's hard to see why you would run that risk. Of course, a similar analysis of the leading Republicans would also show a heavily white, heavily male slate - but a little less so: Republicans are running two non-white incumbents in South Carolina, Nikki Haley and Tim Scott, two incumbent Hispanic Governors in Brian Sandoval and Susanna Martinez, and a Native Hawaiian gubernatorial candidate, former two-term Lieutenant Governor Duke Aiona, as well as a number of white female candidates. And more to the point, Republicans are already doing fine with white voters; they're not the ones who are existentially dependent upon firing up non-white voters with racial appeals. Democrats are - and so their failure to recruit and develop more non-white candidates adds yet another cause for alarm in what is already shaping up to be an alarming election season.
And if the results are ugly, that may make the Democrats rethink running a 69-year-old white woman as their national candidate in 2016.
(NOTE: The original version of this article stated that Joyce Dickerson's opponent in the race to run against Tim Scott had a felony record - actually, it's Jay Stamper, who is running against Lindsay Graham, who has a felony record. I've corrected that as quickly as I could.)
December 9, 2013
POLITICS: Hispanic Voters Lose Faith In Barack Obama
We all know how far President Obama's approval rating has fallen, 13 months after his re-election. Gallup had a fascinating look at who exactly has lost faith in Obama, among poll respondents who approved of him a year ago. And prominent among the groups with the biggest drop is the supposed bedrock of the Permanent Democrat Majority™ - self-identified Hispanic voters.
As always, bear in mind that Gallup is only one pollster, and not the most reliable one at that, and that sub-samples tend to be smaller sample sizes than an entire poll. That said, a comparison between two polls by the same pollster at different points in time is an apples-to-apples comparison, and so of some use in tracking trends. I'd supplement this with similar data from other pollsters, but surprisingly few of the daily, weekly or monthly presidential-approval tracking polls provide this kind of breakdown on a regular basis (although a mid-November Quinnipiac poll showed Obama's approval underwater with Hispanics, 41-47, compared to 67-18 approval a year ago). That's precisely why Gallup's results are so interesting.
Here's Gallup's chart, which measures the net drop in points among different groups in their approval of Obama between his post-spring-2009 high water-mark in December 2012 and November 2013:
Bearing in mind that some of these are overlapping groups, you can see not only that Hispanics register a 23-point drop in approval, the largest of any group, but others near the top are also essential elements of any winning Democratic coalition: the youngest voters (18-29 year olds), the poorest (incomes under $24,000), the least educated (high school or less), various stripes of moderates and independents, women, the unmarried, the irreligious and voters in the Northeast and Midwest. Obama has lost the least support among those where he had the least support to start with: conservative Republicans, conservatives, Republicans. He's dropped at least 8 points among everyone else.
But a second way to look at these numbers is in percentage terms, to adjust for the fact that it's easier to lose more support among groups where you had more to start with. So, here are those figures:
Here, we see unaffiliated independents - perhaps unsurprisingly - at the top of the list, along with liberal Republicans, but Hispanics in a very close third place, with the poor, the uneducated and Midwesterners also high on the list (the latter is a danger sign for Democrats in Michigan, Iowa, and Minnesota next fall). By contrast, on a percentage basis, the three least-shaken elements of Obama's base are - unsurprisingly - black voters, among whom he's lost just 10% of his prior support, and liberal Democrats and Democrats as a whole, the only other two groups to show less than a 16% decline - and the dropoff among those two groups would doubtless be higher if you excluded his unshakeable support among black voters.
And bear in mind, these are losses among people who approved of the job Obama was doing after watching him in office for four years. These aren't voters who reject liberals or Democrats out of hand, or - if you buy into the excuse for every criticism of Obama - voters who don't like him because he's black. Most of them likely voted for the man, twice. And the most likely reason they are turning on him is simply because he's not getting the job done - his policies aren't working, the economy continues to disappoint, and he still isn't delivering the things he promises. As the LIBRE Initiative's Executive Director, Daniel Garza, put it, after noting that the November Quinnipiac poll had showed just 44% approval among Hispanics for Obamacare:
The dramatic drop in support from the U.S. Hispanic community should not come as a surprise to anyone. Instead of effectively addressing Immigration, the slow economy, the lack of access to affordable care, and other critical issues over these past five years, President Obama has delivered mostly empty rhetoric and a record of stagnant Unemployment , diminished household incomes and a tepid GDP growth rate. Americans deserve better.
The immediate lesson here is that, for all the Democrats' bluster, Hispanics are simply not African-Americans. They may have identified to some degree with him against his critics as the first non-white President, they may like some of the things the Democrats stand for, and they may even feel - not without reason - that Republicans want to kick them out of the country. But none of that alone is enough to make them permanent Democratic partisans if they don't see results.
Republicans face a variety of challenges in appealing to Hispanic voters, even moreso than some of the other voter groups that are increasingly disenchanted with Obama. They will not be cheap dates for the GOP. But Democrats are learning that they are growing increasingly tired of being told to just sit back and pull the lever for Obama's pursuit of MacGuffins. Republicans have an opportunity, if they will work for it.
November 25, 2013
POLITICS: We Are The 55% (Or: Mitt Romney Was Wrong About Obama's 47% Floor)
Daily polls can make your head spin, and getting too excited or distressed by a single poll is never advisable. But sometimes, a clear trend emerges in poll after poll that cannot be denied. And that trend is now beyond dispute: a majority of American voters think Barack Obama is not doing a good job as our President. Looking at the RealClearPolitics polling average, Obama has:
-Had more voters disapprove than approve of his job performance every day since early June;
-Seen over 50% disapproval consistently since early August;
-Seen the disapprovals outnumber the approvals by double digits on November 8 for the first time in his presidency, and stay there; and
-Hit 55% disapproval five days ago and stay above that level.
Obama's approval is now down to 40.5%, registering above 42% in only one poll in the average, Rasmussen Reports (Rasmussen, once a reliable if somewhat GOP-leaning pollster, has become increasingly volatile and less transparent since the departure of its founder Scott Rasmussen, who started scaling back his involvement earlier this year and left the company in August). It's possible that it may drop below 40 for the first time any day now, as it's no longer rare to see individual polls with a 38 or 39% rating - an approval rating lower than crack-smoking Toronto Mayor Rob Ford. Over at the Huffington Post's HuffPost Pollster page, Obama's approval rating is so low it has literally fallen off the chart; you have to adjust the default settings (which bottom out at 42.5%) to find it:
In one bit of irony, consider Mitt Romney's famous remark:
There are 47 percent of the people who will vote for the president no matter what. All right, there are 47 percent who are with him, who are dependent upon government, who believe that they are victims, who believe the government has a responsibility to care for them, who believe that they are entitled to health care, to food, to housing, to you-name-it.
As far as approving of this President, Romney has been proven wrong: Obama's approval rating hasn't seen 47% since June 10.
It's true, of course, that a chunk of the people disapproving of Obama are his own 2012 supporters - but that's the point! Presidents don't really get in trouble until they start disenchanting their own side. It's also true, as we recall from 2012, that national polls of this nature are not as precise as state-by-state polling in predicting voter behavior and turnout - but a persistent and growing gap this size is hard to hide, and if you use HuffPo's widget (which lets you examine a sub-sample of pollsters) to back out the impact of Gallup and Rasmussen, Obama's numbers get worse, not better.
Why does all this matter? Let's quote a few of the arguments.
Sean Trende looks at the impact on 2014 House races, although of course the calculus as to the map is quite different for Senate races:
[P]residential job approval is still the most important variable for how his party fares in midterm elections, explaining about half of the variance. The relationship is highly statistically significant: For every point in job approval the president loses, his party loses 0.6 percent of its caucus. (The chart doesn’t measure drop in job approval; just job approval.) So, at 60 percent, the president should lose 5 percent of his caucus; at 50 percent, it is around 12 percent of his caucus lost; at 40 percent, it’s about 18 percent of his caucus lost -- which would be 36 seats.
Chris Cillizza notes the impact on Obama's ability to get things done, and that the trendline of his approval rating is more in common with that of George W. Bush than more popular second-term presidents like Reagan and Clinton who left their party in good shape in the next national election:
The loss of the Senate majority and a smaller minority in the House after November 2014 would make any attempt to rack up second-term accomplishments before he left office extremely difficult for Obama. Combine that with the reality that Obama’s second term has not exactly been larded with major wins to date and you understand why Obama and his legacy are on the ballot in 2014 — even if his name is not. And that means his poll numbers matter. A lot.
John Sides at The Monkey Cage noted back in June how much this can matter:
[I]t matters for whether the President gets what he wants from Congress—with some caveats. Here’s a sense of some of the scholarly literature on the relationship between presidential approval and legislative success. One question is whether Congress simply passes legislation that the president supports. In one study (gated) of 208 roll call votes in the House between 1989-2000, political scientists Brandice Canes-Wrone and Scott de Marchi found the House was more likely to do what the president wanted when the president was more popular. This effect was only significant among legislation that was both salient (mentioned a lot in news coverage) and somewhat complex (focusing on regulatory matters in particular). But, of course, that’s exactly the kind of legislation—e.g., immigration, gun control—that Obama would like to sign right now.
Harry Enten noted in September that the odds are against Obama recovering by 2014: "The president's approval rating has never increased by more than 7pt from this point after re-election until the midterm election."
Enten, after looking at 2014, notes that the impact goes beyond it to 2016:
[T]he president's approval plays a role in the election to find his successor. Once we control for the economy, every 5pt increase in a president's net approval rating increases his party's candidate's margin by 1pt in the presidential election per Drew Linzer. An election his party might have won by 1pt had the incumbent president had a +5pt net approval rating becomes an election the incumbent party loses by 1pt with a -5pt rating.
By and large, presidents whose parties have done badly in 6th-year midterm elections have also seen their party lose ground in the national popular vote in the next election. Here, I charted out the parties from best to worst showings in holding onto their share of the popular vote in the next presidential election following a two-term presidency, and how they had done in the prior midterm - for example, the Democrats lost 0.9 points in the popular vote from 1996 to 2000, and I line that up here with their showing in the 1998 midterms; Republicans lost 7.8 points in 1960 from 1956, and I line that up with the 1958 midterms.
Here's an expanded chart with a few more of the post-1860 presidents who don't fit as neatly (for example, the GOP in 1904 had held the White House for 8 years, but its candidate was an incumbent, not a new contender trying to run on the party brand).
Here's two charts lining up the showings overall of parties seeking to defend a presidential race after re-electing an incumbent; historically, Democrats have struggled slightly more than Republicans in hanging onto their share of the voters:
Note that Obama only won the popular vote in 2012 by 3.9 points, and there was no significant third-party candidate, so if the Democrats lose 2 or more points off their 2012 showing, they lose the popular vote (and you can win the Electoral College while losing the popular vote, as Bush did in 2000, but mathematically it's almost impossible to do so unless it's so close as to be almost a tie). In other words, if the Democrat in 2016 falls off Obama's 2012 showing by 2 or more points, there's a high likelihood that the next President will be a Republican - and the only non-incumbents running after an incumbent was re-elected (and thus, seeking a de facto third term with a new candidate) to suffer less than a 4-point falloff in popular vote were Gore in 2000 and Hoover in 1928.
None of this should suggest that Republicans don't have problems of our own, or that success is about to fall inevitably into our laps. But with 55% of the public disapproving Obama and unlikely to change their minds in significant numbers, there's a major opportunity for the GOP ahead.
September 9, 2013
POLITICS: The New York City Mayor's Race: Race, Crime, Unions, Taxes, and...Other Stuff
Tomorrow - Tuesday, September 10 - New York City voters go to the polls to pick the major-party candidates for their next Mayor. Candidates need 40% of their party's vote to avoid an October 1 runoff election among the top two finishers in the primary. At this writing, it appears that the nominees - possibly without a runoff - will be Republican candidate Joe Lhota (a former Rudy Giuliani aide and more recently Andrew Cuomo's appointee to head the MTA transit system) and Democrat/Working Families Party candidate Bill de Blasio (a former David Dinkins aide and the city's Public Advocate). To get from here to Election Day, the City may reopen old racial wounds and have to grapple with the legacy of its last three Mayors.
Five Decades, Four Mayors: Koch, Dinkins, Giuliani and Bloomberg
The Dinkins Debacle
It pays to begin with a thumbnail sketch of the past four mayors, beginning when David Dinkins toppled Ed Koch in the Democratic primary in 1989. Koch, elected in 1977 as the City reeled from a financial crisis, was then seeking his fourth term as Mayor after a scandal-riddled third term; Dinkins was running to become the City's first African-American Mayor. Koch, once a liberal Congressman, had governed as a relatively pro-business Democrat (he was both the Democratic and Republican nominee in 1981) and rejected liberal political correctness on crime, although his anti-crime initiatives were less vigorous and successful than his revival of the City's economy and finances. New York under Koch enjoyed the prosperity and Wall Street boom of the 80s, but it was neither particularly clean nor safe, with the burden of high crime rates falling most heavily on poor black and Hispanic neighborhoods; Tom Wolfe's Bonfire of the Vanities aptly captured New York in the Koch era. But Democratic primary voters made the problem worse. Facing a 'historic' black candidate in Dinkins, Koch lost roughly 95% of the black vote in the primary, a showing on par with former KKK Grand Wizard David Duke's 4% of the black vote in the general election for Louisiana Governor in 1991. (The Jewish Koch, a fighter for civil rights in the 60s, also had an increasingly acrimonious relationship with Jesse Jackson after Jackson called New York City "Hymietown" - an anti-Semitic slur - in 1984, as well as with rabble-rousing street preachers like then 35-year-old Al Sharpton.) Dinkins rolled up similar margins among black voters along the way to winning the general election against Rudy Giuliani, best known at the time as a crusading US Attorney who took on the mob, drug dealers and Wall Street insider trading.
To describe Dinkins as a failure as Mayor would be a massive understatement. Not even Jimmy Carter managed to discredit liberalism in action as garishly as Dinkins, who saw the murder rate explode and the city descend into the sort of chaos and racial strife that had liberals declaring it inherently ungovernable. In a 1993 rematch providing the same sort of perfect storm of opportunity to move the electorate rightward as Ronald Reagan's 1980 victory over Carter, Giuliani ousted Dinkins - but, this being New York, only narrowly. "Bad as the previous four years were - about 1,700 private-sector jobs lost every week on average, homicides surpassing 2,000 per year, more than 1 million residents on welfare - just about half the city was reluctant to give up on its first black mayor, and the voters in November 1993 ratified change only grudgingly. Incumbent David Dinkins was widely seen as ineffectual, but out of 1.75 million votes cast, in so heavily Democratic a town, Giuliani won by just 50,000." To this day, Dinkins contends that racism rather than the catastrophic state of the City was behind Giuliani's law-and-order campaign and victory (that's not hyperbole: according to his 2013 autobiography, "I think it was just racism, pure and simple").
The Rudy Revolution
What followed was a staggering turnaround in the City's fortunes in general and its law enforcement in particular, completely revising everything people believed about the city's safety and governance. Left-wing frequent Giuliani critic Michael Tomasky wrote that "[m]odern New York, with its safe streets, its gentrified Brooklyn, and booming tourist economy, was born on January 1, 1994. And, love him or hate him, it was Rudolph Giuliani who made the city what it is." George Will called Rudy's tenure "the most successful episode of conservative governance in this country in the last 50 years," Reagan included. Thanks in good part to aggressive, hands-on policing that started with a "broken windows" theory of going after petty offenders like squeegee men, New York became the safest large city in America, the stratospheric murder rates a distant memory. The welfare rolls were cut in half, the public-sector unions brought in line, a few particularly onerous taxes cut, Times Square reclaimed from the hookers and the sex shops to be a place so family-friendly Disney would (literally) later open a store there. Rudy's New York was still socially liberal and far from a libertarian paradise, but he had made it governable again.
Of course, there were always those who never accepted Rudy or his methods, most of all the Sharptonite resistance to Giuliani's law enforcement policies. But a combination of internal Democratic division and external force kept them from unifying when Rudy's tenure was up. Term limits put in place with the Koch third term in mind had made Rudy a lame duck by the time New York voters went to the polls on primary day of his final year in office: September 11, 2001. Between a racially divisive primary against Fernando Ferrer that saddled ultimate Democratic nominee Mark Green with low black turnout in November and the long shadow of the September 11 attacks, the voters elected to stay the Giuliani course with Mike Bloomberg. When the dust settled (literally), the billionaire who had built his fortune catering to Wall Street with trading desk terminals and business news was given the task of rebuilding the City's shattered downtown.
Bloomberg Finds New York's Center
While it's easily forgotten in national conservative circles that revile him, Bloomberg's 12 years in office (he shoved aside the term limits with the collusion of City Council Speaker Christine Quinn) have struck a course that is essentially centrist within the context of NY City politics, complete with drifting in and out of the Republican Party as it suited his purposes. Bloomberg continued and in some ways refined Rudy's approach to law enforcement and management, making it more sophisticated in its use of increasingly detailed data. Rudy had been tough on guns; Bloomberg raised it to the level of anti-gun zealotry. He's held the line against pressure to raise income taxes, and left the private sector mostly free of interference for economic purposes (as Bloomberg responds to charges of being 'in the tank for Wall Street': "I'm in the tank for industries in New York City! That's my job. That's the way people here eat!"). But Bloomberg often sticks his nose into business to advance one of his lifestyle crusades like banning big sodas, tossing smoking out of bars or inveighing against salt, and he also signed off on multiple rounds of property tax hikes. He's pursued a neoliberal policy on education, accumulating more power in the Mayor's office (his long-time schools chancellor, Joel Klein was Bill Clinton's antitrust enforcer), promoting charter schools and using government controls to hold public schools more accountable - which, combined with a negotiating line that prevented reaching a contract with the teachers' union the last several years, has earned him the enmity of the teachers. Befitting a former business executive, Bloomberg has proven a highly capable manager of day-to-day government operations, but has struggled when crises ranging from heavy winter snows to attempted terrorist attacks have called for him to rise to the occasion. And in myriad ways, when Bloomberg needed to buy off support or acquiescence to his policies and ambitions, he's done it by throwing around his own considerable wealth rather than the taxpayers' money.
Many New Yorkers have wearied of Bloomberg's personality, soured on his evasion of term limits and dissented from this or that policy - as Jonathan Chait notes, Bloomberg's contempt for the liberty or good opinion of the individual citizen has over 12 years worn poorly even in New York - but most observers of the New York scene would be hard-pressed to find evidence that the electorate wants a return to Dinkins-era progressivism run wild. A February 2013 Quinnipiac poll showed Bloomberg with a 53-40 positive approval rating and found that 31% of New York City voters cited Giuliani as the best Mayor of the past 50 years, with 25% saying Koch and 24% Bloomberg - compared to just 6% for the archliberal Dinkins, 6% for Great Society liberal Republican John Lindsay and 1% for conventional Democrat Abe Beame.
Bloomberg's Heir: Christine Quinn Misreads The Primary Voters
The conventional wisdom entering the 2013 race, therefore, was that the City's inherent Democratic partisanship (Democrats control nearly everything but the Mayor's office) was overdue after two decades out of power to reassert itself, but most likely in the form of a candidate who would not run dramatically far to Bloomberg's left. Enter Bloomberg's reliable ally and co-conspirator, Christine Quinn. By any reckoning, Quinn was the heavy favorite when the primary began and well into the summer, and is the establishment candidate in the race, winning a rare trifecta of primary endorsements from the New York Times, the Daily News and the Post and running with Bloomberg's blessing and de facto backing. Quinn backed Bloomberg on term limits and has mostly supported his education policies while joining forces on his various nanny-state crusades, and in a broad sense is seen as his heir. Yet, Quinn is distinctly more liberal than the billionaire, most notably on issues relating to private sector business, ranging from a more pro-union stance where Bloomberg has been generally neutral in private-sector labor disputes to an insane law permitting suits for discrimination against the unemployed that she passed over Bloomberg's veto to caving to union pressure on a paid-sick-leave bill. (Quinn has been endorsed by the Teamsters and the building-trades unions). She's also been an increasingly strident critic of the NYPD's "stop and frisk" policies, and has a sour relationship with the NYPD's union.
The massively powerful teachers' union has been a particularly big obstacle for Quinn. Having worked four years without a contract, the teachers are looking for a budget-busting retroactive pay raise that (along with demands by the rest of the City's 300,000 unionized workers) carries an estimated price tag in the $7-8 billion range (that's a thousand dollars from every New Yorker). The 70,000-member UFT hasn't endorsed a successful candidate since Dinkins, prompting Bloomberg to label its endorsement a "kiss of death." The GOP candidates have opposed retroactive raises, with Lhota taking the lead on the issue; Weiner has suggested they be conditioned on concessions on the unions making contributions to their own healthcare (along the lines of what Chris Christie negotiated in New Jersey); Thompson has eagerly endorsed the teachers' demand for retroactive raises, earning him the UFT's endorsement; City Comptroller John Liu's left-wing platform earned him the endorsement of District Council 37, the City's largest public-worker union. Quinn and de Blasio have been cagier on the issue; Quinn has refused to deal in "hypothetical" questions about budget deals, while de Blasio has refused to rule in or out retroactive raises. The difference in their positions, however, is less important than issues of trust: teachers angry at the long stalemate with Bloomberg and his efforts to bring the schools under more Mayoral control have been unwilling to trust his wing-woman.
Quinn's lead in the polls has crumbled so badly that she's now seen as unlikely to make the runoff, if there is a runoff; liberal as she is, and as hard as she pushes identity politics (she touts herself as the potential first woman Mayor of New York; she would also be the City's first openly gay Mayor - Koch's sexual preferences were the subject of much speculation but never confirmed), she's not a likeable campaigner and the Democratic primary voters seem inclined after two decades out of power to reassert their differences with Bloomberg. All of Quinn's substantive dissents from Bloomberg haven't managed to separate her in the public mind from the Mayor. And perhaps the most enduring lesson of Quinn's imminent failure, in light of her alliance with Bloomberg on education, is that a white female Democrat simply cannot afford to be at odds with the teachers' unions.
Dante's Identity: Bill de Blasio and Race
For much of June and July, the Democratic primary race was divided into four tiers: Quinn and Anthony Weiner battling for the top spot in the low/mid 20s, de Blasio and Bill Thompson fighting for position in the low teens, Liu stuck in fifth place due to a lurid campaign-finance scandal, and the rest of the candidates (because what this goat rodeo really needs is more candidates) not even worthy of being polled and not registering when they were. As this great New York Times infographic illustrates, New York City politics remains a labyrinth of racial, religious, ideological and union voting blocs. Many observers, myself included, thought that it was premature to count out Thompson, given that he had won 48% of the vote in the 2009 general election against Bloomberg (who broke all known records for per-vote spending in a major election) and could draw on the traditional loyalty of African-American voters to black candidates, a particularly pronounced tendency in New York City over the years.
The race was rocked on July 23 when Buzzfeed broke the blockbuster revelation that the disgraced Weiner had continued his 'sexting' ways on the internet after being driven from Congress, beginning the process of the bottom dropping out of his support. By August 8, the RCP polling average had Quinn with a 10-point lead and Weiner, de Blasio and Thompson all tied up around 16%:
The Family Card
That's when de Blasio rolled out this ad, featuring his 15-year-old son Dante (visibly reading off cue cards, but hey, he's 15) and his eye-catching throwback-70s Afro:
The ad was such a hit that Dante has had to field questions about his own future political ambitions (he'd probably have to keep the now-signature Afro if that's his plan). Quinnipiac, which has polled this primary race more than any other pollster, illustrates the dramatic effect that followed:
The "Dante" ad wasn't the only factor at work, but it worked on two levels: one, at the level of raw identity politics, it spread awareness that de Blasio's wife and son are African-American; and two, it hit directly on the racial hot-button accusation that the NYPD's "stop and frisk" policy "unfairly targets people of color." The cherry on top is de Blasio's pledge to raise income taxes.
As to the former, on Saturday, Mayor Bloomberg made headlines with his response to the ad, which he initially described as "racist" before backtracking (and later pressuring NY Magazine to drop the "racist" reference):
I mean he's making an appeal using his family to gain support. I think it's pretty obvious to anyone watching what he's been doing. I do not think he himself is racist. It's comparable to me pointing out I'm Jewish in attracting the Jewish vote. You tailor messages to your audiences and address issues you think your audience cares about.
On one level, Bloomberg misses an important point: a guy named Bloomberg doesn't have to tell voters he's Jewish, any more than Bill Thompson has to tell people he's black or John Catsimatidis needs to tell people he's Greek. But de Blasio, a white guy with an Italian surname, has to show people his family to make the point. Identity politics is a sad reality of politics, but we usually don't see candidates remind voters of it quite so bluntly.
If the goal was to prevent Thompson from consolidating black voter support behind the only black candidate in the race, it succeeded wildly. The latest polls show de Blasio leading Thompson among black voters by double-digit margins: 37-26 according to Quinnipiac, 42-26 according to PPP, 39-25 according to Marist.
The Politics of Stop and Frisk
But de Blasio's push is about more than just identity itself; the bigger element of the ad is the racial wedge issue of "stop and frisk." Now, it's important to stop here for a minute and do what Bloomberg and so many others fail to do, and define our terms, because talk about race in politics is chronically beset by confusion over words that have distinctly different meanings:
Racism is a set of ideas and beliefs, about the superiority or inferiority of different groups of people as defined by "race." Of course, race itself is largely an artificial set of distinctions among people who are biologically different only in superficial ways, which is one reason (not the only one) why racism is idiotic nonsense. In practice, it's less important what you believe than what you do - Abraham Lincoln and his generation of Republicans had many beliefs that would strike us today as racist, but what mattered was that they put their blood and treasure on the line to improve the lives of enslaved black people. That said, voters are rightly interested in the beliefs of political candidates, which are often more enduring than their promises. The problem with accusations of racism in politics is not that they're unimportant but that they're non-falsifiable: that is, there's no type of evidence that can be presented to disprove them (indeed, citing any available type of evidence is usually regarded as additional proof that you're actually a racist).
Racial discrimination is the real offense: treating people differently because of their race. That's the case whether the discrimination is individual or whether it's a systematic structure of discrimination like Jim Crow or apartheid. In theory, everybody's against discrimination - except, you know, racial preferences in education and employment.
Disparate impact is treating people the same, but in a way that affects people within different racial groups differently. To use a recent and fairly mild example of this, some people have referred to a tax on tanning beds as "racist," which is ridiculous; that said, for obvious reasons tanning is mostly a thing white people do, so if you're against every possible form of disparate impact, that's the kind of thing you end up crusading against.
Racialism is the habit of viewing everything through the lens of race, a terribly destructive habit but - of course - a hard one to shake when you attempt to write about New York City politics, or national politics in the Age of Obama.
Racial wedge issues, or race-baiting, or the race card, or any number of similar terms refer to pressing political issues or appeals that divide people along racial lines, and that's where we get to what's really at stake with the stop-and-frisk debate and how we talk about it.
The NYPD has - using crime data and statistics - conducted an increasingly active campaign of preventive law enforcement built around stopping individuals on the street for questioning (a tactic blessed by the Warren Court in 1968 in Terry v Ohio so long as there is "articulable suspicion") and frisking them for weapons when deemed appropriate. Beat cops have been concentrated in high-crime areas. A lawsuit charging the NYPD with discrimination did not produce evidence of any policy of racial profiling (thus, no overt evidence of racial discrimination on a non-isolated basis), but relied on statistical evidence to argue (persuasively, to the district judge) disparate impact and an inference of discrimination. In a nutshell, the evidence showed that black New Yorkers were stopped in numbers far disproportionate to their numbers among the population - while the City noted that the demographics of people stopped matched well with the actual population of criminals (as determined both from arrests and victim reports). The evidence also showed that the NYPD was more likely to have "false positive" stops of blacks - ie, stops without a well-explained basis or a resulting arrest.
We have not racial-profiled, we've gone where the crime is....
The merits of the debate demand a more detailed look than space permits here; Slate's Eric Posner explains why the decision got the wrong result:
Twelve percent of the stops resulted in an arrest or summons...So, police stopped black people more often than they stopped whites even though whites constitute a larger fraction of the city's population; they used force against blacks more often; and yet they found weapons and contraband less often when they searched blacks than when they searched whites.
There's the rub: an ideal stop-and-frisk policy will never be error-free, and liberals of all people should know better than to denounce any government program that's not error-free. The question is whether the errors are worth the benefits, and whether they are fairly distributed. As to the former, it's hard to look at New York today compared to the Dinkins years and not want to give a lot of leeway to the NYPD's nearly-miraculous record of crime reduction, a record few government programs in any field can match. As to the latter, an ideal policy will produce a rate of stops that looks like the actual criminal population - that's not a Bayesian fallacy but a recognition that the distribution of errors should mirror the distribution of successes. If the impact on innocent black men is disproportionate, it's because - unfortunately - they disproportionately live in neighborhoods victimized by black male criminals. Any effort to skew the numbers away from the proportions they would hold if they were 100% accurate, simply for purposes of spreading the pain to other racial groups, is not fairness, but its opposite.
For political purposes the question is less about the merits than about how it resonates with the voters - and despite being the major beneficiaries of lower crime rates, black voters are especially hostile to stop-and-frisk and particularly receptive to explicit political appeals arguing that it's racist, discriminatory or at least racially unfair. Is it fair to raise a political issue that divides voters so explicitly on race? I'm no fan of racial politics, for a lot of reasons: playing the race card and gaining voter loyalty on racial lines is often a way of distracting from the real issues and insulating inept or corrupt politicians from accountability. But fundamentally, public policy issues like stop-and-frisk are important issues, just as things like prison furloughs or racial preferences are important issues the voters should be heard on. Liberals who spent two decades freaking out over Willie Horton and the famous Jesse Helms "hands" ad against Harvey Gantt have no moral standing to defend de Blasio's use of a similar racial wedge issue - but even if de Blasio's wrong on the issue and his supporters are hypocrites, that doesn't mean it's an illegitimate issue. Indeed, the parallel to the Helms ad is pretty obvious: in both cases, the candidate is appealing on explicit racial lines to the group that is asked to have innocent members pay the cost of a government social policy (the difference being that preferences are a true zero-sum issue, whereas lower crime rates benefit everyone). That may be ugly and it may be divisive, but at some point, it's still the voters' business.
The Dinkins Legacy
With de Blasio and the other Democratic candidates vowing to put an end to stop-and-frisk (all they'd need to do is drop the City's appeal of the ruling), there's little question that this election will now put at stake 20 years of thinking about law enforcement. The Wall Street Journal reports that it's already taking its toll on the cops' willingness to perform stops.
If all of this seems like de Blasio is trying to overturn 20 years of Giuliani/Bloomberg consensus on law enforcement - and all the crime-fighting success that entails - it's no accident. De Blasio appears to be already looking down the road to make the general election against Lhota a referendum on relitigating the Dinkins-Giuliani races, as evidenced by his reaction to a report (denied by the Lhota camp - "David Johnson has zero affiliation with our campaign and no one on our campaign has ever heard of him") that a Georgia pollster had been polling the public's reaction to de Blasio's interracial marriage:
"I did see the Lhota campaign try and distance themselves from it," Mr. de Blasio replied when asked about the report. "I hope that's true. Because I know who Joe Lhota worked for. He worked for Rudy Giuliani; he was the top deputy for Rudy Giuliani when Rudy was dividing this city as a matter of political strategy."
Given the relative public standing of Dinkins and Giuliani, de Blasio's desire to refight those elections and tie himself and Lhota to Dinkins and Rudy seems mystifying. But by his own admission, de Blasio is a Dinkins guy all the way down:
[After working on his campaign as a volunteer coordinator, w]hen Mr. Dinkins won, Mr. de Blasio secured a job as a City Hall aide, a four-year position "foundational for everything I've done since then," he said. Not only did Mr. de Blasio acquire a taste for politics, he made a series of instrumental contacts, including another young Dinkins aide named Chirlane McCray, whom he met in 1991 and eventually married.
A key element in de Blasio's appeal to liberal activists and older African-Americans has been attempting to revive the reputation of Dinkins, New York's first black mayor. The Dinkins mayoralty meant a great deal to de Blasio, who met his wife and came of age in local politics while both were working for the Dinkins administration. Dinkins, according to de Blasio, was, unbeknownst to the public, a staunch and effective crime fighter. His problem, argues de Blasio, was communications, not substance, though many who lived through the paralyzing fear of the Dinkins years would probably disagree.
Ironically, Dinkins himself - playing to caricature to the end - joined Charlie Rangel in endorsing Thompson rather than de Blasio.
Higher Taxes: What Can't They Do?
The other area where de Blasio has stuck out is taxes: while the Democratic candidates have mostly been non-committal on property taxes, de Blasio pledges to raise income taxes on the City's top tax bracket. There are two obvious problems with this, even beyond the usual problems with tax hikes. One is that these are the very same taxpayers who were already hit when President Obama raised the top federal tax rate by letting the Bush tax cuts expire in 2012, and when Andrew Cuomo raised the top state tax rate in 2011; going back to that same well could push the top marginal rate to a very bad part of the Laffer Curve. And second, for reasons of New York state law and economics, the City can't raise income taxes without the permission of the State Legislature in Albany, and so de Blasio's plan would be dead on arrival in the GOP-controlled State Senate.
That leaves the other big piece of de Blasio's economic message, union organizing in the private sector, an approach that won him the endorsement of SEIU Local 1199, the largest union in the City:
Public Advocate Bill de Blasio, the front-runner in recent polls, has been most explicit about the role of the mayor in such efforts. Asked on Wednesday about his approach to private-sector unionization, he said elected officials "have to think like community organizers."
His community-organizer approach has come with predictable downsides. Though it's gotten less attention than Liu's campaign finance scandal, de Blasio is dogged by his own scandal, deriving from his support by the Working Families Party, an all-but-socialist third party (the WFP is closely akin to the PIRGs, ACORN and similar groups). The US Attorney's Office launched an investigation into the WFP and de Blasio shortly after his 2009 election as Public Advocate, based on accusations during the campaign that the WFP was pervasively violating campaign finance rules. In 2011, the WFP paid $100,000 to settle state and federal lawsuits charging it had used a for-profit arm to provide below-cost services to campaigns as a facade to provide unreported contributions, something the City's campaign finance board had warned de Blasio and other WFP-backed candidates about during the 2009 race. In 2012, a special prosecutor was appointed to dig further. Earlier this spring, the WFP itself went to court to fight subpoenas in the special prosecutor's investigation.
(de Blasio's other job after the Dinkins years was with Andrew Cuomo and Kirsten Gillibrand in the Clinton-era HUD, which started the nation down the road to the 2008 housing crisis. It's not hard to see why de Blasio would be hesitant to tout that.)
The City's 20 years without a Democratic Mayor is a vivid illustration of Robert Conquest's First Law ("everyone is most conservative about the things he knows best"); even leading liberals like Josh Marshall aren't willing to claim that de Blasio's economic policies will actually work:
The interesting thing about de Blasio...is that he is running as an unabashed progressive....[F]or three decades rising economic inequality has been a cornerstone of the Democratic critique of the direction of the country. It's been a theme of many campaigns. Yet most elected Democrats, particularly those in executive positions, have shied away from implementing the set of policies that might actually change or ameliorate the trend...I don't know if those 'things' will work in the big picture. (That's not just a throwaway line. I'm cautious and somewhat skeptical about our ability to shift these trends through policy....)...
Where is the GOP in all of this? With a much more low-key 3-man race of its own, but really two-man:the most conservative candidate, George McDonald, has drawn positive notices everywhere he's gone but hasn't gained any traction and lacks the resources to get his message out. That leaves the unpronouncable against the unspellable: Lhota (the "H" is silent), endorsed by his old boss Rudy, against Catsimatidis, a Greek-born self-made grocery billionaire endorsed by George Pataki. Catsimatidis' immigrant-makes-the-American-dream bio may be inspiring, but as a candidate he's been prone to all the well-known pitfalls of clueless-rich-guy, consultant-meal-ticket GOP politics, and has made his closing argument that Lhota can't win because he doesn't have enough money. Catsimatidis also doesn't sound like a guy who is willing to go for de Blasio's jugular. Then again, the "Cats Man" did manage to put Lhota in the position of having to make this classic debate denial:
Lhota was forced to answer for a controversial statement he made to New York before Labor Day, when he said he would not have stopped train service to rescue beloved vagrant kittens August and Arthur. "I'm not the anti-kitten candidate," Lhota insisted. "Let's talk about the facts, let's talk about the real facts here. First off, as you all know, I have pets. I love pets. I grew up with cats ... We have thousands of cats, literally thousands of cats, that are in the subway system every single day, day and night, scurrying across the tracks and they don't get killed." The remark removed all doubt that lost and adorable cats would be left to fend for themselves under a Lhota administration.
We get daily mailers from both these guys, and the latest Lhota mailer has more pictures of Rudy than of Lhota; at the end of the day, while Lhota knows he needs Democratic votes to win, he can't avoid the shadow of his old boss and shouldn't try. Lhota isn't a perfect fit for conservatives (given his stances on abortion and same-sex marriage) or libertarians, given his support for stop-and-frisk, but he's easily the best thing GOP voters could have hoped for after Police Commissioner Ray Kelly declined to run. Matt Welch of Reason Magazine argues that Lhota would actually have something to offer to fans of smaller government; as Lhota told the NY Post:
Asked how he differs from Bloomberg, Lhota cites health initiatives in general and Bloomberg's bid to bar sales of super-sized soda in particular.
Tomorrow, we find out if this three-ring circus is finally ready to reduce to a two-man race.
July 17, 2013
POLITICS: Fear of the Missing White Voters
RealClearPolitics election analyst Sean Trende has come under coordinated red-hot rhetorical fire from the Left for his thesis that one of the major causes of Mitt Romney's loss in 2012 was that a disproportionate number of white voters - mostly downscale whites outside the South - stayed home. Much of the criticism of Trende's thesis is based on deliberately misreading his policy prescriptions - but it's also based on a simpler failure to grasp the basic math behind his calculations. Like any exercise in reading exit polls and census data, Trende's assumptions (which he lays out explicitly) can be critiqued by people who are serious about understanding the issue; there are no definitive answers in this area other than final vote counts. But the vehemence directed at Trende's number-crunching suggests a Democratic establishment that fears honest debate intruding in its narrative of an inevitable, permanent Democratic majority built on a permanently racially polarized electorate.
Given the intense and growing racial polarization of the Obama-era electorate, it is sadly necessary to look at the race of voters to make sense of what happened in 2012 and what it says about the two parties' coalitions going forward; on this, analysts on all sides agree. Indeed, those who argue for a long-term Democratic majority do so primarily on the basis of maintenance of current levels of racial division. It is also agreed among all analysts that turnout was down in 2012 from 2008; the raw numbers show that about 2.2 million fewer people voted, while the population grew. The issue is how to measure the rates of turnout among each racial group.
Who The Missing Voters Were
Trende's original thesis was based on the initial exit polls released immediately after the election as compared to Census Bureau population estimates, and developed in a four part series beginning last month. Naturally, given the nature of the data sets involved, his numbers changed as more precise sources of data became available. He conducted a simple five-step exercise:
1. Look at voter turnout - total and by race - in the 2008 election;
2. Look at Census data to determine the growth of eligible voters in each racial category;
3. Project what the 2012 electorate would have looked like if each category turned out at the same rates as in 2008, but adjusted for the 2012 population;
4. Look at voter turnout - total and by race - in the 2012 election;
5. Compare Step 3 to Step 4 to determine how each group's rates of turnout changed from 2008 to 2012.
This is not a controversial methodology; total population growth and total election turnout are hard data, and the only real issues are which of various Census reports you use to compute population growth by racial category, and which of various election sources you use to compute turnout by racial group. With a little cutting and pasting to combine his two charts, here is what Trende's June analysis concluded:
In other words, compared to 2008 levels of turnout, white voter turnout was down far more than non-white voter turnout (6.2% vs 3.8% for black voters and 1.6% for Hispanic voters), and there were approximately 6 million "missing white voters," as compared to about 871,000 black and Hispanic voters. Trende also finds about 1.3 million missing "other" voters ("other" being mainly mixed-race voters, as well as Native American, South Asian, and other groups - not necessarily a bloc as heavily Democratic as black or Hispanic voters). The "other" group is a statistically significant part of the analysis, but, as Trende's later analysis shows, that last figure may be an anomaly due largely to mathematical rounding issues, without which the number of "missing" non-white voters in total drops in half when you use later, more accurate data - more on that below.
There is one error in Trende's computation, which brings his total short of the 129.2 million votes cast in 2012, and that's Asian voters (who have broken heavily Democratic in recent elections after being a GOP voting bloc in the Reagan-HW Bush years). Trende finds about 70,000 Asian voters missing, when in fact Asian turnout was up enough that he should be showing about 575,000 extra Asian voters (I contacted Trende and he confirmed this). Asian voters are an oft-overlooked and growing piece of the puzzle, and they still turn out in very, very low numbers compared to their (still-small) share of the US population, but reaching out to them is an important consideration going forward. In any event, when you adjust for the proper counting of Asian voters, you find that it actually strengthens Trende's thesis that white voter turnout was down relative to turnout of the major non-white voting blocs.
Where The Missing Voters Were
Trende also produced a map showing where the missing voters were most likely to be found, although the map can't break them out by race; the areas in blue on the map represent the biggest drops in turnout, red represents growth in turnout compared to population growth.
As you can see from the map, a good number of the "missing" voters were in uncontested states like New York and Oklahoma where they would not have made a difference. And the big chunks of deep blue in New Mexico, Oklahoma and South Dakota (as well as the heavy dropoff in turnout in Alaska, not shown on the map) are consistent with a dramatic decline in Native American voter turnout.
But a significant number of others were in Pennsylvania (Obama by 309,840 votes out of 5.75 million cast), Ohio (Obama by 166,272 votes out of 5.59 million cast), Michigan (Obama by 449,313 votes out of 4.74 million cast), and Minnesota (Obama by 225,942 votes out of 2.94 million cast). This is consistent with Trende's conclusion that - while these voters were not, in and of themselves, the cause of Romney's loss - they were a contributing factor large enough to consider, and one that may loom even larger in a closer future contest between a better Republican candidate and a Democrat who has less visceral appeal to non-white voters. (The lower turnout throughout the Northeast also surely reflects the influence of Hurricane Sandy).
The White-Voter Path To GOP Victory
Later in his series, Trende moved on to a second thesis: that it's possible - not likely, but possible - that depending how turnout develops (eg, if African-American turnout and voting patterns revert to pre-Obama levels), that the GOP could start winning national elections on the basis of winning a growing share of the white vote without eroding the Democrats' hold on non-white voters. As Trende notes, while this scenario requires some leaps from where we stand today, white voters have been trending gradually more Republican:
Democrats liked to mock the GOP as the "Party of White People" after the 2012 elections. But from a purely electoral perspective, that's not a terrible thing to be. Even with present population projections, there are likely to be a lot of non-Hispanic whites in this country for a very long time. Relatively slight changes among their voting habits can forestall massive changes among the non-white population for a very long while. The very white baby boom generation is just hitting retirement age, and younger whites, while unsurprisingly more Democratic than the baby boomers (who, you may recall, supported George McGovern), still voted for Romney overall.
Nowhere does Trende argue for the GOP to turn up its nose at Hispanic outreach, or counsel a harder line on immigration; rather, he argues simply that there are enough different variables that it's unwise to write off the GOP just yet on the basis of mathematical and demographic determinism, even if the GOP does defeat the current iteration of "comprehensive immigration reform." There is more than one way to build a winning electoral coalition.
The main salvo against Trende can be found in a belligerent ThinkProgress blog post by Ruy Teixeira and Alan Abramowitz entitled, "No, Republicans, 'Missing' White Voters Won't Save You." Stripping away the rhetorical overkill ("GOP phone home! Your missing white voters have been found, and it turns out they weren't really missing"), the main point of contention is that Teixeira and Abramowitz simply reject the notion that turnout was down at differential rates:
Trende was using an estimate of around 2.7 million additional eligible whites between 2008 and 2012. That's wrong: Census data show an increase of only 1.5 million white eligibles....[U]sing Census data on eligible voters plus exit poll data on shares of votes by race, we calculate that turnout went down by about equal amounts among white and minority voters (3.4 and 3.2 percentage points, respectively).
This attack on Trende was predictably amplified by Paul Krugman, who doesn't seem to have even read Trende's essays, calling them a "Whiter Shade of Fail"; Krugman concludes on the basis of reading the ThinkProgress blog post that "the missing-white-voter story is a myth." (Josh Marshall takes a similar line).
But careful reading is your friend. And a careful reading shows why Teixeira and Abramowitz are long on vitriol - because they are short on trustworthy data.
The immediate problem here is that Teixeira and Abramowitz don't show their work to explain how they come up with these percentages, so a certain amount of deductive detective work is required to figure out what they did (not making your computations transparent is generally not a sign of confidence in your data). From the links in their post, it appears that the main issue is that they and Trende are using different Census Bureau reports for their data. Also, critically, Teixeira and Abramowitz don't break out turnout among the component elements of "non-white" voters, who they treat as a monolithic mass.
The CPS Bait and Switch
As Trende observes, Teixeira and Abramowitz "look at a different data set -- the CPS [Current Population Survey] data," a monthly survey in which people self-report employment data and (after the election) self-reported voting participation. Trende, by contrast, used Census Bureau population estimates derived from the actual 2010 Census.
If you take it at face value, however, the CPS survey has a serious flaw that should be obvious even to a Nobel Prize winner:
[T]he CPS data conclude that there were 1.4 million more Hispanics who voted in 2012 than in 2008, 547,000 more Asians, 1.7 million more blacks, and 2 million fewer whites. That works out to a total of 1.8 million more votes cast in 2012 than 2008, according to the CPS survey.
In short: the CPS turnout figures cannot possibly be correct. It's like a preseason baseball prediction where the whole league is over .500. It's mathematically impossible. Now, it's certainly possible that CPS is wrong proportionally - that is, that it overreported the turnout of all groups equally. It's also certainly possible that it's not proportionate. But there's no way from looking at CPS alone to know, so relying on it as an authoritative source without caveat or explanation is a very questionable choice.
Teixeira and Abramowitz stake their whole argument on the CPS - but then they use it selectively. As noted above, the actual turnout figures produced by CPS support the idea that white turnout was down in absolute terms, while black, Hispanic and Asian turnout was up. If they broke these groups out individually, as they appear in the CPS data itself, that would destroy their entire argument that turnout was down equally across all groups. So they do two things to cover their tracks. One is to clump these groups together with "other" non-whites; but as Trende notes, "the large mass of missing 'other' voters is probably a rounding issue. This isn't a minor point; those voters represent 60 percent of all the non-whites that Teixeira and Abramowitz are discussing."
Second, as Trende demonstrates, Teixeira and Abramowitz are only able to use the CPS data to their advantage by mixing and matching it with other sources (specifically, exit polls) - if you use only the CPS, "the CPS data actually show a larger decline in the white vote than do the exit polls." (Trende, because he's using non-election-related Census data, has to use the exit polls for his turnout figures - but if Teixeira and Abramowitz think CPS is the more reliable source, why do they avoid using it to compute turnout?)
There is no perfect answer to these questions. The Census is the best possible population figure, but the interstitial estimates involve some inherent guesswork. Exit polls may be biased in who answers them, and the CPS is obviously biased to over-report voting and may be biased in who over-reports; we can't know. (One difference being that exit poll respondents don't know who won the election when they respond; CPS respondents do, so there may be a possible bias towards overreporting by non-voting supporters of the winner. But that's speculation.) What we do know is that Trende has put his methodological choices on the table and they are reasonable ones; Teixeira and Abramowitz have not, nor offered any defense for their methods, nor explained how they can square their theory of perfectly proportional decline in turnout across groups with the fact that the very source they use shows the opposite. Under those circumstances, it's not hard to decide who to trust.
The Wider Universe of Missing Voters
For all the heat over Trende's computations, it should not be forgotten that the "missing white voters" are only the difference in turnout patterns between 2008 and 2012 - both elections in which uninspiring and poorly-organized GOP campaigns faced off against Barack Obama (a uniquely inspiring figure to non-white voters due to his status as the first non-white President), and the first of which - the baseline - already involved a uniquely bad political environment for Republicans. In fact, voter turnout is a volatile variable that changes from one election to another; while it can be useful to perform an exercise like Trende's, it requires a serious failure of imagination to regard the 2008 and 2012 turnout environments as the outer boundaries of potential voter turnout.
How many voters are "missing" depends very much what your baseline is - a baseline that never stops moving. It's debatable exactly how many eligible voters there are at any given time (different sources use different measurements) but consider that Michael McDonald of George Mason University (on whom Teixeira and Abramowitz rely) figures a "voting eligible population" of 221,925,820 in 2012 - which means that compared to the entire universe of eligible voters, there weren't six or eight million missing voters, there were 92.7 million missing voters, 40% more than the total that voted for Obama. On the other hand, McDonald calculates that, while 58.2% of eligible voters voted in 2012, only 51.7% voted in 1996 when Bill Clinton ran for re-election. If you take McDonald's figures and use 1996, the last election with an incumbent Democrat on the ballot two years after a GOP rout in the Congressional midterms, as your baseline, suddenly you're not talking about missing voters at all - you're asking where 8.4 million extra voters came from.
I do not, of course, mean to suggest that there were 92 million voters that either campaign could reasonably have expected to turn out. My point is, the dropoff of some 6 million eligible white voters and 1.6 million eligible non-white voters as compared to the 2008 baseline is just one segment of a much broader universe of eligible non-voters, some of whom will doubtless be turned out by the winning presidential campaign in 2016 or 2020, just as some of the folks who turned out for Obama, Romney or McCain will surely drop out of the process in the next two elections even if they remain eligible voters. Turnout as a whole can be volatile over time, as McDonald's estimates show:
Turnout rates dropped as more previously ineligible voters entered the system, particularly after the voting age was reduced; it spiked in 1992 when Ross Perot's campaign offered an outlet to voters unhappy with both parties, and again in 2004 and 2008 as the Bush and Obama campaigns found distinctly different paths to bring previously disaffected voters to the polls. The increasing volatility of turnout rates in recent years suggests that improvements in technology, combined with changes in voting practices (e.g., early voting) may be improving campaigns' ability to locate new voters. And there's a second, equally important piece of the puzzle that I frequently stress: the two parties' bases, turnout methods and reasons for appealing to voters are asymmetrical. Look at this chart of the vote totals of the Democratic and Republican tickets in presidential races between 1976 and 2012:
The Democrat vote grew steadily year to year throughout this period, consistent with the view (implicit in all of Teixeira's analyses and those who follow a similar demographics-are-everything approach) that the Democrats are mainly a collection of interest groups that grow with the populations they represent. The Republican vote, by contrast, was much more volatile (and susceptible to being drawn off by dissenters like Perot), reflecting the fact that Republicans are a more ideological party and therefore more dependent on the issue environment (particularly the presence of national security as a major issue) and the quality of the party's candidate and platform to draw votes. Candidates and campaigns still matter, and matter more to Republicans. It's entirely plausible to think the GOP will run better candidates and better campaigns in the future - that the McCain and Romney campaigns were not the best of all possible Republican campaigns.
Specifically, we are not so far removed from George W. Bush and his Karl Rove-led political operation figuring out how to increase the GOP vote from 39 million voters in 1992 and 1996 to 50 million in 2000 and 62 million in 2004, a feat that astounded liberal observers at the time and upended conventional wisdom that the GOP could only succeed in a low-turnout environment. The 2004 election came after Ruy Teixeira and John Judis had published their "Emerging Democratic Majority" book in 2002, and Teixeira spent the 2004 election arguing so vociferously that the polls were overestimating Republican turnout that Mickey Kaus acidly remarked the day after the election "Bush 51, Kerry 48: Pollster Ruy Teixeira demands that these raw numbers be weighted to reflect party I.D.!"
Teixeira wasn't the first or last election analyst to assume that dramatic changes in the turnout environment were implausible; many observers on the right, myself and Trende included, spent a good deal of 2012 questioning how Obama could recreate the dramatic shift away from 2004's turnout that we saw in 2008. The point here is the danger of assuming that present trends will continue unabated forever.
In short, we're discussing the current margins - of the 92.7 million eligible voters who passed on the Obama-Romney contest, around 9% of those would have shown up at 2008 levels of turnout; of the 129.2 million who did vote, around 6% of those would have stayed home at 1996 levels of turnout. But until we run the next election, we don't know how far each side can push those margins, or with which populations of eligible but not certain voters. The history of American politics suggests that we have not seen the last new development that will surprise observers of the political scene.
Ruy Teixeira's Methods Seem Familiar
Reading through Texiera's flailing assault on Trende, I felt a strange sense of deja vu - because I had read this before, in Teixeira's review of Jonathan Last's excellent book What to Expect When No One's Expecting. And as with his dismissal of Trende, Teixeira's review was greeted by the usual head-nodders on the Left as an excuse not to deal with Last's arguments but rather dismiss them out of hand.
Last's book develops his argument that birth rates in the United States and around the world are falling to a point that threatens a declining population, that many changes in society, economics and government can flow from such a demographic shift, and that a lot of these could be very, very bad. Last's book notes that Hispanics (particularly recent arrivals to the U.S.) have been the only major group having enough children to keep the United States from falling below "replacement level" birthrates, but that trends among Hispanics suggest that they may also fall back towards the rest of the U.S. population over time. Finally, he argues that, while immigration has helped the U.S. stave off the more dire declines in population faced by countries like Japan, Russia and Southern Europe, there are downsides to relying too heavily on new immigrants to replace the native-born population, and reasons (especially given Mexican birthrates) to suspect that a steady supply of immigrants may dry up down the road.
Teixeira applies the same rhetorical sledgehammer to Last's carefully-researched, copiously footnoted and even-handedly argued book that he deployed against Trende: "If Last's claims sound hysterical and overwrought, that is because they are....If Last's claims about the impending population crash are fanciful, his claim that fertility decline will lead to economic collapse is completely ridiculous." But as with his attack on Trende, Teixeira's assertions don't stand up well under scrutiny.
To begin with, Teixeira's review of the book is astoundingly parochial: something like half of the book and scores of its examples (both anecdote and country-level data) look at birthrates around the world and in history, and a good deal of Last's argument addresses how the U.S. will be impacted by demographic changes in other countries, some of them very dramatically underway. But aside from one hand-waving reference to UN projections (more on which below) and a reference (which Teixeira refuses to engage) to Last's reliance on slowing Mexican population growth, Teixeira completely ignores everything happening outside the United States and all the book's discussion of history as an example.
Teixeira accuses Last of being "truly the man with a hammer who sees nails everywhere," yet the entirety of his critique of Last's solutions is to ask, "why not support immigration reform, as well as generally higher immigration levels?" and accuse him of being an immigration restrictionist who "just isn't very interested in seeing more immigrants in the country." It seems Teixeira is the one who only has a hammer, given that the argument for more immigration and growing political power for Democratic-leaning Hispanics is also the entirety of his attack on Trende, his attack on the 2004 polls and the 2002 book that made his name. As Last noted in response, if you actually read the book, you'd see that Last is not arguing against more immigration, just explaining why it's not the whole answer to every problem.
As for number-crunching, Teixeira didn't even bother to grapple with Last's marshalling and sifting of the demographic data; he just appeals to authority:
The Census Bureau does project that the fertility rate will diminish, but only by a modest .09 over the next 50 years. And while the fertility rate is likely to remain below the replacement rate for the next 50 years, the Census Bureau expects us to add another 100 million people by 2060 due to immigration and "demographic momentum." (Despite sub-replacement fertility rates, a relatively large proportion of the population will be in prime reproductive years for decades to come.) So much for population collapse.
At least this time, Teixeira looked at what the Census Bureau had to say. But he offers no reason why we would expect declining fertility rates to reverse themselves down the road, and while Last explains the implausibility of the UN's 2010 projection (which represented an abrupt and not credibly explained about-face from its prior stance), Teixeira incuriously accepts it at face value and then asserts these inherently speculative projections as fact. Last himself admits that many of the future projections involve uncertainty - but the past and current trends are hard facts. As Last notes, this is far from the only area in which Teixiera just hand-waves whole detailed sections of the book - unlike Teixeira, Last actually considers the experiences of other countries to see what works and what does not, rather than just blithely assuming that demographic trends will reverse themselves of their own accord.
Liberal pundits and Democratic activists - and the line between the two can be hard to locate - have increasingly overinvested in two excuses for insulating themselves in a bubble: that no data can possibly support any arguments by analysts on the Right, who can be dismissed with an ad hominem, a quick hand-wave and a lot of nodding, and that demographics alone will deliver them a permanent electoral majority without the need for their side to actually win any more arguments. These are hazardous trends, and the imbalance between Teixeira's rhetoric in dealing with pundits like Trende and Last and the actual substance of his critiques is an illustration of the dangers of the need to sustain this illusion at all costs to a writer's own credibility.
Posted by Baseball Crank at 3:30 PM | Politics 2012 | Politics 2013 | Politics 2016 | Poll Analysis | Comments (2)
May 13, 2013
POLITICS: PPP on the Brown-Warren Senate Race: A Polling Post-Mortem
Polls are back in the news, with the release of four public polls and an internal Gabriel Gomez campaign poll in the June 25 Massachusetts special Senate election to replace John Kerry. 3 of the 4 public polls show Ed Markey with a distinct but still surmountable lead, an average of 6 points; the fourth shows him up by 17 and looks like an outlier, adding 2.7 points by itself to Markey's lead in the RCP average. The Gomez campaign's internal poll shows Markey by 3; if you use the general rule of thumb that a campaign conducts multiple internal polls and will only release its most favorable internal, that's consistent with this currently being a 5-7 point race. Which is not a bad place for a Republican to be in Massachusetts five weeks before the election - it gives Gomez a puncher's chance in a special election - although you'd clearly still put better than 50/50 odds on Markey.
The closest public poll so far was put out by progressive Democratic pollsters PPP; its first poll of the race has Markey up by 4, 44-40. Let's take a look at how PPP polled the last Senate race in Massachusetts, the 2012 race between Scott Brown and Elizabeth Warren, which Warren ultimately won 54-46. That race may be less predictive of this one than the 2010 special election between Brown and Martha Coakley (in which PPP was one of the more reliable pollsters), but it's interesting as an exercise in examining how PPP samples the electorate.
Read More »
PPP v Rasmussen
PPP is one of two national pollsters - the other being Rasmussen - to publish regular statewide polls across the country, and that frequency in and of itself has made both pollsters into major players in public polling. At the moment, PPP is riding high, having posted a better record in 2012 than Rasmussen, while Rasmussen (who did some polling for Republicans in the past and retains a presence in conservative media) had built its reputation in better years for the GOP like 2002 and 2004. There's a chicken-and-egg question in there, as to whether pollsters with something of a partisan tilt are just getting lucky in years when the electorate favors their side; as I noted in my November post-mortem, the stricter likely-voter screens of pollsters like Rasmussen worked better in the turnout environments of 2002 and 2004, while PPP's looser screen better approximated turnout in 2012.
For an example of a race that Rasmussen polled much better than PPP, consider the 2010 North Carolina Senate race between Richard Burr and Elaine Marshall. PPP is based in North Carolina and closely tied to that state's Democratic Party; I have no idea whether it was doing non-public polling for the Marshall campaign, but its polls were clearly less favorable for Burr than Rasmussen or other pollsters. PPP conducted 11 of the 30 polls taken in 2010 that were included in the RCP average. 7 of those 11 polls showed Burr polling below 45%, including two that had him below 40%; only 3 of the other 19 polls conducted by other pollsters showed Burr below 45% at any point in the race. 8 of PPP's 11 polls had Burr with a single-digit lead; only 4 of the other 19 did. Between September 8 and October 25, the race was polled 8 times; the two PPP polls in that period showed Burr leading 49-36 and 48-40, while the other six polls all showed a lopsided race, with an average result of Burr 52, Marshall 35. PPP's final poll suddenly got with the program (Burr 52, Marshall 40), and the race ended Burr 55, Marshall 43. You can see in this graph the PPP and Rasmussen polling of that race, which Rasmussen also polled 11 times in 2010:
PPP's polling single-handedly drove the coverage suggesting that Burr was vulnerable, it was out of step with the other pollsters, and it was flat-out wrong. PPP was hardly the first or last pollster to blow a race, of course; the point here is to identify an example of a race where Rasmussen's approach was clearly a lot more reliable guide to what was happening than PPP's. All pollsters require careful scrutiny if we are to use their work product as tools for advancing rather than obscuring our knowledge of elections, and the halo effect surrounding PPP after the 2012 election should not exempt it from that scrutiny.
As we have seen before, PPP's methodology for polling differs from that of Rasmussen and other pollsters; in lieu of likely-voter screens, Tom Jensen (who runs PPP) has described how he models the demographic composition of his polls:
Jensen conceded that the secret to PPP's success was what boiled down to a well informed but still not entirely empirical hunch. "We just projected that African-American, Hispanic, and young voter turnout would be as high in 2012 as it was in 2008, and we weighted our polls accordingly," he explained. "When you look at polls that succeeded and those that failed that was the difference." Given the methodological challenges currently confronting pollsters, those hunches are only going to prove more important. "The art part of polling, as opposed to the science part," Jensen said, "is becoming a bigger and bigger part of the equation in having accurate polls."
And those hunches can change over the course of a race - consider PPP's shifting estimate of the racial makeup of the Florida electorate in 2012, as compiled by Sean Davis:
As you can see, PPP dramatically shifted gears based on something starting in its October 14 poll, and that enabled it to get closer to the final result as reported by exit polling. The vagueness of Jensen's public explanations of his methodology suggest a man who is not really sharing his secret sauce. My guess - which I stress here is pure speculation in the absence of evidence, so I'm happy to be corrected by Jensen - is that PPP may have been doing non-public work for OFA or some other Democratic or liberal-interest-group client, and gathered information (maybe about OFA's superior turnout operation, maybe about non-public polls) that was not generally available, and modeled its polling assumptions around that. (There's nothing illegal or necessarily unethical about doing so, but if that's the case, it would suggest that public polling is more driven by non-public information than you might think). If that's how Jensen formed his assessments of the electorate, then we can assume that PPP's turnout-projection models will continue to be superior so long as Jensen has access to better-informed sources of non-public information than other pollsters (it's clear that in 2012, Democratic campaigns knew things about turnout that Republican campaigns didn't, and Republican campaigns did not know any things about turnout that Democratic campaigns didn't; one hopes this will not always be the case, as it plainly was not in 2004).
On the other hand, if we take Jensen at face value and assume he really is just guessing, then his guesses are as variable as those of any other reasonably well-informed pundit. The 2012 Massachusetts Senate race is just one of many case studies from which we can look at how PPP's shifting assumptions about turnout interacted with its polling.
Brown v Warren
PPP polled the Brown-Warren Senate race in polls released on August 21, September 16, October 12, and November 2. (I'm citing here the dates the polls were publicly released; PPP generally takes polls over the prior 3 days). Here's the topline result of those polls:
PPP showed Brown with a 5-point lead in August, at a time when Brown held about a 1-point lead overall in the RCP average - but according to PPP, the race shifted 7 points in Warren's direction in September, 4 more points in October, and 2 more into November. PPP's final result actually understated Warren's margin of victory by 2 points. What changed? In each case, I will compare PPP's internals to the exit polls.
The truck-driving Brown against the archetypically schoolmarmish Warren was a race tailor-made for a gender gap, and what is perhaps most striking about PPP's polling down the stretch run of this race is the sharp turn towards a female-dominated electorate:
The exit polls - assuming those are accurate, which is never really certain - suggest that PPP went overboard in projecting that the Massachusetts electorate would be 55-57% female in 2012, and the final poll understated the gender gap, which ended up closer to what PPP's October poll had estimated. That said, clearly the shift to more female voters in the sample had a measurable impact on Warren's support. For what it's worth, the current poll showing Markey up 4 also has a 55% female sample, although the gender gap is less pronounced (Markey leads 46-36 with women, Gomez leads 45-41 with men).
As it did in Florida, PPP started upping the non-white share of its sample in Massachusetts near the end of the race. But this time, it overshot the mark: the exit polls showed an electorate that was 14% non-white, lower than the 16-18% in PPP's polls from September-November.
A share of the vote that small is bound to be "noisier" - that is, there's a small-sample-size problem where it's harder to get enough non-white voters to poll them reliably if they're less than 20% of the people in your overall sample. While the trend in PPP's poll was a dramatic dropoff in Brown's non-white support, the actual numbers are just nuts: there's no way Brown went from leading among non-white voters with 47% of the vote in August to down 35 points in November to losing by 72 on Election Day. It's possible that there's some instability here among whether Hispanic voters were reporting themselves as white, but that result just doesn't make any sense at all.
Among white voters, PPP's final poll showed Brown pulling ahead for the first time since August, even as his overall standing was at its low ebb; the exit poll showed him winning white voters 51-49 but losing the race due to his poor standing with the remaining 14% of the electorate.
PPP's Gomez-Markey poll has an 84% white electorate with Gomez down 2 points among white voters (41-43), underperforming Brown but with a lot of undecideds, and 13 among non-white voters (36-47), a showing with non-white voters that's consistent with Brown's August-September polling but wildly inconsistent with the exit polls.
The other big dividing line in 2012 was the youth vote. Let's look at what PPP's poll sample looked like by age:
The number of 18-29 year olds answering the September 13-16 poll dropped in half from August and didn't recover until November - that could be the result of any number of anomalies, but I have to assume that it had something to do with it being the start of the college calendar (Massachusetts has an enormous proportion of college students). But even so, not one of PPP's polls matched the 19% under-30 turnout of the exit polls (bear in mind here, however, that younger voters may be more likely to answer exit polls). The youth-voter sample is, again, noisy - Warren was bleeding Brown's support among young voters into October, but PPP's final poll had Brown winning 50% of the under-30 vote; the exit polls had him lose it by 12 points.
(Note one other minor methodological discrepancy: PPP's use of an age 30-45 bracket, whereas the exit polls used a 30-44 bracket).
The Gomez-Markey poll had just 9% voters under 30, which may not be a bad estimate for a special election after the end of the college year, with Gomez down 40-33 among young voters.
The great polling debate of 2012 was weighting of polls by party ID and the corresponding importance of independent voters. Massachusetts is a strange animal: independents are traditionally close to half the electorate, while there are more than twice as many Democrats as Republicans. Most independents in Massachusetts are not closet Republicans, as the usual strength of statewide Democrats can tell you, but many are persuadable - and given the tiny number of self-identified Republicans, a Republican candidate needs to clean up in a large way with independents to win. Brown's support among independents never faltered in PPP's polls or the exit polling (although if both sets of polls are treated as credible, it would suggest that Warren corralled all the late-deciding independents), but the partisan polling shows a different story:
The October PPP poll oversampled Democrats and the August poll with Brown up 5 actually undersampled Republicans, but otherwise PPP's polls were fairly consistent in their partisan sampling for this race. But what PPP was reporting was a steady consolidation of Democrats behind Warren as she made the race more ideological and put the primary behind her. On the other hand, every PPP poll, especially the final one, understated Brown's support among his own party.
(By contrast, the self-reporting of voter ideology was about the one element that was very consistent month to month in PPP's polling, but I've never really put much stock in self-reported voter ID).
The Gomez-Markey poll had a D-heavy sample at the expense of independents, 41/17/42 D/R/I, with Gomez up 47-31 among independents, Gomez drawing 21% of Democrats and Markey drawing 17% of Republicans - all signs of a race the voters haven't really focused on yet.
The new conventional wisdom on polling is that the key to accurately forecasting elections is getting the demographics right: if you predict who will vote, by race, gender, and age, you can pretty accurately profile who they will vote for. Yet despite PPP's status as one of the icons of this brave new era of polling, its record in the 2012 Brown-Warren race - a race in which its topline results ended up on the mark - suggests that it was actually better at modeling the electorate by party ID and ideology than by race, gender or age, that its assumptions about the demographics of the electorate changed a good deal from month to month, and that within each demographic group its reported results were often highly unstable.
Given the difficulty of accurately projecting turnout in special elections, even the polling averages have to be taken with some grain of salt in the best of times, but polling in the age of Obama is ever more art than science.
« Close It
November 16, 2012
POLITICS: Sometimes, It Really Is Different This Time - A Polling Post-Mortem (Part III of III)
This is the third and final part of my three-part polling post-mortem. Part I here looks at the national and state polls, and Part II at the likely voter screens and the electorate.
V. Polls vs. Non-Poll Tools
One of my premises in reviewing projections of turnout was that other items of information besides the polls were worth reviewing. Many of these same indicators favoring Romney in 2012 had forecast the rise of Obama in 2008. Why did so many of them prove useless or misleading?
The Bush era saw a modern-historic rise in the partisan component of the electorate, i.e., the percentage of votes that were either Democrats voting Democrat or Republicans voting Republican - and while the partisan component has become more Democratic, the trend has not significantly abated under Obama:
As long as that remains the case, knowing the partisan composition of the electorate remains critical. Yet non-poll data on the topic proved elusive. The data point I stressed that failed most spectacularly was the Rasmussen and Gallup surveys of party ID. I don't regret looking at those; they had a proven track record in the past of being right, and I like looking at data that has a proven track record in the past of being right. Rasmussen's surveys had been right in presidential and non-presidential election years, in years before and during/after the rise of Obama, and had never before overstated GOP turnout. And the Rasmussen survey in particular is based on an enormous sample of something like 15,000 interviews a month. But both proved to be way off the mark: Gallup had the electorate at R+1, Rasmussen's final survey R+6. Despite their record of accuracy before 2012, I will almost certainly put no stock in those surveys again. It really was different this time.
On the other hand, I still stand by my scorn for TPM's party ID survey average; it was useless slop that failed once again. It got 2010 wrong, and 2012 too: even if you adjust the numbers upward proportionally from the 90.7% of the population it purported to survey, it projects a D+8 electorate of D 36/R 28/I 36, when the exits told us it was D 38/R 32/I 30. The survey assumed significantly more independent voters than Republican voters, but Republicans outnumbered independents at the polls, just as they have every year since 1980.
That said, as in 2008 and unlike in the off-year electorate of 2010, Republicans were at a recent-historically low share of the vote relative to independents, suggesting that their turnout problem was not solely one of high Democratic turnout - this chart computes the GOP share of (Republicans+Independents), and the Democratic share of (Democrats+Independents), so as to avoid letting one party's turnout cloud estimates of the other's:
No two ways about it: there were not enough Republicans at the polls. The question for the GOP going forward is how to bring the people who stayed home or left the party back into the fold and the voting booth.
Then there's voter registrations; I relied on a bunch of studies showing that Democrats were registering new voters at a slower rate than 2008 and suffering a net decline in voter registration in key battleground states, while Republican registrations were up slightly and independent registrations were up dramatically. This hard data told the same story as the national party ID surveys and the voter enthusiasm self-reporting. While Democrats said they could just turn out the voters they'd registered in 2008, I was skeptical on two grounds: voters age 22-25 were likely to have moved since 2008, and voters age 18-21 could not have been registered then.
I may have overrated these problems. It would seem that OFA's digital outreach must have kept a handle on transient recent college grads. And we have yet to see final voter-registration figures; while 2008 featured yearlong registration drives, it's still possible that the Obama campaign just registered a whole lot of people in October and/or the day they voted. I'll be very surprised if we do not see, in the data that comes out after the fact, a surge in last-minute registrations.
There was also early voting and absentee ballot data; I didn't have systematic data, but lots and lots of the individual hard-data points, especially from Colorado and Ohio, showed that early voting and absentee ballot requests were up in Republican areas or among registered Republicans, and down with Democrats, at least compared to 2008. Many of these data points came from official state records; they were not just the usual vaporous campaign emissions about how many doors they knocked on. Yet again, all this data turned out to be misleading. For example, the Colorado Secretary of State at one point was showing an R+2 electorate in the state after 62% of the state had voted early. Coming from official records, that seemed to me a non-crazy reason to think the electorate would be pretty good for the GOP, given that early voting is more of a Democratic strength in most states. Exit polls showed the Colorado electorate ended up D+4.
Relatedly, one of the realities the GOP has to come to grips with is the extent to which early voting has changed both the process of turning out voters and the process of polling even as compared to a decade ago - early voting makes it easier to turn out less-motivated voters, but also harder to use traditional tools to figure out who will turn out. Many of the polls in October hugely oversampled early voters (you'd get samples that were around 40% early voters when about 20% of the state, according to official records, had voted early) - but of course, with voter turnout overall below 60% of the voting-age population, you probably do need to oversample people who you now know are 100% certain to vote, if your sample is going to reflect final turnout. I suspect that at least in some states, the polls taking a turn towards Obama at the end reflected, not a change in public opinion, but a change in the poll samples as more of Obama's early vote got locked in. That suggests that past patterns in how the polls moved at the end of a race in the days before early voting may be a poor guide to how they will move in years to come.
Another indicator I factored in, from within the polls, was polls of self-reported voter enthusiasm. Many, many polls reported GOP voters more enthusiastic about voting. Such polls have been indicative of an "enthusiasm gap" borne out on Election Day in the past, including in 2010; they were not this time. Ditto the less scientific indicator of the large, enthusiastic crowds Romney and Ryan drew on the trail. By contrast, one thing I didn't put a ton of stock in, small-dollar donations, favored Obama, and in retrospect it was probably a sign of the effectiveness of his digital outreach (the much-mocked three-a-day fundraising emails), and a proxy for real base enthusiasm just as it had been for Bush in 2004. Romney never really did particularly well with small donors.
I also failed to consider that Dick Morris predicted a Romney win, which should have set the probability of a Romney win to zero all by itself.
All of which does make me wonder whether, despite my longstanding philosophy of wanting to use external sources as a sanity check on the polls, there are any left we can trust. If relative or in some cases absolute advantages in voter registration, early voting, absentee balloting, party identification, and self-reported voter enthusiasm are not worth anything, we may be stuck trusting the pollsters' hunches - and may be blindsided the next time they are wrong.
VI. Presidential vs. Off Year Polling
Many of us quite reasonably thought that 2010 proved the GOP had recovered from its 2006 and 2008 wipeouts, and that we should expect an electorate in 2012 that looked at least as much like 2010 as like 2008; at a minimum, a midway point between the two, which would be D+3.5. After all, 2010 was the more recent sample, and both parties had contested it vigorously. But one of the real emerging lessons of 2012 is that we are in an age where turnout in mid-term elections is genuinely not predictive of the electorate that will show up in a presidential election, and vice versa. As with many things in the Age of Obama, it remains to be seen if this effect will persist after Obama is gone - but it is clearly with us now, and suggests both that (1) Democrats on the ballot in 2014 should not count on the 2012 electorate showing up and (2) even a strongly Republican-tilted electorate in 2014, if one resurfaces, will not tell us much about the 2016 electorate. Right now, I would not want to be Mark Warner facing the electorate that voted in Bob McDonnell by 19 points, or Ron Johnson facing the electorate that re-upped Obama by 7.
This chart shows each party's swing between the off year elections and the prior and subsequent general election:
As you can see, Republican turnout in the era from 1984 to 2000 was extremely steady every two years, in both general and off-year elections, around 35% of the vote. Democrats would go up and down relative to independents, but the GOP share was a constant. But since 2000, that has fluctuated much more wildly, with high GOP turnout in the 2002, 2004 and 2010 elections and low turnout in 2006, 2008 and 2012. That volatility is even higher than the volatility of the Democrats. What it suggests is, more or less, that there are a lot more casual Democratic voters than casual Republican voters - the GOP's determined base turns out rain or shine every two years except in a real washout like 2006, but the extra people who come out only every four years are (at present) composed more heavily of Democrats. That's terrible news if you're a Democratic candidate for Senator or Governor in 2014 (even aside from the usual carnage that attends a president's sixth-year elections), but it's also frightening news for Republicans considering the long-term strength of the party.
VII. Models vs. Averages
My criticism, and that of other informed skeptics on the Right, of Nate Silver's 538 model was on three grounds. First, most of the major controversies in this election cycle centered around how much faith to place in the state polling averages, a debate for now largely resolved in favor of the state polling averages. Since the 538 model runs on those averages, it successfully called the election - but so did the averages themselves, without the assistance of the model.
Second, the model has been oversold. This really has nothing to do with the model itself, and everything to do with making people understand that it was only as good as its inputs. That criticism still stands: as noted in Part II, the polls had to make some very unscientific adjustments to keep up with the electorate this year, and there are significant reasons to question their ability to do so in the future. If the pollsters' "hunches" are wrong next time, the model contains no mechanism to avoid failing just as it has failed in virtually every past instance where the state polls were wrong. If you view the 538 model as a way of aggregating imperfect inputs - like the RCP average, but with some additional bells and whistles - you can get value from it as an informed consumer. If you view it as an infallible Oracle to be obeyed, you are likely to sooner or later be disappointed.
Third, the most questionable part of the model is its projections of the likelihood of how late-deciding voters will break, which by definition is the part not anchored to the polls. (You can read Nate Silver's breakdown of past incumbent-challenger races here, and while as he notes it suffers from the usual small-sample-size problems of any presidential poll analysis, you can also see that the challenger has traditionally tended to gain more ground than the incumbent as compared to his standing in the October polls). This is an area where others in this field have done more work than I have, so I won't repeat the controversies, but one of my prior concerns was Ted Frank's point that the 538 model was placing heavy emphasis on the 2000 election in projecting that voters were less likely to break against an incumbent party when the Democrats are in office than the Republicans. Ted's point was that Bush's DUI story was an unusual end-of-race event not likely to recur here (I had a good deal of confidence that Mitt Romney had never been busted for DUI). But we did, yet again, have an end-of-the-race late-October surprise, in the form of Hurricane Sandy, and we did, yet again, have voters break towards Obama right at the end. Unless you place a lot of value on the ability of the media to spin a late-breaking story in the Democrats' favor, however (not a factor in Bush's case, since the story was self-explanatory), it's hard to see how you build a credible mathematical model that assumes this sort of thing will happen with regularity.
The model's usefulness in presidential polling is also not necessarily translatable to other races, especially in off-years when the electorate is not as predictable. There was no running 538 forecast this year, at all, for the Democrats' chances of re-taking the House (which they did not). In 2010, the 538 forecast in August 2010 gave Republicans only around a 60 percent chance of taking the House, and still had Democrats with about a 20% chance of holding their House majority as late as Election Day - a much higher chance than the model gave Romney of winning this year. But of course, the Democrats got shellacked in a landslide, far worse by historic House standards than Romney's loss by historic presidential standards.
As to the parts of the 538 model that go beyond just plugging in the state poll averages, I continue to take Bill James' view of expert and expertise:
"[G]etting the answers right" had almost nothing to do with the success of my career. My reputation is based entirely on finding the right questions to ask - that is, in finding questions that have objective answers, but to which no one happens to know what the objective answer is...When I do that, it makes almost no difference whether I get the answer right, or whether I get it a little bit wrong. Of course I do my very best to get the answers right, out of pride and caution, but it doesn't actually matter.
I add to that my own prior view of experts:
[T]he expert who learns that the recitation of jargon and the appeal to authority effectively exempts him from moral or social scrutiny has made the most dangerous discovery known to man: the ability to get away with virtually anything. Because if people will let you talk your way into money and influence with good science on the grounds that they do not understand it or have no right to obstruct it, what is to stop the expert from using bad science from accomplishing the same end, if they layman isn't equipped to tell the difference between the two?
We have not arrived now at the End of History or the End of Science. The polling controversies of past election cycles forced pollsters and poll analysts to learn important lessons. The polling controversies of this election cycle have, in my view, done the same. I wish my conclusions had carried the day this time, but I make no apology for challenging assumptions that were being treated as Holy Writ by liberals merely because, on this occasion, those assumptions proved correct. That's what I often do in my day job as a lawyer, in which I often encounter two contending experts with irreconcilable conclusions: probe their competing assumptions to expose what each side's conclusions assume to be true. There will always be a role for a Socrates, asking well-compensated analysts and pundits to explain themselves and put their assumptions on the line to be judged. The day we stop asking those questions is the day we let the "experts" know they can get away with anything just by hanging some numbers on it.
For the reasons explained in Part II, the old model of what kind of voter represents the swinging center of the electorate didn't work in 2012 - in fact, the center wasn't the decisive factor at all, but rather the huge margins in one corner of the electorate matched against a party that saw falling turnout among its natural base. Here in a single chart is the winning candidate's share of the two-party vote among five groups traditionally thought of as swing voters since 1972 - independents, suburbanites, voters age 30 and up, white women, white Catholics:
Mitt Romney's coalition among these five groups would have been the foundation of a clear national majority throughout the political era that ran from Harry Truman to George W. Bush, and the polling practices that grew up in that era would have captured that majority's formation. Pollsters have had to unlearn a lot of what they knew about polling in order to stay ahead of those changes in the electorate, and poll analysis does as well. The harder question will be how we can tell, other than after the fact, if the pollsters' guesses don't capture future shifts. Merely appealing to the idea that the majority of pollsters will always be right is an unsatisfying answer, especially given the follow-the-herd tendencies in the industry.
So the state poll averages were right, and really nothing that contradicted their narrative was. Does all of this mean that the state poll averages, and the models that run on them, will always be right in the future? Of course not. Anybody who has followed gambling prognosticators or stock pickers knows that winning streaks of couple of cycles do not always equal omniscience, even when backed by facially impressive-looking math. Unquestioning faith in mathematical models still has not been adequately called to account for its role in the 2008 financial crisis, for example. (On the other hand, climate models can only dream of the predictive success record of the state poll averages). Just because somebody gives you a prediction with numbers on it and is right a few times in a row doesn't mean they always will be. If you look at this as a science, you have to recognize that we don't have nearly enough data from presidential elections to constitute a meaningful sample size. And the fact that the poll averages were right because the pollsters changed the way they poll - in a world of ongoing technological and demographic change, and via methods that are themselves far from scientific - leaves us with a lot of uncertainties about whether they will make the right guesses again next time. Tom Jensen's next "hunch" could be wrong. There might be elections in the future in which polls using likely voter screens are more accurate than polls that all but abandon the project. Skeptical examination of the assumptions behind the polls' turnout forecasts will not go away, and should not go away.
But all that said, we're conservatives; we learn from experience, and even when the process is questionable, results talk. The case for trusting state poll averages over all other indicators, at least in the stretch run of presidential elections, has been strengthened a good deal by a third consecutive cycle of those averages calling the result right in 48 or 49 states out of 50. The case for treating other indicators as predictive of turnout has been weakened and in some specific cases pretty badly discredited. And while I remain a little less firmly convinced of the value added by modeling of how undecided voters will break at the end - over and above the value of the poll averages themselves - the 538 model had a good election in that regard.
A final word. While I've been following elections for a long time, I really cut my teeth reading polls in the 2002 and 2004 elections. I recall well from those races seeing a lot of polls that were registered-voter polls or polls with D-heavy samples very favorable to Democrats, yet the end results were much more favorable to Republicans. In the presidential elections of 2000, 2004 and even 2008 (before the financial crisis), we repeatedly saw the polls shift towards the GOP when we got past Labor Day and most pollsters started using likely voter screens. I learned a lot from that experience, some of which is clearly still true, and some not.
But I also saw a lot about human nature that is eternally true. A lot of Republican pundits and poll-readers looked like geniuses in that period by projecting Republican wins in a lot of the competitive races. The lesson, then as now, is that it is easy to look smart when your own side is winning all the close ones. It doesn't make you a bad or dishonest advocate for your side if you are better at predicting your side's victories than its losses. But it means you are still one side's advocate - and while I work hard to call things as I see them (I genuinely believed every word I wrote in this race), I make no bones about being an advocate.
But you have not really made it as a neutral arbiter of presidential polling - let alone a scientific one - until you have given both sides news they desperately do not want to hear. We will know Nate Silver has really made it as a presidential pollster when people on his own ideological side are screaming in terror at his conclusions, and not before.
Posted by Baseball Crank at 10:36 AM | Politics 2012 | Poll Analysis | Comments (8) | TrackBack (0)
November 15, 2012
POLITICS: Sometimes, It Really Is Different This Time - A Polling Post-Mortem (Part II of III)
The second part of my 3-part post-mortem on the polls and the 2012 election. See yesterday's Part I here.
IV. Likely vs. Registered Voters
A. Pay No Attention To The Man Behind The Screen
Near the heart of every major polling controversy this year was the issue of sampling and likely-voter screens. Polls traditionally report results among either "all adults" (whether or not registered to vote), "registered voters," or "likely voters." Historically, there's a well-recognized pattern: all-adults and registered-voter polls have tended to skew a couple of points in favor of the Democrats, and in the past this has usually been to the detriment of the accuracy of the polls. The pattern was especially pronounced this year; Bob Krumm noted before the election that Obama's strength in national polls was directly correlated with how lenient the poll's likely-voter screen was, and Nate Silver found the same effect in August, concluding:
It's all a bit of a mess, frankly. I suspect that part of the problem is that polling firms are applying likely voter methods that might have been designed 30 years ago to a modern polling universe of extremely low response rates (even the most thorough polling firms can only get about 10 percent of voters to return their calls), cellphone-only households, and an increasingly diverse and partisan electorate - and that is producing erratic and unpredictable results. There's always some uncertainty about just who will turn out to vote, but there is more of it than usual this year.
He also noted at the same time that Obama's support was strongest among those poll screens considered least likely to vote.
Those screens have worked in the past; if they didn't, the poll averages would have been useless a long time ago. This is why one of the regular rules of thumb in reading polls is that if a campaign is citing polls of registered rather than likely voters, especially late in the campaign, it's doomed. Yet that's exactly what Obama supporters were doing in the closing weeks, and more or less what Jim Messina was saying even after the election was over. For once, the registered-voter numbers were more accurate than the polls that put rigorous effort into likely-voter screening. The question is whether the pollsters actually had a good reason to do this, or whether they just got awfully lucky.
Part of what is supposed to make polling valuable is pollsters' ability to judge which voters are likely to show up to vote. They get to the likely-voter number by first, constructing a sample of registered voters, and second, applying a series of screening questions to determine which of those voters is likely to vote. That's what Messina was talking about in his argument that "traditional polling" was "broken" - the Obama campaign's theory throughout the election was that pollsters using the kinds of likely voter screens that have worked in the past (like Gallup and Rasmussen) would be wrong this time. The lesson we learned, at least this year, is that Messina was right and the traditional, professional pollsters were wrong - and that the nature of Obama's coalition made the application of likely voter screens particularly likely to affect the accuracy of the polls.
But determining which voters are likely to vote is the part of polling that is most inherently subjective and least scientific. Moreover, pollsters are often not very forthcoming about how they make these determinations, so sometimes when they release a poll, the best you can do is compare the number of registered and likely voters in the sample. As a result, a certain amount of deductive work is required to figure out why some polls give different results from others. There's only so much we can know from the outside, but it appears that pollsters like Gallup and Rasmussen were, for much of the cycle, using likely-voter screens that made traditional assumptions about who would make the effort to vote - and those assumptions just didn't hold this year, as illustrated by Gallup's final poll envisioning an electorate that was 78% white (the exits said 72%). By contrast, a number of the polls that were later vindicated were reporting results that defied all historical precedents, classifying as many as 99% of registered voters as likely voters. Their process seemed problematic precisely because it was so different from the things that made polls trustworthy in the past, but they got results.
One of the pollsters that projected a Democrat-friendly electorate and ended up getting high marks in the post-election rankings of final polls was PPP, a Democratic pollster employed by SEIU and Daily Kos, among other clients whose identities are not known. PPP's overall accuracy throughout election cycles is a longer story, but they did end up having a good record at the very end. Here's Tom Jensen, the principal of PPP, discussing how his firm determines who is likely to vote:
Jensen conceded that the secret to PPP's success was what boiled down to a well informed but still not entirely empirical hunch. "We just projected that African-American, Hispanic, and young voter turnout would be as high in 2012 as it was in 2008, and we weighted our polls accordingly," he explained. "When you look at polls that succeeded and those that failed that was the difference." Given the methodological challenges currently confronting pollsters, those hunches are only going to prove more important. "The art part of polling, as opposed to the science part," Jensen said, "is becoming a bigger and bigger part of the equation in having accurate polls."
In other words, the successful pollsters this cycle were doing exactly the same thing the poll skeptics were doing: making a more-or-less informed guess as to what the electorate would look like and weighting their results to match that. As Neil Stevens notes:
I don't remember anyone willing to say PPP was actively rigging the polls to reach chosen results, but there it is in black and white. Jensen decided in advance what he wanted the electorate to look like, and so tweaked the numbers until he got what he wanted. This isn't a whole lot different from what Research 2000 admitted to doing, folks.
Research 2000, as you may recall, was PPP's predecessor as DailyKos pollster, but had to be canned for more or less manipulating its data to get to results it wanted; Kos eventually sued them for fraud, which was settled out of court. Here's what Nate Silver had to say about R2k at the time:
[I]n practice, a pollster will usually have enough knobs to twist between likely voter screens, weighting and sampling assumptions, etc., that they could back into almost any result they wanted more often than not. But there would usually be some scientific pretense for it.
In fact, Jensen's hunches changed over the course of the race. Sean Davis calculated the demographic composition of PPP's Florida polls over the course of the race, yielding the following percentages of white voters:
4/17: 71%...6/5: 70%...7/3: 69%...9/12: 70%...9/23: 69%...10/14: 66%......10/28: 64%...11/5: 66%
From April until September 23, PPP assumed an average white vote of 69.8%. From October 14 through November 5, PPP assumed an average white vote of 65%. What changed? Who knows?
It's entirely possible, of course, that Jensen has some other source of information he's not disclosing here, but taking him at his word, the "poll averages are science!" crowd should have just a little more humility about exactly what it is that they are placing their unquestioning faith in - Jensen believed that this year's electorate would favor the candidate he favored, and he was right, but right in roughly the same way pundits are right when they say their side will win. Nuclear physics, this is not.
How PPP got it right while others, including polling titans Gallup and Rasmussen, got it so wrong goes back to a difference in method for how the firms identify likely voters and how long they conduct a poll.
It also means the likely voter screen - like Jensen's hunches - is completely opaque to the consumer of the poll. All you see is the opinions of the people Jensen decides should be polled, and who agree to talk to him.
Even pollsters themselves conceded that the combination of demographic and technological changes had made their supposed science more inexact than ever. "We're in sort of what I would call polling's dark age," Jay Leve, who runs the polling firm Survey USA, told me earlier this fall. "We're coming out of a period of time where everyone agreed about the right way to conduct research, and we're entering into a time where no one can agree what the right way to conduct research is."
[Nate Silver's] appeal, of course, is that he's scientific. And last night, his science worked because the polls themselves worked. But as polls become more art than science, Silver's approach could become more tenuous. The good thing about pollsters - at least the good ones - is that they're constantly reassessing and tweaking their approaches. That's the bad thing, too, at least when it comes to having any certainty that about how they'll perform in the future.
B. Why The Screen Mattered So Much This Time
Why did these differences in projecting the electorate matter? In an ordinary election, they would not: ordinary winning presidential candidates have a broad enough base of support that you can see it coming pretty clearly without needing the right "hunch." But Obama was not an ordinary winning presidential candidate. The racially polarized electorate of the Obama era means that every slight shift in demographics can have an outsized effect on outcomes.
Let's look at what the exit polls tell us. 81% of the electorate was voters age 30 and up; Romney won those by 2 points, 50-48. Drilling into the state-by-state exit polls, here's a map of what the election would have looked like just among voters age 30 and up - losses with young voters cost Romney six states worth 95 electoral votes, more than enough to flip the election:
Historically, that is game-set-match; the last candidate to win the national popular vote while losing voters age 30 and up was Jimmy Carter in 1976:
Another 11% of the electorate was white voters under 30; Romney won those too, by 7 points, 51-44. These were Paul Ryan's "faded Obama posters" voters - they swung 17 points from Obama winning them by 10 in 2008. Obama's pop culture cache with young white voters had worn off by 2012 in the face of his record. That's 92% of the electorate accounted for, and Romney up 50-48 and with a decisive lead in electoral votes. In other words, the 8% of the electorate consisting of non-white voters too young to have voted in the Bush v. Gore race in 2000 accounted for the entirety of Obama's national margin of victory.
The "gender gap" was similarly a feature of race and racial turnout patterns. Romney won white women, who made up 38% of the electorate, by 14 points, 56-42; this was the biggest margin of victory among white women since Reagan in 1984. Obama in 2008 was the first winning candidate since Carter in 1976 to lose white women, but Carter lost them by 6, Obama last time by 7. Yet, Romney lost women overall by 11, 55-44. Why? He lost non-white women 85-15, including Hispanic women 76-23 and black women 96-3. Among non-white voters, Obama again maximized the group most favorable to him: black and Hispanic women were 14% of the electorate, compared to 10% black and Hispanic men, both of which Romney lost by less severely lopsided margins (Obama won black men 87-11 and Hispanic men 65-33). In other words, the black and Hispanic segment of the electorate was something on the order of 58% female; black voters were over 60% female. Romney had no similar redoubt of lockstep support - exit polls showed that even among Mormon voters, he didn't crack 80%. So polls measuring turnout had to match two highly asymmetric campaigns, one winning majority groups with support in the 50s and 60s, the other winning much smaller groups by enormous margins.
You can slice the exit polls a few different ways and see similar results along racial lines (without reference to age or gender). Nate Cohn notes that black turnout in general was key to winning Ohio, as black voters were 15% of the electorate there, up from 11% in 2008. If you look solely at white and black voters and leave out Obama's margins with Hispanic, Asian, and Native American voters, Romney wins five states he lost - Florida, Colorado, Nevada, Pennsylvania (very narrowly) and New Mexico - enough to swing the race:
Of course, white and black voters together are only 53% of the vote in New Mexico. Neither a campaign nor a poll analysis can safely ignore such segments of the electorate. But my point is that, because Obama's margin of victory (both nationally and in the critical swing states) was entirely the result of his outsized margins with very narrow but homogenous segments of the electorate, the accuracy of polls was highly sensitive to the relative size of this segment in turnout compared to other voters.
Yet, voters under 30 in particular have rarely been a reliable source of voter turnout; for years and years, it has almost invariably been the case that a campaign losing with the rest of the electorate but placing its entire faith on high turnout from young voters was a losing campaign. Even Obama in 2008 didn't do that: he won voters over 30, independents, and young white voters handily. His coalition was broader then, before he had a record.
It is true that Carter set a precedent in 1976 for appealing to the under-30 voters. But thanks to the Baby Boomers, the oldest of which were just hitting 30 at the time, voters under 30 were 32% of the electorate in 1976; today, thanks to shrinking birthrates and a graying population, they are just 19% and demographically likely to decline even further:
The last year in which under-30 voters were 20% of the electorate was 1992, not coincidentally the last election less than 20 years after Roe v. Wade. Take away 50 million abortions, and the demographics of the electorate look quite different in a race where the winning candidate will end up around 62 million votes. But while young voters are less numerous and traditionally a below-average turnout group, Obama for the second straight election cycle managed to increase them as a share of the electorate, closing in on their share of the population (Census data from 2000, 2004, 2008 and 2010 show 18-29 year olds as a steady 22% of the voting-age population; by contrast, with the Boomers graying, 30-44 year olds dropped in that time from 31% to 26% of the population, while 45-64 year olds rose from 30% to 35%). Here's the major age groups' turnout relative to their share of the general voting-age population:
And while the smallness of the age breakdowns among black and Hispanic voters creates rounding-error issues that make the math a little fuzzy, this chart illustrates rather vividly that the proportion of young voters among non-white voters as a whole was much, much larger than the proportion of young voters among white voters:
Voters under 30 made up somewhere north of a third of all Latino voters, compared to less than 15% of all white voters. Partly that, too, is demographics; the median age of Hispanics is 27 compared to 42 for white non-Hispanics. But it's also the case that OFA maximized the showing of the few loyal groups that provided its entire margin of victory. Rasmussen came to a similar conclusion in evaluating why his polls were off:
A preliminary review indicates that one reason for this is that we underestimated the minority share of the electorate. In 2008, 26% of voters were non-white. We expected that to remain relatively constant. However, in 2012, 28% of voters were non-white. That was exactly the share projected by the Obama campaign. It is not clear at the moment whether minority turnout increased nationally, white turnout decreased, or if it was a combination of both. The increase in minority turnout has a significant impact on the final projections since Romney won nearly 60% of white votes while Obama won an even larger share of the minority vote.
As Rasmussen notes, the demographic shift from 2008 could be higher non-white turnout, or lower white turnout (or both). Sean Trende has estimated that white voter turnout was down in absolute terms and in particular in proportion to white Americans' share of the voting-age population:
Had the same number of white voters cast ballots in 2012 as did in 2008, the 2012 electorate would have been about 74 percent white, 12 percent black, and 9 percent Latino (the same result occurs if you build in expectations for population growth among all these groups). In other words, the reason this electorate looked so different from the 2008 electorate is almost entirely attributable to white voters staying home. The other groups increased their vote, but by less than we would have expected simply from population growth.
The 74% would be in line with Rasmussen's assumptions, which were more reasonable than Gallup's projection of a 78% white electorate. Byron York has more on the collapse of white voter turnout in Ohio by about 200,000 voters, which led to Romney getting fewer total votes there than John McCain in 2008.
The actual proportions in the voting-age population depend on how you read the Census data and break it out to exclude the non-voting-age. The Census showed non-Hispanic whites as 63.7% of the overall population (of all age groups) in 2010, compared to 69.7% in 2000, dropping to 63.4% in the 2011 Census estimate. Looking at Pew Hispanic Center data, the nation's 215 million eligible voters are 72% white, 13% black, 11% Hispanic and 4% Asian; the electorate was 72% white, 13% black, 10% Hispanic and 3% Asian, which when you do the math means that 59% of eligible black voters voted, 58% of white voters, 53% of Hispanic voters and 41% of eligible Asian voters. In other words: high black voter turnout, especially by historic standards; low Hispanic and Asian turnout but rising by historic standards and in particular rising relative to the rest of the electorate:
Geographically, this map shows the distribution of states, with the higher percentages of non-Hispanic whites in darker blue:
Leaving race and age aside, other aspects of the exit polling, like the pre-election national polls and the internals of pre-election state polls, mostly present a picture of an incumbent president doomed to defeat in any ordinary political environment in recent memory. As noted, Romney won independents by 5 points; the last candidate to lose independents by more than 2 points and win the presidency or the popular vote was, again, Carter in 1976, who lost independents by 11 points but took advantage of a depressed, decimated and divided Republican base in the aftermath of Watergate and Reagan's primary challenge to Ford. With the economy the number one issue throughout the election, voters told exit pollsters they trusted Romney more than Obama, albeit narrowly, 49-48. 77% of exit poll respondents said the economy was in not so good or poor shape, and Romney won those voters by 22 points, 60-38. 59% of the voters cited the economy as the top issue; Romney won them 51-47, plus winning 66-32 among the 15% of voters who cited the budget deficit. And this exit poll question was perhaps the most dramatic of all:
In basically any American election before 2012, I would tell you with great confidence that a candidate, much less an incumbent, is toast if he (1) loses independents; (2) loses voters age 30 and up; (3) loses white women by double digits; (4) loses white voters under 30; (5) is less trusted than his opponent on the economy when 59% of voters cited the economy as the dominant issue in the election; and (6) loses voters who prioritized leadership, strong values or a vision for the future. That has never before been the electoral profile of a winning candidate. As I said before the election, if it that changed, we need to rethink everything we know about elections.
We do - at least, for now, when Obama's on the ballot. The OFA theory of the electorate was that "really, it's different this time" - that neither the economic doldrums nor any other factor would dampen the historic levels of enthusiasm for Obama among non-white voters under 30. And as it turned out, OFA was right: they turned out in numbers totally out of step with their historic turnout patterns relative to their share of the voting-age population, and delivered the entirety of Obama's national margin of victory.
Viewing this, to some extent, I feel like a guy who shorted the NASDAQ in 1998, reasoning that the tech bubble couldn't last forever, and ended up getting mocked by the guys who watched all their "new economy" stocks rise without an end in sight. History teaches us that those guys, of course, eventually lost their shirts - but they were awfully proud of their "this time, it's different" reasoning for quite a while. Time will tell if the believers in the new Democratic turnout model go the same way.
If it does, how will we know? Will polls run by partisan Democrats like Tom Jensen readjust their hunches? Or will we have to look outside the polls? I will look at these questions tomorrow in Part III.
Posted by Baseball Crank at 10:40 AM | Politics 2012 | Poll Analysis | Comments (4) | TrackBack (0)
November 14, 2012
POLITICS: Sometimes, It Really Is Different This Time - A Polling Post-Mortem (Part I of III)
As promised, a mea culpa on my pre-election poll analysis: why I was wrong, why the state poll averages were right - and why I'd say most of the same things if I had to do it over. I suppose I have lost a good deal of credibility with a number of people by making the kind of out-on-a-limb prediction I don't usually make, and being wrong. But my assumptions have always been out in the open. Let's examine why they led to the wrong answer, and which of those assumptions should be re-evaluated in the future.
I. More Evidence Is Better Than Less
Discussions of polling often lend themselves to more heat than light. A lot of the post-election poll commentary is even dumber than the pre-election poll commentary, as victorious liberals spin a narrative that conservatives were all "poll deniers" or "poll truthers" ignoring the polls. Now, it's true that there were more than a few people on the Right who made intelligent discussion of the polls harder rather than easier. It's also true that some of the efforts at "unskewing" the polls were unhelpfully ham-handed; in this pre-election essay I explained why Dean Chambers' unskewedpolls.com was more alchemy than science. Chambers was projecting a strong Romney lead in the polls back in September, when neither I nor almost anybody I knew believed that Romney was actually ahead. My mantra on Twitter in August and September, like that of many conservatives I respect, was simply that it was a close race, that Obama had problems with independent voters and hadn't closed the deal yet, and that there was still time for Romney to catch up to him. While a lot of Romney's problems were baked-in by mid-summer, I don't think that was an unreasonable view to take at the time; as it turned out, there would be twists in the race throughout October. But those of us who attempted to take the polls seriously, and drew conclusions from the polling evidence itself as well as external evidence, were not denying anything; we were just looking under the hood.
A. Believe The Polls, But Don't Believe Only The Polls
Let me start by restating my philosophy of polling, and indeed my philosophy of examining most any question. Polls are not reality. They are a tool for measuring reality. They are traditionally the best single tool, and polling averages make them a better tool by evening out the outliers. But they are not perfect tools: public polls have called the outcome of races wrong before, or been off on the margins of victory by a significant amount. There are reasons why campaigns spend a lot of money on their own polls, which - other than for purposes like push-polling or testing messages - they would not do if an average of public polls was as flawless a guide to the electorate as a thermometer. If you think public poll averages are an infallible predictor, then the uniform practice of actual campaigns is totally irrational. Indeed, listen to Jim Messina, the nuts-and-bolts guru of the Obama campaign, on his view of public polling:
Every night, Obama's analytics team would run the campaign 66,000 times on a computer simulation. "And every morning we would come in and spend our money based on those simulations," said Messina.
Is Messina a "poll truther" or "poll denier" for saying that "American polling is broken" or rejecting "traditional" polls? Was he wasting Obama's money by running his own state and local polls instead of just reading 538? Or is he reflecting the fact that public polls involve a certain amount of guesswork about voter turnout that can only be definitively tested each new election cycle by the final vote tallies?
The most recent set of publicly available polls are also not the only tool for measuring reality. There are all sorts of metrics - some more hard and quantifiable than others - that have traditionally been useful in assessing the state of play: voter-registration numbers, early voting data and absentee ballot data, trendlines in the polls, what the pollsters themselves are reporting about voter enthusiasm, and hazier indicators like small-dollar donations and the size of crowds on the campaign trail. It's hardly anti-empirical to examine these additional facts, and allow them to affect your conclusions. The people who correctly predicted that Harry Reid would hang on to beat Sharron Angle when the final polls said otherwise were looking at this kind of information. If all you did was read the RCP average or 538, you would have missed that.
None of these are reasons to disregard the polls; they should still be the primary item of evidence. But neither does it make sense to treat them with blind reverence and ignore all the external evidence - especially when (1) different sets of polls are in conflict with each other and (2) the internal breakdowns of the polls, which can act as a sanity check on the plausibility of the polls' assumptions, are telling you that the polls in question are predicting something that is historically unlikely to happen.
To go back to one more example from baseball analysis (sabermetrics), this is sort of like the "statheads vs scouts" debate: whether you can better project a player's future accomplishments by looking at his statistics or by listening to experts analyze his "tools." I'm in the stathead camp: I believe strongly in quantifiable data, and I place it at the center of just about any baseball analysis I do. In fact, I probably err more often than not on overemphasizing the numbers. But as Bill James, the father of sabermetrics, has often said, sabermetrics (or any other field of hard science, soft science or would-be science) is not the search for statistics, it's the search for truth, and in searching for the truth, you don't blind yourself to some of the available evidence simply because it doesn't fit in a column of your model. Most statheads today will tell you that no matter how good your statistical measurements and how highly you prioritize them, you can learn things from also listening to the scouts that you might not learn anywhere else. The scouts can't tell you a guy with a terrible stat line is actually a major star - but in a close case, their view can make the difference.
B. Regrets, I Have A Few
If I had it to do over, maybe I'd be less definitive in pronouncing Obama "toast," but for the most part I'd look at the same evidence, see the same things and draw the same conclusions I drew at the time. I do, however, have two main regrets.
First, which is an unfortunate side effect of doing punditry as a part-time unpaid hobby rather than a full time job, is that I should probably have spent more time breaking down the internals of all the battleground-state polls to see exactly how geographically consistent the national trends were: I had seen those trends confirmed in Ohio and a couple of the other key battleground states, and didn't do the deep, time-consuming dive into the internals of every single battleground state. As it turned out, Romney did very well with independent voters in a number of the battleground states, but critically underachieved with them in others and lost those states. Romney beat Obama with independents by 5 points or more in Ohio (+10), Virginia (+10), North Carolina (+15), Nevada (+8), and Pennsylvania (+5). In more lightly-contested states, he won them in Missouri (+26), Indiana (+11), New Mexico (+8) and Arizona (+6). But in two other battleground states, his margin with independents was narrow: Colorado (+4), and Michigan (+1). And he actually lost independents in Iowa (-14), New Hampshire (-7), Florida (-3), Wisconsin (-1), and Minnesota (-4). (It says something about Romney that his biggest failure with independent voters, especially in states with lily-white electorates that didn't present the demographic challenges apparent elsewhere, came in the places where the voters had been exposed at length to his scorched-earth 2008 and 2012 primary campaigns). In non-battleground states where exit polls were taken, independents mostly trended pro-Romney in red states - Alabama (+52), Mississippi (+34), Montana (+7) - and pro-Obama in blue states - California (-12), New York (-4), Illinois (-5), Vermont (-40), Maine (-28), Maryland (-12), Massachusetts (-7), Oregon (-7), Washington (-3), Connecticut (-3). In just two totally non-competitive states did independents buck the state's partisan tilt: Romney won independents in New Jersey (+1), and Obama tied him in Kansas. If Romney had won all the states where he carried independents, he would have won, but not in all the states I thought he would win:
I might have caught some of this with more time to commit to each state, although in some cases, that weakness came out only in the very late polls. Or not; for example, PPP had Romney up by 2 with independents in New Hampshire on October 28, and up by 7 with independents in Florida the same day.
Second, and far more importantly, I didn't do enough to re-evaluate my conclusion after declaring it on Twitter on October 19 and laying it out in detail on October 26, eleven days before Election Day - and that week turned out to be Romney's high water-mark in the national and state polls. Nationally, Romney led in the RCP average all but two days between October 9 and October 31, and Obama was below 48% in the average every day from October 8 to November 5. In four key states, Romney had seized the lead in the RCP average: in Colorado, he led from October 9-29; in New Hampshire, he led from October 19-21; in Virginia, from October 19 to November 2; in Florida, from October 8 to the end of the race. Those leads gave me confidence that Romney's momentum in the national polls was real and would leave him needing to pick off just one more state to win. Instead, he lost all four, and the state poll averages were predicting three of those four losses by Election Day. When I wrote my initial post, I assumed (given the lateness in the cycle and the evident deterioration of Obama's position in the national polls and with independent voters by October 26), that there was not time for any new game-changing events. I was wrong.
I try hard to avoid confirmation bias in evaluating the evidence before reaching a conclusion - but it can be a lot harder to avoid it after publicly committing to a conclusion, which of course is the exact same "painted in a corner" problem I suspected Nate Silver of having. I spent the week after my "toast" post without electricity, internet, heat or hot water and only sporadically able to get to my office (I had written most of this essay when the lights went out on October 29), and probably did not spend enough of my remaining time going back over the subsequent polling to reconsider whether the conditions that seemed to show Obama in an untenable position were abating. (By Election Day, for example, the national poll averages were running in Obama's favor as well). I remain doubtful that Hurricane Sandy had enough impact to swing the election and particularly doubtful that it did much to affect turnout, which decided the election; but its interruption of the dynamic of the endgame probably did do what the final week state and national polls showed happening: cut into Romney's lead with independents and possibly cost him his chance to win New Hampshire, Virginia, and/or Florida. Exit polls showed that late-deciding voters broke in Obama's favor, which I had not expected to happen for an incumbent whose approval ratings and favorability with independent voters had been underwater for so much of his term.
II. Poll vs. Poll
The polls, particularly the national and swing-state polls in the third and fourth weeks of October, were telling consistent stories about the opinions of different groups of voters, but contradictory stories about the numbers in which those groups would show up to vote. I looked at internal evidence, and saw that Romney was winning independents handily and Obama was drawing nearly no crossover support from Republicans, which meant Obama needed to win entirely by having enough extra Democrats vote to overcome Romney's independent advantage. The first premise was borne out by Romney's win with independents and slight edge with crossovers, although his 5-point margin with independents was at least 3 points smaller than the margin most polls had been showing when I first made my "toast" call. Had Romney carried independents by 8 or 9, as the polls were showing at the time, and not lost them in a few key states, Obama would have needed at minimum a D+4 electorate to win nationally. As it turned out, Obama won by about 2 points with a D+6 electorate compared to 2008's D+7; Democratic turnout was down a point from 2008, but Republican turnout matched the 32% of 2008. So, my analysis correctly judged what Obama needed to do. The historic turnout conditions of 2008, which I believed were necessary for Obama to win, were effectively repeated.
There were all sorts of reasons, based in history and observable fact, to believe that those turnout conditions were unlikely to repeat themselves. Republican turnout had been unusually low in 2008 - the lowest since before Ronald Reagan started converting a lot of people to the party - following the historic financial crisis and 8 years of Bush, and had bounced back in 2010. While 2010 was a midterm election and involved a distinct electorate, its results underscored that there seemed no obvious reason to believe it would approach that nadir again. History tells us to treat with caution the assumption that elections held in the aftermath of a catastrophic event like the 2008 financial crisis are representative. The harder metrics, which I will discuss in Part III, also suggested GOP turnout was still doing well, if not as well as in the off-year elections of 2010. And the vaunted OFA turnout machine could only count for so much: certainly, no matter how sophisticated Obama's turnout operation, low GOP turnout was something he could not manufacture through operational efficiency. Obama had polled poorly in areas like job approval for most of his tenure, and campaigns that lose independent voters tend to be losing at least some of the enthusiasm of their base as well. As I will discuss below, virtually all the features of Obama's profile in the polls - mirrored in the post-election exit polls - were traditionally characteristic of losing candidates, as were the nature of the "this time it's different, really!" arguments made by his boosters. It had to be different for him to win. The customary laws of political gravity had to be defied. And they were.
As I'll get to more in Part II, with regard to turnout, what the exit polls show is that Obama really did do something that was very, very historically unusual - and on top of that, and perhaps even more importantly, Romney's GOTV operation (both the mechanics and his ability to inspire marginal Right-leaning voters to show up) turned out to be far less than projected. This time, it really was different. And the challenging question going forward is whether it will stay that way.
Read More »
III. State vs. Nation
For the third consecutive presidential election, the RealClearPolitics averages of state polls correctly called nearly all the states correctly - 49 of 50 states (all but Wisconsin) in 2004, 48 of 50 (all but Indiana and North Carolina) in 2008, 49 of 50 (all but Florida) in 2012. (The four missed calls all involved states that broke late and narrowly to the Democrats). There was a lot of hoopla about Nate Silver's 538 model, and in my poll-reading pieces I noted - as have others - some quibbles with his model, with which I still have issues. But it's not clear that the methodological quirks of the 538 model matter much anyway; they've yet to deliver a different answer than the RCP average in a presidential contest or in any significant number of statewide elections. Ditto for other state poll averages, all of which were basically on the mark. The state poll averages implicitly projected an electorate with a large Democratic turnout advantage, and that's what happened.
That does not mean "the polls" were all correct. The national poll averages at both RCP and TPM had Romney ahead for much of October and ahead at the time I wrote my original "toast" post. This race was often compared to 2004 and ended up much the same, but in 2004, John Kerry never led in the RCP average in October. There were a variety of similar precedential indicators that Obama was in worse shape in the national polls than anybody who had gotten re-elected in the past. And traditionally, we have tended to see movements in the national polls followed by movements in the swing state polls - exactly what was reflected in the RCP average in Colorado, New Hampshire, Virginia and Florida. By October 29, Rasmussen's state-by-state polls were showing Romney leading in enough states to win the Electoral College. Of course, Rasmussen was wrong.
It is somewhat Orwellian for Obama supporters to paint as "poll deniers" those of us who looked at a conflict between two sets of polls containing two divergent sets of assumptions, and conclude that one of those sets of polls was more likely to be correct because external evidence rendered those assumptions more reasonable.
There was great debate over whether it mattered at all to look at the national polls, since presidential elections are determined on a state-by-state basis. This was really two questions in one: how likely was it that there would be a popular vote/electoral college split, and was it more likely that you could predict the national popular vote from state polls, or predict the battleground state vote from national polls. It is worth noting, on the first score, that the defenders of the state poll averages were only half right. On the one hand, they spent months arguing that Ohio in particular was a firewall that would remain favorable to Obama even if the national vote went slightly for Romney - and if Romney had hung onto his leads in Colorado, New Hampshire, Virginia and Florida, Ohio would have been enough to put him over the top. I thought the theory that Ohio would vote more Democratic than the nation was unlikely: Ohio has had an exceptionally consistent record from 1860 through 2008 of being just barely more Republican than the country as a whole nearly every election. I was right about that: Obama won Ohio by less than his national margin. As The New Republic's Nate Cohn notes:
[T]he exit polls show that Obama did worse among Ohio's white voters than John Kerry.... if there was anywhere that the president should have excelled due to the auto bailout, it would have been northeast Ohio. But the president lost northeastern Ohio's two classic white middle class bellwethers: Lake County, home to the overwhelmingly white suburbs and exurbs east of Cleveland, and Stark County, home to Canton. The president also lost additional ground in traditionally Democratic stretches of eastern Ohio, where Obama performed worse than any Democrat since McGovern in a stretch of "coal country" along the Ohio River. And Obama’s problems weren't limited to eastern Ohio. The president performed poorly in southwestern Ohio, including one deeply conservative and culturally southern county where Obama's performance was the worst by a Democrat since at least 1868.
But score one as well for the firewall theory as a whole: Obama could have lost Ohio and Florida (both of which voted more Republican than the nation) and Virginia (which voted more Democratic than the nation) and still won 272 electoral votes.
In theory, you should always look only at the state polls, since that's where the election is; if the state pollsters are as good at what they do and poll as regularly as the national pollsters, there would never be a reason to look at national polling. While we all look at national polls, especially early in the race when states are being polled less regularly, I mainly looked at state poll averages in 2004 and 2008. Which gets to the second point: how likely state polls were to offer a better picture of the national environment than national polls. I agreed with Sean Trende that the potential flaw in this theory was that many of the state pollsters at issue (other than pollsters like Rasmussen or PPP that poll both nationally and in the states) were relatively new to the game or had individually sketchy track records that couldn't be automatically substituted for the record of "state polls" as a whole.
As it turned out, though, the state pollsters were closer to the mark, and should be the main focus of our attention in future races, at least in the closing months. But why were they right when the national pollsters were wrong? For the answer to that, you need to look at how the pollsters decided what the electorate would look like. I will cover that topic tomorrow in Part II.
« Close It
Posted by Baseball Crank at 11:00 AM | Politics 2012 | Poll Analysis | Comments (21) | TrackBack (0)
November 5, 2012
POLITICS: My Final Election 2012 Predictions
It's time to start making final predictions for the 2012 election. I'm also rounding up predictions from others who are out on the limb with me predicting a Romney victory. I still feel fairly confident about my bottom line: Romney will win. But until we see the actual voter turnout, it's hard to project more than educated guesswork as to the size of that win.
The Electorate and the Popular Vote
The final week of polling has been even more of a mess than usual in a season in which the polls have made less and less sense both internally (their assumptions about turnout and the conflict between the toplines and Romney's margins with independents) and externally (how the polls' views of turnout conflict with every other item of evidence outside the polling). Josh Jordan notes the particularly unstable polling of independents in the aftermath of Hurricane Sandy.
But while some of the more garish double-digit margins are gone, the latest battery of national polls shows Romney's standing vs Obama with independents most likely somewhere around a 7 point lead: CNN/ORC (+22, but with the smallest of sample sizes), Monmouth/SUSA (+16), Rasmussen (+9), WaPo/ABC (+7), NBC/WSJ (+7), Pew (+4), Gallup (-1).
What will turnout look like? First up, we have Rasmussen's October 2012 Party ID survey; I've revised my chart to look at the historical track record since 2004 of his October survey:
That's an R+6 electorate. Obama is clearly, in my view, doomed if the electorate is D+3 (the historical average for presidential elections since 1984) or less, and probably needs about D+6, maybe D+5 (the 2000 electorate) to win. We haven't had an R+ electorate for a presidential election since probably the 1920s (2004 was even); we may not have had an R+6 electorate since Reconstruction ended. If Rasmussen's survey is even half right, Democrats could be in for a very, very rough night across the board. Even as accurate as Rasmussen has been, I'm hesitant to go out all the way on that limb - but it's hard to argue with his record on this front. The survey encompasses a huge sample, on the order of 15,000 interviews.
It may be tough to measure the final electorate, because exit polls won't capture early voters, and in some states that's a lot of people. (The Denver Post cites Colorado Secretary of State figures showing more than 62% of the state's registered voters have voted already, with a turnout of R+2). My prediction for the national turnout is a conservative one: D+2, D 37/R 35/I 28. Assuming Romney wins Republicans 94-6, Obama carries Democrats 93-7, and Romney wins independents 53-47 - again, a conservative projection given the polls - that gives us Romney 50.3%, Obama 49.7%.
(If you run those same assumptions in the electorate from Rasmussen's party ID survey, you get Romney 53.7%, Obama 46.2% - and it gets wider from there if the spread among independents gets into double digits. But I'm being conservative here, as I still expect the more likely outcome to be fairly close).
The Electoral College
I start with this map, with Romney up 235-190:
Obama is still running ads in North Carolina and still contesting Florida; Florida is usually close, but I see no real likelihood that it goes for Obama again.
Then, let's take off the board the states where Romney is only going to win if it's really breaking big for him - I'm actually now including Nevada and not Pennsylvania in that category, and to be cautious, Maine's 2d Congressional District - and the states I'd almost written off in September where I now think Romney is in very solid shape (Virginia, Colorado and New Hampshire):
Romney 261, Obama 223. Iowa becomes irrelevant at this point - Romney wins one of the remaining three, or Obama wins all three, and it's game over. But I don't actually see Pennsylvania being the one to get Romney over the line if he loses Ohio and Wisconsin. Playing it safe, I end up with Romney taking just one of those four - Wisconsin - and a narrow electoral college win, 271-267:
If Romney wins, as I project, I strongly suspect that he will win at least one of the other three, maybe all three. But Wisconsin is my pick for the state that puts him over the hump.
Here's RCP's current map of the Senate races, projecting each side picks up 3 seats, netting no change to the Democrats' 53-47 advantage:
That's a very disappointing outcome from where the GOP should have been, but probably not as bad as it has looked most of the past two months. I'm going to be absurdly bullish and say R+2 Senate seats because I can't look across Pennsylvania, Ohio, Wisconsin, Virginia, Massachusetts, Indiana, and Missouri and see the Democrats going better than 5-for-7. There are different reasons in different races - Mourdock and Akin will have strong GOP turnout advantages at their backs, Scott Brown is just a tough campaigner, and Smith, Mandel, Thompson and Allen all have the swing state ground games behind them. Add in Montana and it will take a big set of Romney coattails for the Republicans to win half or more of those eight races - but 3 out of 8 hardly seems unreasonable if the presidential race is going well.
(It's also impossible to be sure how the Maine Senate race will come out - three-way races are notoriously hard to poll - but I'll nonetheless be surprised if Angus King doesn't win and caucus with the Democrats. A 50 D/49 R Senate with Paul Ryan as the VP could put King in position to be a tremendous power broker.).
Like most people, I'm not even bothering with a House prediction, other than to reiterate a point Neil Stevens has made: if the electorate was really going to look like the D+7 Democratic wave of 2008, we'd be talking about a ton of Democratic House pickups (redistricting or no) and a threat of the return of Speaker Pelosi. But at this point, even the DCCC seems to have all but thrown in the towel; Nate Silver doesn't even seem to be tracking odds for a Democratic House pickup. That suggests that down-ticket Democrats are looking at a much more realistic universe.
Around The Horn
Michael Barone has Romney 315, Obama 223. I'm always in good company agreeing with Barone.
Posted by Baseball Crank at 6:00 PM | Politics 2012 | Poll Analysis | Comments (35) | TrackBack (0)
November 2, 2012
POLITICS: No, Independents Are Not Just Discouraged Republicans
The centerpiece of my thesis (discussed here and here) that Mitt Romney will win Tuesday's election is his consistently strong showing among independent voters (ie, voters who identify as neither Republicans nor Democrats) across the majority of national and state polls, pretty much regardless of whether those are polls he's winning or losing. On that score, I believe the polls; if they're wrong about the independent vote, my analysis is essentially irrelevant. But if they're right, I believe I'll be proven right. If Romney goes on to win independents nationally by 5+ points and carry independents by more than a few points in states like Ohio and Wisconsin, he will win.
Obama can only overcome that kind of deficit among independents by decisively winning the partisan turnout battle - indeed, the polls that show him winning nationally or on a state-by-state basis do so almost uniformly by projecting a decisive advantage in Democratic turnout - but when you look for evidence outside of the polling samples themselves of that Democratic turnout advantage, you won't find it.
There are two main arguments currently circulating for why Obama will win in spite of these factors. Both are premised upon the notion that the Obama-favoring polls are correctly projecting an electorate on the order of 2008's D+7 (D 39/R 32/I 29) electorate, in which Obama's turnout advantage will outweigh his loss of independent voters. I dealt on Wednesday with the first of those: Nate Silver's polling model, which simply assumes that state-level polls are correctly projecting the turnout, on the theory that state polling averages have historically been trustworthy.
The other main argument comes from Josh Marshall. Marshall's thesis is that independents are supporting Romney because the ranks of independent voters have been swelled by "an exodus from the GOP to the right":
In other words, a lot of people left the Republican party, in identification terms. But they didn't become Democrats. And it doesn't seem (at least from the politics of the last two years) like they became more moderate of 'centrist' in ideological terms. They simply reidentified themselves as independents... I think in a lot of cases they actually re-identified because the GOP wasn't right-wing enough, call it a Tea Party exodus from the GOP.
This is a plausible-sounding theory, if you think independent voters are some sort of strange new phenomenon never before seen on the American electoral landscape, and Marshall backs it up with a colorful line graph showing the results of what he describes as various national party ID surveys. But it does not stand up to scrutiny.
Let's look at a screenshot of Marshall's chart; you can go click over to TPM if you want to play around with the various bells and whistles on it:
First of all, as Marshall himself admits, "we keep this data set of 'adults' rather than registered or likely voters. That makes it somewhat different from the voting electorate. ...These are polls which simply ask people over 18, how do you identify in partisan terms." As anyone who follows elections even remotely closely knows, polls of "all adults" are completely worthless, and a campaign whose supporters are citing polls of "all adults" rather than registered or likely voters six days before an election is doomed.
Second, notice something about the math here: Marshall is citing a collection of surveys that say the population is 32.5 D/25.2 R/33 I at present - which adds up to 90.7% of the people. What happened to the other 9.3%? As of the line in the middle representing the 2010 election, he shows the population as 34.2 D/30.2 R/28 I - again, 92.4% of the people, with 7.6% unaccounted for. Around Election Day 2008, it shows 39.4 D/30.6 R/25.2 I - 95.2% of the population, with 4.8% unaccounted for. It's impossible to translate those kinds of large omissions into a useful tool for analyzing the electorate. (In fact, Marshall shows independents outnumbering Democrats - and if that happens, I promise you, Obama is toast).
Third, it doesn't match up to the actual voter turnout. Exit polls in 2008 show 39% D and 31% R, numbers consistent with the chart, but 29% rather than 25% I. For 2010, it's way off: Marshall's chart has the population at 30.2% Republican and a D+4 advantage for Democrats, when in fact we know the exits showed a D 36/R 36/I 28 electorate. Somehow, the 7.6% of the people not accounted for turned out to almost all be Republicans. Marshall makes no effort to test how any of these surveys (or his rolling average of surveys) has matched up historically to the actual electorate, unlike my comparison of the track record of the Gallup and Rasmussen party ID surveys (both of which he mysteriously leaves out of his average) dating back over multiple elections. I will trust the people who have done this before and been proven reliable.
If it was true that success with independent voters was the result of defectors from the party, you would expect recent and longer-term history to show an inverse relationship between success with independents and partisan turnout - that is, you'd expect to see Republicans doing better with independents when GOP turnout is low, and Democrats doing better with independents when Democratic turnout is low. There is, in fact, some evidence that that was true before 1984, when a lot of independents and "Reagan Democrats" started self-identifying as Republicans. But since then, if you look at the presidential election years and the last two off-year Congressional elections (2006 and 2010), what you see in general is more like the opposite relationship: parties tend to do better with independents when they are turning out a lot of their own partisans. This chart shows the percentage of Democrats and Republicans in the electorate each of those years, along with each party's share of the two-party vote among independents (that is, I dropped out the percentage of independents voting for Ross Perot, John Anderson, etc.):
As you can see, Democrats did well with independents in years like 1996, 2006 and 2008 when Democratic turnout was up, indicating that good partisan environments/candidates drew Democrats to the polls and attracted independent voters. Republicans did better with independents in 2010 when Republican turnout was up. Independents were closely divided in 2000 and 2004. This is what anybody who has spent any time working in the fields of campaigns, polling or election punditry would expect to see - strength with independent voters nearly always goes hand-in-hand with the party-base enthusiasm that drives good voter turnout. A big Democratic turnout year and small Republican turnout year at the same time as a big Republican surge with independent voters is out of whack with history.
I don't doubt that, anecdotally, a fair number of people left the GOP after 2008 to join the long-time Perotista faction and build a core bloc of Tea Party-friendly independents. Rasmussen's surveys suggest that a good chunk of those people came home to the GOP by the time of the 2010 election, and if they didn't, they were replaced by other Republicans, because GOP turnout in 2010 was the best it had been since 2004. Marshall's theory that Republicans have collapsed to something resembling 25% of the electorate is frankly inexplicable in light of the 2010 elections (and the 2012 recall election, in Wisconsin).
For an example of why this makes no sense, let's look at one of the latest pieces of evidence outside the polling that supports a more Republican electorate: as Ed Morrissey notes, a recent study shows voter regstration across 8 states that register voters by party (FL, NC, CO, NV, NM, IA, PA & NH; states like OH & VA don't) shows a net 1.3% increase in Republican registration since 2008 and a net 2.5% decrease in Democratic registration, while independent registration has boomed, up 14.4%. You can read that registration data to show that being an independent is still a lot more popular choice than being a Republican these days; you can't sensibly read it to show that the growth of independent voters is the result of a decreasing base of Republican voters, and you can't possibly read it to show that the total share of Republicans and independents is holding steady or declining relative to the Democrats.
I would be very shocked if Republicans are just 25% of the voters in this election. I bet Josh Marshall would too.
Posted by Baseball Crank at 2:00 PM | Politics 2012 | Poll Analysis | Comments (12) | TrackBack (0)
October 31, 2012
POLITICS: On Polling Models, Skewed & Unskewed
There's a very large gulf between my conclusion, explained on Friday, that Obama is toast on Election Day and confident projections like Nate Silver's poll-reading model still giving the president (at last check) a 77.4% chance of victory. Let me explain why, and what that says about the difference between my approach and Nate's.
The Limits of Mathematical Models
"A page of history is worth a volume of logic"
Mathematical models are all the rage these days, but you need to start with the most basic of facts: a model is only as good as the underlying data, and that data comes in two varieties: (1) actual raw data about the current and recent past, and (2) historical evidence from which the future is projected from the raw data, on the assumption that the future will behave like the past. Consider the models under closest scrutiny right now: weather models such as hurricane models. These are the best kind of model, in the sense that the raw data is derived from intensive real-time observation and the historical data is derived from a huge number of observations and thus not dependent on a tiny and potentially unrepresentative sample.
Yet, as you watch any storm develop, you see its projected path change, sometimes dramatically. Why? Because the models are highly sensitive to changes in raw data, and because storms are dynamic systems: their path follows a certain logic, but does not track a wholly predictable trajectory. The constant adjustments made to weather models ought to give us a little more humility in dealing with models that suffer from greater flaws in raw data observations, smaller sample sizes in their bases of historical data, or that purport to explain even more complex or dynamic systems - models like climate modeling, financial market forecasts, economic and budgetary forecasting, or the behavior of voters. Yet somehow, liberals in particular seem so enamored of such models that they decry any skepticism of their projections as a "War on Objectivity," in the words of Paul Krugman. Conservatives get labeled "climate deniers" or "poll deniers" (by the likes of Tom Jensen of PPP, Markos Moulitsas, Jonathan Chait and the American Prospect) or, in the case of disagreeing with budgetary forecasts that aren't really even forecasts, "liars." But if history teaches us anything, it's that the more abuse that's directed towards skeptics, the greater the need for someone to play Socrates.
Consider an argument Michael Lewis makes in his book The Big Short: nearly everybody involved in the mortgage-backed securities market (buy-side, sell-side, ratings agencies, regulators) bought into mathematical models valuing MBS as low-risk based on models whose historical data didn't go back far enough to capture a collapse in housing prices. And it was precisely such a collapse that destroyed all the assumptions on which the models rested. But the people who saw the collapse coming weren't people who built better models; they were people who questioned the assumptions in the existing models and figured out how dependent they were on those unquestioned assumptions. Something similar is what I believe is going on today with poll averages and the polling models on which they are based. The 2008 electorate that put Barack Obama in the White House is the 2005 housing market, the Dow 36,000 of politics. And any model that directly or indirectly assumes its continuation in 2012 is - no matter how diligently applied - combining bad raw data with a flawed reading of the historical evidence.
Different sets of polls are, more or less, describing two alternate universes in terms of what the 2012 electorate will look like, one strongly favorable to Obama, one essentially decisive in favor of Romney. The pro-Obama view requires a number of things to happen that are effectively unprecedented in electoral history, but Nate Silver argues that we should trust them because state poll averages have a better track record in other elections than national polls. The pro-Romney view, by contrast, simply assumes that things have gone wrong in a number of the polls' samples that have gone wrong before. Sean Trende argues that the national pollsters currently in the field are more reliable, and that this (rather than the history of state and national pollsters in the abstract) should be significant:
Among national pollsters, you have a battle-tested group with a long track record performing national polls. Of the 14 pollsters producing national surveys in October, all but three were doing the same in 2004 (although AP used Ipsos as its pollster that year rather than GfK, and I believe a few others may have changed their data-collection companies). Of the 14 pollsters surveying Ohio in October, only four did so in 2004 (five if you count CNN/USAToday/Gallup and CNN/Opinion Research as the same poll).
In my opinion, which view is correct is not one that can be resolved by mathematical models, but rather by an examination of the competing assumptions underlying the two sets of polls and an assessment of their reasonableness in light of history and current political reality.
Where Polls Come From
Polling is "scientific," in the sense that it attempts to follow well-established mathematical concepts of random sampling, but political polls remain as much art as science, and each polling cycle presents different challenges to pollsters' ability to accurately capture public sentiment. Quick summary: dating back roughly to George Gallup's introduction of modern political polling in the 1936 election, a pollster seeks to extrapolate the voting behavior of many millions of people (130 million people voted in the 2008 presidential election) from a poll of several hundred or a few thousand people. In a poll that seeks only the opinion of the public at large, the pollster will seek to use a variety of sampling techniques to ensure that the population called actually matches the population as a whole in terms of age, gender, race, geography and other demographic factors. In some cases, where the raw data doesn't provide a random sample, the pollster may re-weight the sample to reflect a fair cross-section.
Political polling is a somewhat different animal, however: not all adults are registered voters, and not all registered voters show up to vote every time there's an election. So, a pollster has to use a variety of different methods - in particular, a "likely voter" screen designed to tease out the poll respondent's likelihood of voting - to try to figure out whether the pollster's results have sampled a group of people who correspond to the actual electorate for a given election. This is complicated by the fact that voter turnout isn't uniform: in some years and some states Republican enthusiasm is higher than others, in some Democratic enthusiasm is higher than others. You can conduct the best poll in the world in terms of accurately ascertaining the views of a population that mirrors your sample - but if your sample doesn't mirror that season's electorate, your poll will mislead its readers in the same way that the Literary Digest's unscientific poll did in 1936, or the RCP averages in the Senate elections in Colorado and Nevada in 2010, or the polls that failed to capture the GOP surge in 2002.
Technology, economics and other factors affect polling. The rise of caller ID in particular has dramatically reduced response rates - that is, pollsters have to call 8 or 9 people for every one who will answer their poll. That raises the level of difficulty in ensuring that the people who actually do answer the questions are a representative sample. Liberals argue that pollsters undersample people who have only cell phones (a disproportionately younger and/or poorer group) and non-English speakers; conservatives counter that Tea Partiers may be less likely to talk to pollsters and that polls in some cases can suffer a "Shy Tory Factor" where voters are less likely to admit to voting Republican. Partisans dispute the relative merits of in-person versus automated polling and the structure of polls that ask a lot of leading questions before asking for voter preferences. And the economics of the polling business itself is under stress, as news organizations have less money to spend on polls and pollsters do public political polling for a variety of business reasons, only some of which have anything to do with a desire to be accurate - some pollsters like PPP make most of their money off serving partisan clients, news organizations do it to drive news, universities do it for name recognition.
2012, even moreso than past elections, is apt to produce another round of reflection and recrimination on all of these issues, as a great many of the individual polls we have seen so far have been largely or wholly irreconcilable, especially in terms of their view of the partisan makeup of the 2012 electorate. If you assume that (1) the various players in national and state polling have essentially random tendencies towards inaccuracy in modeling the electorate in all conceivable environments and (2) each state's poll average includes a large enough sample of different polls by different pollsters to bear out this assumption - in that case, state polling averages and the models that rest on them should be good predictors of turnout, as they have been in most (but not all) past elections. But when you consider that 2008 was a very unusual environment and that every turnout indicator we have other than the state poll averages is pointing to a different electorate, these become far more questionable assumptions.
Toplines and Internals
Nate Silver's much-celebrated model is, like other poll averages, based simply on analyzing the toplines of public polls. This, more than any other factor, is where he and I part company.
If you read only the toplines of polls - the single number that says something like "Romney 48, Obama 47" - you would get the impression from a great many polls that this is a very tight race nationally, in which Obama has a steady lead in key swing states. In an ordinary year, the toplines of the polls eventually converge around the final result - but this year, there seems to be some stubborn splits among the poll toplines that reflect the pollsters' struggles to come to agreement on who is going to vote.
Poll toplines are simply the sum of their internals: that is, different subgroups within the sample. The one poll-watchers track most closely is the partisan breakdowns: how each candidate is doing with Republican voters, Democratic voters and independent voters, two of whom (the Rs & Ds) have relatively predictable voting patterns. Bridging the gap from those internals to the topline is the percentage of each group included in the poll, which of course derives from the likely-voter modeling and other sampling issues described above. And therein lies the controversy.
My thesis, and that of a good many conservative skeptics of the 538 model, is that these internals are telling an entirely different story than some of the toplines: that Obama is getting clobbered with independent voters, traditionally the largest variable in any election and especially in a presidential election, where both sides will usually have sophisticated, well-funded turnout operations in the field. He's on track to lose independents by double digits nationally, and the last three candidates to do that were Dukakis, Mondale and Carter in 1980. And he's not balancing that with any particular crossover advantage (i.e., drawing more crossover Republican voters than Romney is drawing crossover Democratic voters). Similar trends are apparent throughout the state-by-state polls, not in every single poll but in enough of them to show a clear trend all over the battleground states.
If you averaged Obama's standing in all the internals, you'd capture a profile of a candidate that looks an awful lot like a whole lot of people who have gone down to defeat in the past, and nearly nobody who has won. Under such circumstances, Obama can only win if the electorate features a historically decisive turnout advantage for Democrats - an advantage that none of the historically predictive turnout metrics are seeing, with the sole exception of the poll samples used by some (but not all) pollsters. Thus, Obama's position in the toplines depends entirely on whether those pollsters are correctly sampling the partisan turnout.
That's where the importance of knowing and understanding electoral history comes in. Because if your model is relying entirely on toplines that don't make any sense when you look at the internals with a knowledge of the past history of what winning campaigns look like, you need to start playing Socrates.
Moneyball and PECOTA's World
Let me use an analogy from baseball statistics, which I think is appropriate here because it's where both I and Nate Silver first learned to read statistics critically and first got an audience on the internet: in terms of their predictive power, poll toplines are like pitcher win-loss records or batter RBI.
At a very general level, the job of a baseball batter is to make runs score, and the job of a baseball pitcher is to win games, so traditionally people looked at W-L records and RBI as evidence of who was good at their jobs. And it's true that any group of pitchers with really good W-L record will, on average, be better than a group with bad ones; any group of batters with a lot of RBI will, on average, be better than a group with very few RBI. If you built a model around those numbers, you'd be right more often than you'd be wrong.
But wins and RBI are not skills; they are the byproducts of other skills (striking people out, hitting home runs, etc.) combined with opportunities: you can't drive in runners who aren't on base, and you can't win games if your team doesn't score runs. If you build your team around acquiring guys who get a lot of RBI and wins, you may end up making an awful lot of mistakes. Similarly, you can't win the votes of people who don't come to the polls.
Baseball analysis has come a long way in recent decades, because baseball is a closed system: nearly everything is recorded and quantified, so statistical analysis is less likely to founder on hidden, uncounted variables. Yet, even highly sophisticated baseball models can still make mistakes if they rest on mistaken assumptions. Baseball Prospectus.com's PECOTA player projection system - designed by Nate Silver and his colleagues at BP - is one of the best state-of-the-art systems in the business. But one of PECOTA's more recent, well-known failures presents an object lesson. In 2009, PECOTA projected rookie Orioles catcher Matt Wieters to hit .311/.395/.546 (batting/on base percentage/slugging). As regular consumers of PECOTA know, this is just a probabilistic projection of his most likely performance, and the actual projection provided a range of possible outcomes. But the projection clearly was wrong, and not just unsuccessful. While Wieters has developed into a good player, nothing in his major league performance since has justfied such optimism: Wieters hit .288/.340/.412 as a rookie, and .260/.328/.421 over his first four major league seasons. What went wrong? Wieters had batted .355/.454/.600 between AA and A ball in 2008, and systems like PECOTA are supposed to adjust those numbers downward for the difference in the level of competition between A ball, AA ball and the major leagues. But as Colin Wyers noted at the time, the problem was that the context adjustments used by PECOTA that season used an unusually generous translation, assuming that the two leagues Wieters had played in - the Eastern League and the Carolina League - were much more competitive in 2008 than they had been in previous years. By getting the baseline of the 2008 environment Wieters played in wrong, PECOTA got the projection wrong, a projection that was out of step with what other models were much more realistically projecting at the time. The sophistication of the PECOTA system was no match for two bad inputs in the historical data.
My point is not to beat up on PECOTA, which as I said is a fantastic system and much better than anything I could design. Let's consider for a further example one of PECOTA's most notable successes, one where I questioned Nate Silver at the time and was wrong; I think it also illustrates the differing approaches at work here. In 2008, PECOTA projected the Tampa Bay Rays to win 88-89 games, a projection that Nate Silver touted in a widely-read Sports Illustrated article. It was a daring projection, seeing as the Rays had lost 95 or more games three years running and never won more than 70 games in franchise history. As Silver wrote, "[i]t's in the field...that the Rays will make their biggest gains...the Rays' defense projects to be 10 runs above average this year, an 82-run improvement." I wrote at the time: "this is nuts. Last season, Tampa allowed 944 runs (5.83 per game), the highest in the majors by a margin of more than 50 runs. This season, BP is projecting them to allow 713 runs (4.40 per game), the lowest in the AL, third-lowest in the majors...and a 32% reduction from last season...it's an incredibly ambitious goal."
PECOTA was right, and if anything was too conservative. The Rays won 97 games and went to the World Series, without any improvement by their offense, almost entirely on the strength of an improved defense. I later calculated that their one-year defensive improvement was the largest since 1878. Looking at history and common sense, I was right that PECOTA was projecting an event nearly unprecedented in the history of the game, and I would raise the same objection again. But the model was right in seeing it coming.
If you looked closely, you could see why: the frontiers of statistical analysis had shifted. Michael Lewis' book Moneyball, following the 2002 Oakland A's, captured the era when statistical analysts stressed hitting and de-emphasized fielding on the theory that it was easier to use sophisticated metrics to find better hitters, but harder to quantify the benefits of defense. By 2008, the metrics were creating more opportunities to study defense, and - as captured in Jonah Keri's book The Extra 2% (about the building of that Rays team) - the Rays took advantage.
But for the Rays, the 2008 environment was not so easily repeated in subsequent years. While still a successful club with a solid defense in a pitcher's park (and still far better defensively than in 2007) they have led the league in "Defensive Efficiency Rating" only once in the past four years. It's what Bill James called the Law of Competitive Balance: unsuccessful teams adapt more quickly to imitate the successes of the successful teams, bringing both sides closer to parity. Trende, in his book The Lost Majority, applies the same essential lesson to political coalitions. Assuming that the 2008 turnout models, which depended heavily on unusually low Republican turnout, still apply to Obama's current campaign ignores the extent to which multiple factors favor a balance swinging back to the Republicans. And the polls that make up the averages - averages upon which Nate Silver's model rests - are doing just that. Nate's model might well work in an election where the relationship between the internals and the toplines was unchanged from 2008. But because that assumption is an unreasonable one, yet almost by definition not subject to question in his model, the model is delivering a conclusion at odds with current, observable political reality.
Painted Into A Corner
Poll analysis by campaign professionals often involves a large dollop of conscious partisan hackery: spinning the polls to suggest a result the campaigns know is not realistic, in the hopes of avoiding the bottom-drops-out loss of voter confidence that sets in when a campaign is visibly doomed. For the record, unlike some of my conservative colleagues, I don't think Nate is a conscious partisan hack. I have a lot of respect for his intelligence and his thoroughness as a baseball analyst and we have mutual friends in the world of baseball analysis, and I think he undoubtedly recognizes that it will not be good for his credibility to be committed to the last ditch to defending Obama as a prohibitive favorite in an election he ends up losing. (It's true that the 538 model is just probabilities, but as Prof. Jacobson notes, Nate won his reputation as an electoral forecaster with similar probabilistic projections in 2008; if you project a guy to have a 77% chance to win an election he loses, that will inevitably cause people to put less faith in your odds-laying later on).
I do, however, think that - for whatever reasons - Nate has painted himself into a corner from which there is no easy escape. If I'm right about the electorate and the polls are right about the internals, Romney wins - and if Romney wins, the 538 model will require some serious rethinking. There's a bunch of reasons why he finds himself in this position. One is that his model has been oversold: he made his poll-reading reputation based on a single election cycle, in which he had access to non-public polls to check his work. Nate is, in fact, not the first poll-reader to get 49 states right: RedState's own Gerry Daly did the same thing in 2004, missing only Wisconsin (which Bush lost by half a point) in his Election Day forecast, and Gerry did this through careful common-sense reading of the state-by-state polls checked against the national polls, not through a model that purported to do his thinking for him. (As it happened, the RCP averages at the end of the cycle did the same thing, as they did in 2008.) I'm inclined to listen to guys like Gerry who have been doing this for years and have not only recounted the numbers from past elections but lived through the reading of polls while they were happening. In 2010, the 538 model fared well - but no better than the poll averages at RCP. And that was only after Nate was much slower to pick up on the coming GOP wave than Scott Rasmussen, who called it a lot earlier in the cycle.
There are a raft of methodological quibbles with the 538 model (some larger than others), many of which reek of confirmation bias (ie, the tendency to question bad news more closely than good). For example, while Nate's commentaries have included lengthy broadsides against Rasmussen and Gallup, his model tends to give a lot of weight to partisan pollster PPP. Ted Frank noted one example that perfectly captures the value of knowing your history; the 538 model's assumptions about how late-deciding undecided voters will break are tilted towards Obama by including the 2000 election, when Gore did far better on Election Day than the late-October polls suggested. But Gore wasn't an incumbent, and there was a major event (the Bush DUI story) that had a major impact on turnout and undecided voters. If you make different assumptions based on a different reading of history, you get different conclusions. The spirit of open scientific inquiry should welcome this kind of scrutiny, even in the heat of election season.
None of this is a reason to conclude that the 538 concept is broken beyond repair. If you regard poll analysis as something like an objective calling, you can learn from your failures as well as your successes. If Obama wins, my own assumptions (and indeed, nearly everything we know about winning campaigns) will have to be re-examined. If Romney wins, the model of simply aggregating the topline state-by-state poll averages will have to be sent back to the drawing board. But there will be no hiding, in that case, from the fact of its failure.
One of the more widely-discussed efforts to fix the problem of topline poll data varying by turnout models is Dean Chambers' UnskewedPolls.com, which takes the internals of each poll and re-weights them for a more Romney-friendly turnout model. In concept, what Chambers is doing is on the right track, because it lets us separate how much of the poll toplines is due to the sentiments of different groups and how much is due to assumptions about turnout. But his execution is a methodological hash.
I haven't pulled apart all the pieces of Chambers' model, but my main objection to UnskewedPolls is that it re-weights the electorate twice:
The QStarNews poll works with the premise that the partisan makeup of the electorate 34.8 percent Republicans, 35.2 percent Democrats and 30.0 percent independent voters. Additionally, our model is based on the electorate including approximately 41.0 percent conservatives, 20.0 percent moderates and 39.0 percent liberals.
The problem with this method is that neither the raw data (the current polls) nor a lot of the historical data (past years' exit polls) has crosstabs showing how the votes of each partisan group break out by ideology. That is, for example, we have nearly no separate polling (certainly none on the polls Chambers is "unskewing") showing how Romney is polling among independents who self-identify as moderates, or how Obama is polling among Democrats who self-identify as conservatives. That's aside from the question of whether ideological self-ID is nearly as predictive a variable as party ID. Re-weighting the samples twice by these two separate variables, without access to those crosstabs, means you don't really have any idea whether you are just adding a mutiplier that double-counts your adjustments to the turnout model. It's more alchemy than science.
We can't know until Election Day who is right. I stand by my view that Obama is losing independent voters decisively, because the national and state polls both support that thesis. I stand by my view that Republican turnout will be up significantly from recent-historic lows in 2008 in the key swing states (Ohio, Wisconsin, Colorado) and nationally, because the post-2008 elections, the party registration data, the early-voting and absentee-ballot numbers, and the Rasmussen and Gallup national party-ID surveys (both of which have solid track records) all point to this conclusion. I stand by my view that no countervailing evidence outside of poll samples shows a similar surge above 2008 levels in Democratic voter turnout, as would be needed to offset Romney's advantage with independents and increased GOP voter turnout. And I stand by the view that a mechanical reading of polling averages is an inadequate basis to project an event unprecedented in American history: the re-election of a sitting president without a clear-cut victory in the national popular vote.
Perhaps, despite the paucity of evidence to the contrary, these assumptions are wrong. But if they are correct, no mathematical model can provide a convincing explanation of how Obama is going to win re-election. He remains toast.
Posted by Baseball Crank at 7:00 PM | Politics 2012 | Poll Analysis | Comments (97) | TrackBack (0)
October 26, 2012
POLITICS: Why I Think Obama Is Toast
Barack Obama is toast. This is not something I say lightly. I generally try to remain cautious about predictions, because the prediction business is a humbling one. I have never been especially bullish on Mitt Romney, and I spent most of the summer and early fall arguing that this was basically a neck-and-neck race that would go down to the wire. But in the end, two things stand out:
One, Mitt Romney has a consistent, significant lead among independent voters, which increasingly looks like a double-digit lead. This is especially clear in national polls, but can also be seen in the key swing state polls. It's been a hard enough number for the past few weeks now, even as the last of the debates gets baked into the polls, that there's little chance that Obama can turn it around in the 11 days remaining in this race. In fact, Obama has been underwater with independents almost continuously since the middle of 2009.
Two, to overcome losing independents by more than a few points, Obama needs to have a decisive advantage in Democratic turnout, roughly on the order of - or in some places exceeding - the advantage he enjoyed in 2008, when Democrats nationally had a 7-point advantage (39-32). Yet nearly every indicator we have of turnout suggests that, relative to Republicans, the Democrats are behind where they were in 2008. Surveys by the two largest professional pollsters, Rasmussen and Gallup, actually suggest that Republicans will have a turnout advantage, which has happened only once (in the 2002 midterms) in the history of exit polling and probably hasn't happened in a presidential election year since the 1920s.
Those two facts alone caused me to conclude at the end of last week that Obama will lose - perhaps lose a very close race, but lose just the same. That conclusion is only underscored by the fact that, historically, there is little reason to believe that the remaining undecided voters will break for an incumbent in tough economic times. He will lose the national popular vote, and the fact that he has remained competitive to the end in the two key swing states he needs to win (Ohio and Wisconsin) will not save him.
Read More »
Three types of people vote: Republicans, Democrats, and Independents. Traditionally, Republican candidates get in the vicinity of 90% of the votes of Republicans, and Democrats (for a variety of reasons) get a similar but perhaps slightly smaller percent of the votes of Democrats. This is more or less true over time and in national and state races. In 2010, Republicans carried Republican voters 95-4, Democrats carried Democratic voters 92-7, a 3-point Republican advantage. Absent an unusually large number of party crossovers, then, there are two paths to winning an election: win the remaining, Independent voters; or turn out more of your own.
It is usually the case that if you want to know who is winning an election, you look at who is winning independent voters. This chart, for example, shows the popular vote totals for Republican and Democratic presidential candidates from 1972-2008, juxtaposed with their share of independent voters. As you can see, while the relationship isn't a perfect one (Bush narrowly won independents while narrowly losing the popular vote in 2000, and narrowly lost independents while winning the popular vote in 2004; a lot of independents also voted for third party candidates in 1980, 1992 and 1996), independent voters tend to mirror the trend of the electorate as a whole. This is not surprising: year to year, the preferences of independent voters tend to be a good deal more volatile than the partisan composition of the R/D portion of the electorate.
The extent to which victory depends on independent voters can vary based on the margin of victory and the turnout of party regulars, of course. Exit polls in 1980 showed an electorate that was 45% Democrat and only 30% Republican; in those circumstances, Reagan's 9-point popular vote victory (and downticket victories by many GOP Senate candidates) depended on winning a lot of independents (Reagan won independents by 25 in 1980, 27 in 1984) and crossovers. Nobody since George H.W. Bush in 1988 has won independents by double digits.
More recently, we can use the crosstabs in the exit polls to break this down more precisely. In 2000, for example, Bush had an 0.5 point advantage from his 2-point win among independent voters and a 1.5 point advantage from winning crossovers by 3 (he got 11% of Democratic votes, Gore got 8% of Republicans), but Gore won the popular vote because the 4-point Democratic turnout edge gave him a 1.9 point advantage. This graph shows the component parts of the popular vote margins for Gore in 2000, Bush in 2004, and Obama in 2008:
As you can see, Republicans generally help offset Democratic turnout advantages by drawing a slightly larger number of crossover votes. In Obama's case, 28% of his popular vote margin in 2008 came from winning independents, while the other 72% came from getting more Democratic votes than McCain got Republican votes. But had Obama lost independents by the same 52-44 margin, he would have won 51-48 instead of 53-46, cutting his margin of victory by more than half - turning a 5.9 point margin among loyal partisans into a 2.6 point win.
Here's another way of visualizing how the share of the electorate broke down in each of those elections by Republicans voting Republican (R-R), Independents voting Republican (I-R), Democrats voting Republican (D-R), etc.:
The problem for Obama, as Josh Jordan has pointed out here (with regard to the national polls) and here (with regard to the Ohio polls) and the Romney campaign addressed in a memo on Ohio on Thursday, is that whatever the toplines say, Obama is losing independents and losing them by a significant amount. Jordan's analysis of the polls at the time showed Obama down, on average, 8.3 points with independents nationally and 8.7 points with independents in Ohio. If that holds (more on which below), and unless Obama can sustain the kind of significant edge in loyal partisan votes he had in 2008, he'll end up behind.
This is not a new problem, and if anything it was even worse in the mid-term elections. Independents favored House Democrats by 18 in 2006, House Republicans by 15 in 2010. Independents broke about 2-to-1 for Chris Christie and Bob McDonnell in 2009 and Scott Brown in January 2010, and independents favored Republicans in 18 Senate races (15 by double digits) and 13 governors' races (11 by double digits) in November 2010. In Ohio, independents favored Rob Portman by 39 points and John Kasich by 16; in Nevada, they backed Brian Sandoval by 28 points, Sharron Angle by 4. Turnout still mattered - Ken Buck and Dino Rossi won independents by 18 and 16 and still lost - but there is no question that the big swing of independent voters to the GOP had a huge impact on the 2010 elections, just as the huge swing to Democrats had in 2006 and 2008.
Everything in the latest polls suggests doom for Obama with independents. This morning's Washington Post poll has him down 20 with independents, 58-38. The Rasmussen national tracker has him down 17 today. Today's IBD/TIPP poll has him down 10, 48-38. SurveyUSA/Monmouth has him trailing by 19, 52-33. The outlier, SEIU/DailyKos pollster PPP, had Romney up 2 yesterday with independents, 47-45, after the PPP tracker showed him up 10, 51-41, three days earlier. In this morning's swing state poll, Rasmussen shows Romney leading Obama by 11 with independents.
In Ohio, ARG has Obama down 20 with independents, 57-37, SurveyUSA has him down 8, 47-39; TIME has him down 15, 53-38; PPP has him down 7, 49-42; CBS/Quinnipiac has him down 7, 49-42; Gravis has him down 19, 52-33.
(UPDATE: This afternoon, ARG released a more recent poll showing Romney leading by 21 with Ohio independents, 57-36. CNN/ORC countered with a poll completely out of step with all the other major polls, showing Obama leading with Ohio independents by 5, 49-44).
Obama has lost independents. He will lose them nationally by easily 5-8 points, and quite possibly well into double digits. And he will lose them in Ohio by at least 5 as well. With no sign that he's winning the crossover battle, partisan turnout is his only hope.
We've established that Obama, having won independents four years ago, is now losing them. If the electorate looks like 2008, of course, that's not a fatal problem. But that's a seriously dubious assumption.
Step back and look at the national party ID figures taken from exit polls back to 1980 (2010 exits here and here):
The Democrats' enormous, post-New Deal party ID advantage evaporated after Ronald Reagan's election, but after that, GOP party ID was remarkably steady around 35-36% from 1984 to 2000; what varied from year to year, usually going up a few points in presidential election years, was the relative vote share of Democrats vs independents. All that changed during the Bush and Obama years: Republicans enjoyed a post-9/11 boom in party ID in 2002-04, followed by a crash in 2006-08 (2008 was the first electorate below 34% Republican since 1982), followed by a run-up again in 2010. The obvious conclusion is that the largest factor in the partisan composition of the electorate in 2008 was that Republicans stayed home. Meanwhile, Democrats in 2010 were under 37% of the electorate for only the second time (the other being 1994). Another way of looking at this is to chart the Republican share of the non-Democratic electorate (R/(R+I)) and the Democratic share of the non-Republican electorate:
Looked at in that context, it's pretty clear that (1) both parties have been steadily losing share to independents since the partisan high-water marks of 2002 and 2004, and (2) by far the bigger factor in 2008 was low GOP turnout. And there's precious little Obama's campaign or ground operations can do to keep Republicans home; if anything, the harder he presses social wedge issues to fire up his own base, the more likely it is that he's helping motivate the GOP base. So he has to squeeze out a really large surge in Democrats to offset that.
That's asking a lot. If you average out the past 7 election cycles, you get an average party ID split of D+3 in presidential election years (D/R/I of 38/35/27) and D+2 in off-year elections (37/35/27). To believe that the D+7 electorate of 2008 is likely to be replayed in 2012, you have to believe some sort of fundamental shift has taken place...but the 2010 elections don't support that thesis at all. Nor, in Wisconsin, does the 2012 recall election. Take a look at the charts for Ohio and Wisconsin:
(Per Josh Jordan, I used the adjusted 2008 Ohio D/R/I figures, since the actual 2008 Ohio exits overstated the Democratic turnout advantage - if you added them up, they didn't produce the same results as the actual vote count. Even exit polls are still polls.)
Obama supporters at this point are shaking their heads, saying that those elections were different: not national elections. It's worth examining that in more detail another day. Interestingly, among other things, African-American voters were 15% of the Ohio electorate in 2010, compared to 11% in 2008. A look at the black and Hispanic components of the electorates in the six most hotly-contested swing states suggests that the narrative of a sudden shift to a less-white, more-Obama-friendly electorate is really only an accurate description of one of the six (Nevada); even Colorado saw fewer Hispanic voters in 2008 and 2010 than in 2000 and 2004, and if you're banking on non-white voters to save you in Wisconsin, Iowa and New Hampshire, you are well and truly doomed:
Race aside, there are ways of surveying the landscape to determine what the electorate is going to look like. Rasmussen has conducted a monthly survey of party affiliation since 2004, and Gallup also polls for party ID among likely voters and just this morning released a comparison of its October 2012 figures to its final polls in 2004 and 2008. If you compare the most recent Rasmussen surveys for August and September (averaged out) to how the August-September average predicted the electorates in 2004 and 2008, and compare the October Gallup polls to the final results those years, you can see that they both told similar stories that played out pretty well in November, with Rasmussen's survey being the more accurate of the two:
Extending the Rasmussen chart to the midterm elections in 2006 and 2010 you see a somewhat less accurate picture - Rasmussen understated Republican turnout in both years (in 2010, his survey only captured a big GOP spike in November) - but still one that communicated the basic outline of where the electorate was headed:
Needless to say, if the Rasmussen and Gallup surveys accurately depict the electorate, this will not even be close. Is that possible? Well, historically, big late breaks to the Republican side have tended to surprise the pollsters; as I noted, 2002 was the most-Republican year on record, and the pollsters in 2002 were especially caught unawares. Check the RCP Generic Congressional average for that year to see how badly off the final polls were and how very, very badly off they were until the final week. More here and here.
If you tick off the list of other signs of turnout - voter registration figures, polls asking which side's partisans are more enthusiastic, early voting and absentee ballot figures - you get a mess of data, data that can sometimes be misleadingly incomplete or hard to trace to nonpartisan sources...but nearly everything we can verify shows the Democrats in worse shape than 2008. It's just hard to see where you come up with the evidence of the Democratic turnout wave that Obama needs. Certainly Obama has a sophisticated GOTV operation, well-honed and extensively staffed throughout the swing states. But Republicans had that in 2006, and it was all for naught. The voters themselves still get to decide if they really want to show up and pull the lever for you. And as noted above, Obama can't do a thing to keep Republicans and independents home. At this point, given all the indicators, Obama's plan for a decisive enough turnout advantage to overcome a huge loss with independents looks like the Underpants Gnomes' business plan.
It's hard to make sense of why so many pollsters are showing this as a tight race under these circumstances, with independents consistently breaking heavily to Romney and all the indicators of turnout suggesting at least a much smaller Democratic advantage than 2008 and - if you believe Gallup's and Rasmussen's surveys - a Republican wave unlike any we've seen in a presidential election in our lifetimes. Bob Krumm notes that the GOP advantage in national polls is directly correlated to how tight their likely-voter screens are; Romney also, for whatever reason, tends to do better in polls with larger samples. But the reasons can await the inevitable mid-November recriminations over what the polls missed and why. The important point is, a D+7 electorate is gone, and it's not coming back.
The waterfront of analyzing all the factors that go into my conclusion here is too large to cover in one post, but the signs of Obama's defeat are too clear now to ignore. Given all the available information - Romney's lead among independents, the outlier nature of the 2008 turnout model, the elections held since 2008, the party ID surveys, the voter registration, early voting and absentee ballot data - I have to conclude that there is no remaining path at this late date for Obama to win the national popular vote. He is toast.
Obama's partisans have argued that he doesn't need to; that he can pursue the rare path of winning key swing states without a national win. Time permitting, I'll come back later to why I don't think this flies if you take a close look at the state-by-state polling using the same assumptions about turnout and independent voters. But I don't buy that either.
Mitt Romney will be the 45th President of the United States.
« Close It
Posted by Baseball Crank at 4:00 PM | Politics 2012 | Poll Analysis | Comments (13) | TrackBack (0)
October 9, 2012
POLITICS: Gravity Hits The Obama Campaign
This is a must-read Sean Trende column on why Obama's bandwagon strategy has demanded that he remain in the lead at every point in the campaign. I've been saying for months now that Obama's fundraising in particular - and even moreso, his ability to deter Romney from raising money from business - was hugely dependent on convincing business interests that Obama's regulators would still be calling the shots after the election and they should not feel safe about going all-in to be rid of him. This is also why Obama's team has gone nuclear in its attacks on individual polls that show cracks in his armor, moreso even than usual for political campaigns and much, much moreso than usual for campaigns that are ahead in most of the polls. The same goes for Obama's ability to draw huge turnout from young voters and other traditionally low-turnout groups.
Today's battery of good polling news for Romney (including a boost from Gallup switching from a registered-voter to likely-voter model) is far from proof that Romney will win the election, but it is a blow to the overwhelming narrative leading into the first debate that the race had already been won by Obama, and that any skepticism of polls assuming an overwhelmingly Democratic electorate was conspiracy-level crackpottery. At last check, the liberal-run TPM polling average had Romney up by 2.8 points, a wider lead than the 2.5 point lead Unskewedpolls.com was showing. The state-by-state polling may not be entirely caught up yet, but it usually lags; John McCain was clinging to swing-state polling for weeks after he fell behind for good in the big national trackers.
Republicans have been saying for weeks that this was still a close race. Today, the polls caught up to that. Obama may yet win, but he can no longer do so just by projecting inevitability, running out the clock and letting the media bury any story that threatened to help Romney under horse race coverage. Obama and Biden have been ducking tough questions - Obama does interviews on The View and music and sports radio, while dodging the White House press corps; since joining the ticket, Paul Ryan has done 197 interviews, while Biden in the same time has done 1. You can run like that when you're way ahead; you can't if you actually need to get a positive message of your own out.
There are three debates left to go (including the VP debate), which will let a national audience judge the campaigns for themselves, and despite Democratic dissatisfaction with Jim Lehrer's refusal to act as a gatekeeper running interference against Romney, it's unlikely that the moderators of the remaining debates can protect Obama and Biden from having to win those debates on their own.
Posted by Baseball Crank at 3:38 PM | Politics 2012 | Poll Analysis | Comments (1) | TrackBack (0)
August 20, 2012
POLITICS: The Registration Gap
[A] Globe analysis of voter registration data in swing states reveals scant evidence that the massive undertaking [of Democratic voter registration drives] is yielding much fresh support for Obama.
Read the whole thing for some of the dramatic results from Iowa and New Hampshire - sites of hotly contested GOP presidential primaries - in particular. The Democrats have an explanation for this, but it doesn't address one of their core problems.
Read More »
Here's what the Democrats argue:
"The fact is, there are currently many more Democrats registered in battleground states now than there were before the 2008 primary campaign began, which means there are fewer eligible voters left to register because of the gains we made in 2008," campaign spokesman Adam Fetcher said in an e-mail.
Now, as a matter of logic and basic statistics, it's true that when you pick a lot of the low-hanging fruit in one year, the ceiling on your rate of growth the next time around will be a lot lower. In some groups, like African-American adults (ie, born before 1987), this is undoubtedly a partial explanation.
But it also misses a basic dynamic regarding the Democrats' coalition. The Democrats in general, and Obama in particular, rely on winning college-age voters by large margins. According to the NY Times exit polls, in the past 4 elections, Democrats carried the 18-29 year old vote 53-34 (Clinton vs Dole), 48-46 (Gore vs Bush), 54-45 (Kerry vs Bush) and 66-32 (Obama vs McCain). In the same four elections, Democrats won first-time voters 54-34, 52-43, 53-45 and 69-30. CNN's exit polls had Obama winning voters age 18-24 (10% of the electorate) by 66-32 and 25-29 (8%) by 66-31, while McCain won a slight majority among the 64% of the voters age 40 and over. (The gap was especially pronounced among white voters, who went for Obama 54-44 in the under-30 age cohort but which McCain won by some 17 points among white voters over 30). Both sets of exit polls pegged new voters at 11% of the 2008 electorate.
Here's the thing about voters age 18-21: there is 100% turnover among the group in each presidential election. Voters age 18-21 in 2008 were born between 1987 and 1990. Many of them were college students who probably registered at either their college address or their parents' home. Most are now out of college, and may have moved away from home; in a fairly sizeable number of cases, those voters will need to update their registrations or re-register, or they are lost to the Democrats. Even if they do re-register, Democrats will need to be updating their databases to do GOTV efforts. Chasing young voters costs money that needs to be re-spent every cycle. The core Republican vote of adult homeowners with jobs and families does not present this problem.
By contrast, voters who are age 18-21 this year were born between 1991 and 1994, and - to put a fine point on it - not one of them was registered in 2008. If Democrats are not showing a large volume of voter registration, they are failing to register young voters who will be direly needed to win this election. This is aside from the fact that there has also been an unusual amount of voter dislocation since 2008 due to the economy - homes foreclosed, people moving to look for work. That can affect both parties, but it suggests some caution in Democrats relying on just maintaining their 2008 voter rolls.
This may prove particularly important because turnout will be so critical to this year's election, perhaps even moreso than in 2004. If you read the polls, you have to be struck by the wide gaps between the polls of registered voters or all adults, those that attempt to model the election to resemble the 2008 turnout, or those that assume the electorate will look something like a cross between the 2008 and 2010 electorates. I don't doubt that most of the pollsters are honestly reporting the results they are getting, but they diverge when they apply their as-much-art-as-science judgments to who is actually going to vote.
In the end, nobody knows for sure just yet. And even the voter registration picture may look a lot brighter for Democrats when college students return to campus in the fall. But the registration picture painted by the Globe is inarguably bad news for them as it stands right now.
« Close It
Posted by Baseball Crank at 1:30 PM | Politics 2012 | Poll Analysis | Comments (5) | TrackBack (0)
May 14, 2012
POLITICS: CBS/NYT: Romney 46, Obama 43 Among Registered Voters
In a long election season, it's never wise to get too high or too low over any one poll. Presidential elections are won at the state level, but statewide polling is fairly sporadic at this stage of the race, so we're stuck reading national polls a lot. But the latest poll is bad news for President Obama.
We all know the major issues by now to look for with individual polls: some polls are adults, and are totally useless, because only registered voters can vote. Polls of likely voters, in turn, are vastly more accurate and less Democratic-biased than polls of registered voters, many of whom also don't show up to vote. Most polls are also reported after weighting to achieve some guesstimate of the partisan breakdown of the general electorate among Democrats, Republicans and Independents. Even polls that don't feature egregious hackery are an inexact science, because they rest on the pollster's current assumptions about the D/R/I split and the 'screen' they use to decide who is a likely voter. If the shape of the electorate is not as projected, the poll will be wrong.
Polling averages tend to be steadier than individual polls conducted over a few hundred respondents, and they show a tight race - the RealClearPolitics average shows Obama up 46.5%-45.1%, while the left-leaning TPMPolltracker average shows Romney up 46.1-44.2. Those averages smooth out possible outliers like last Friday's jaw-dropping Rasmussen poll showing Romney up 50-43 among likely voters. And the averages themselves get more reliable as more of the pollsters start polling likely voters - right now, Rasmussen is virtually the only pollster reporting regularly conducted polls that is polling likely rather than registered voters. Looking at RCP, Rasmussen's mid-April poll is the last likely voter poll showing President Obama in the lead.
All that said, the Obama campaign cannot be happy with the results of the latest CBS/New York Times poll - a poll of registered voters done by two organizations notoriously unfriendly to Republicans* - showing Mitt Romney leading Barack Obama 46-43. Some breakdowns below the fold.
Read More »
1. This is a registered voter poll, which as noted above means it tends to favor Democrats. The weighted party-ID split is 36% D, 30% R, 34% I.
2. The trend is negative for Obama - 48-42 lead in February, 47-44 in March, tied 46-46 in April, down 43-46 in May. Whatever the methodology, if you use it consistently and show a clear trend, that says something.
3. Gender gap? Romney leads 45-42 among men, actually down from a 49-43 lead last month, but after all the "war on women" hoo-ha, Obama's 49-43 lead among women has flipped to a 46-44 Romney lead.
4. Oddly, Obama for once is polling behind his approval rating, which is up to 50% in this poll. One of the common themes of the past few years is that he tends to poll above his approval rating - people tend to like him or say they do, but don't think he's getting the job done. This, compared to the general personal and political unlikeability of Mitt Romney, is one reason why I tend to agree with Michael Barone's third scenario that there's a good chance that undecided voters wait until the last minute to resign themselves and break for Romney, much as happened in the primaries among reluctant voters who felt they had run out of better options.
The poll dropped some of the questions in last month's survey about Romney, but that poll had him tied even though voters said by 60-34 that they can't relate to him and by 62-27 that he says what people want to hear, not what he believes. In other words, it's only Romney's inherent flaws as a candidate that are even keeping this race close - people neither especially like nor trust Romney but are still dissatisfied enough with Obama to give him a shot. The longer the polls look tied, the worse things get for Obama, because it means voters haven't bought his various efforts to make Romney radioactive. Remember, all this occurs against a backdrop of voters unhappy with the direction of the country and thus predisposed to change horses.
4. Despite its near-unanimous popularity among the media, entertainment and academia, Obama's support of same-sex marriage is not an asset; by 25-16 (22-14 among independents), more voters say they are less likely rather than more likely to vote for Obama after his change of position on the issue, whereas by 23-17 (20-20 among independents), voters say they are more likely to vote for Romney as a result. Just under 60% of voters don't consider the issue a factor. Notably, the poll found that the public, by 50-46 (50-47 among independents) favors amending the U.S. Constitution to define marriage as between a man and a woman. Meanwhile, by 67-24, voters think Obama's change of position on the issue is politically expedient rather than principled. In other words, the voters think he's being political and doing something unpopular. This is not where you want to be in an election. Yet another reason why I refer to Romney-Obama as the collision between a resistible force and a movable object.
6. 62% of voters named the economy as the number one issue and another 20% named the budget, the deficit or health care. This race will be dominated by the big-picture domestic issues, not foreign policy or social issues, as much influence as those have on the baked-in partisan divides.
POSTSCRIPT: Bad polling news for Obama is also bad for his campaign for another, more immediate reason. Both of these candidates are unusually dependent on raising vast sums of money. Obama, as a number of press accounts recently have noted, has mostly lost the confidence of Wall Street fundraisers, who were a huge element of his fundraising in 2012. Romney, by contrast, as a former private equity guy, has a natural base of support throughout the financial industry. But many potential donors are terrified of donating to Romney and seeing Obama win, given this White House's well-known efforts to target and intimidate private citizens who donate to the opposition. Perceptions shifting away from an inevitable Obama victory could have a disproportionate effect on the fundraising balance of power.
« Close It
Posted by Baseball Crank at 8:00 PM | Politics 2012 | Poll Analysis | Comments (1) | TrackBack (0)
April 5, 2012
POLITICS: The Momentum Finally Shifts, Slightly, To Romney
I've previously looked in detail at the breakdown of GOP primary votes here, here and here; for purposes of this series, I've broken out the votes in three groups - the five conservative candidates (Santorum, Gingrich, Perry, Bachmann and Cain), the two moderate candidates (Romney and Hunstman) and the libertarian (Paul) - for reasons explained in the first post. In my second post, I detailed the signs to look for to see whether and when Romney would start putting the race away with the voters rather than simply plodding through the accumulation of delegates.
After the March 24 vote in Louisiana and Tuesday's votes in Wisconsin, Maryland and DC, we can see the signs of that momentum shift, but only slightly, with stubborn resistance to Romney still continuing. Not-unrelatedly, we can see the collapse of Newt Gingrich's campaign to levels even lower than he was getting in February, the last time he went a month without being on the ballot in any Southern state (recall that Newt was not on the Missouri ballot). Let's start with the month-by-month running tally:
Read More »
This looks like a significant shift to Romney after the deterioration of his numbers in March, but as with his rough showing in March, you have to bear in mind the calendar: all three primaries on Tuesday were in blue territory (deep blue in the case of Maryland and DC, plus Santorum wasn't on the ballot in DC, whereas Wisconsin has been much more favorable lately to Republicans, with Tea Party-backed Republicans Ron Johnson and Scott Walker winning statewide in 2010). The real dynamic remains mostly the same, with Romney's supporters and opponents just not evenly distributed geographically. Let's update the red/blue/purple numbers, previously broken out here, again excluding Virginia because none of the conservatives were on the ballot:
As you can see, the vote totals for April are actually slightly more favorable to the conservatives in general, and less favorable to Romney, than the overall blue-state numbers. That's not to say that Romney underperformed, per se; Santorum had been leading him in Wisconsin until the campaign and its related advertising blitz started there in earnest after Louisiana, and while Romney had won narrow primary victories in Michigan and Ohio, Santorum had previously won caucuses in Minnesota and Iowa. So, Wisconsin was far from a natural gimme for Romney, and his win there - while hardly overwhelming, with he and Huntsman still combining for less than 45% of the vote - can't be ascribed solely to being on blue turf. Sean Trende went in more demographic detail the day before the vote to explain why Wisconsin could be a sign of momentum shift. On the other hand, Romney was unable to muster a majority in deep-blue Maryland, where John McCain won almost 55% of the vote against Mike Huckabee on February 12, 2008; McCain also won almost 55% a week later in Wisconsin. Thus, the break to Romney is happening at the margins, and is not yet the stampede that McCain was enjoying by mid-February in 2008, when a smaller bloc of social conservative voters were holding out for Mike Huckabee (McCain would go on to win majorities in Texas, Ohio, Vermont and Rhode Island on March 4 before Huckabee dropped out). The sense that Romney is winning over the objections of a majority of the party's voters will not be dispelled any time soon; the best he can hope for is a determined resignation to put the primary slog out of its misery and move on to the general election.
The collapse of the Newt campaign may also be overstated somewhat here (he's been getting clobbered in the blue states all year), but there really seems to be no life left in Newt's cash-strapped operation besides protest votes (full disclosure: I cast a vote yesterday for Newt by absentee for New York's April 24 primary, purely as a protest; I might have voted for Perry if he'd been on the ballot). Amusing, neither-here-nor-there detail: with Perry and Cain not on the ballot, Michele Bachmann actually drew more votes in Wisconsin than she has in any other state, 8 more than she got in Iowa.
Finally, let's update the state-by-state tally - Virginia and DC are marked with asterisks because of the absence of major candidates from the ballot, Guam and American Samoa are excluded because I couldn't find popular vote totals:
It's nearly three weeks until the next set of primaries on April 24, all in the deep-blue Northeast (New York, Connecticut, Rhode Island and Delaware) except for blue Northeast/Midwest hybrid Pennsylvania, where Santorum - as the hometown favorite son - looks to make his last stand. Santorum trails badly in the polls in all of them besides Pennsylvania, where the polls are moving against him. The others should be be easy wins for Romney even without a momentum shift, but if Santorum loses Pennsylvania, it should convince Santorum - and Gingrich, who should have already - to throw in the towel and admit that, whether the voters like it or not, Mitt Romney is the 2012 GOP presidential nominee.
« Close It
Posted by Baseball Crank at 12:30 PM | Politics 2012 | Poll Analysis | Comments (2) | TrackBack (0)
March 21, 2012
POLITICS: Red State, Blue State, Mitt State, Newt State
How has the popular vote differed in the 2012 GOP primary if you break out the states by their track record in recent presidential elections? It turns out that there are some distinct patterns, patterns that provide both good and bad news for a GOP contemplating a general election behind Mitt Romney.
Let's start with the 13 "Red" states (i.e., the states won by the GOP in the last 3 presidential elections) to vote so far: SC, MO, AZ, WY, AK, GA, ID, ND, OK, TN, KS, AL & MS. Here's how the vote breaks down, out of 4,052,212 votes cast:
Newt 30.4% (2 wins)
If we combine the votes for the 5 conservative and two moderate candidates as explained here*, we get the following:
Conservative bloc: 60.3%
Unsurprisingly, Romney has struggled in solidly Republican states, where the conservative vote has outpolled him 2-to-1, but the division in that vote means that he, Newt and Santorum have run almost in a 3-way heat, with Newt actually narrrowly in the lead (Santorum will probably close the gap on Saturday). The good news is, unless the Romney campaign really collapses, he's likely to win most of these states against Obama anyway. The bad news is, there are a lot of down-ticket GOP officeholders who could suffer if Romney isn't able to energize voters in these states.
Then we have the 8 Blue states (states won by the Democrats the past 3 elections) to vote so far: MN, ME, MI, WA, MA, VT, HI, & IL, in which 2,460,097 votes have been cast. Unsurprisingly, these states present a diametrically opposite picture:
Romney 47.3% (7 wins)
Moderate bloc: 47.5%
Romney's run much closer to a majority with voters in blue territory, who are accustomed to making a lot of compromises in search of electable candidates; Ron Paul has also run a lot stronger in these states, while Newt has been a complete non-factor with GOP electorates that tend to be mistrustful of the role of Southerners in the party's leadership. That doesn't mean there's no market for conservatives, as the Pennsylvanian Santorum has actually done better in blue states than red ones.
Then there's the 7 Purple or Swing states, each won by each party at least once in the last 3 elections. Excluding Virginia, which skews the sample because the conservatives were not even on the ballot, that leaves IA, NH, FL, NV, CO, & OH, in which 3,345,072 votes have been cast:
Read More »
Romney 41.8% (4 wins)
Conservative bloc: 46.6%
These states - most notably Iowa, Florida, and Ohio - have seen some of the most heated campaigning of the race. The bottom line in the swing states may be the same as in the red states (Romney being outvoted by the combined conservative vote, but winning a plurality by the division among his opponents) but he has clearly outdistanced any one of his adversaries, as Newt was largely a non-factor in Iowa, showed poorly in Ohio and was not on the map in New Hampshire, while Santorum ran a distant third in Florida. And the conservative bloc's lead in these states has been much more modest, with neither commanding a majority of the vote. There's certainly a plausible argument here that - whether or not he can win over skeptical swing-state swing voters - Romney has at least shown that none of his remaining opponents is a consistent vote-getter in those states. Like most of Romney's arguments for the nomination, this is more about the weakness of his opponents than his own strengths.
Finally, there's the vote in the territories (Puerto Rico, the U.S. Virgin Islands, and the Northern Mariana Islands), in which 113,014 votes have been cast:
Romney 87.8% (2 wins)
Not much use breaking these out further; Romney's campaign steamrolled the other candidates in these jurisdictions, but none of them vote in November. In theory, combined with Romney's vastly improved showing this year among Florida's Hispanic primary voters, his overwhelming win in Puerto Rico might seem to indicate that despite Romney's current hardline immigration rhetoric** he may be on a path to do much better with Hispanic voters than I have feared. On the other hand, nothing in the general election polling would seem to support that view; neither Puerto Rican Republicans nor Cuban-American Florida Republicans seem to be all that representative of Hispanic swing voters in places like Colorado and Nevada.
* - After Saturday's Louisiana primary, the last state voting in March - or perhaps after the first April votes - I'll do a more complete update of my running tally of the popular vote in the 2012 primaries, see here and here
« Close It
Posted by Baseball Crank at 4:00 PM | Politics 2012 | Poll Analysis | Comments (4) | TrackBack (0)
March 14, 2012
POLITICS: Mitt Romney: Winning, But Not Getting More Popular (The Popular Vote, March 10-13)
After last night's contests, it's time to update my running tallies of the popular vote in the GOP presidential primary and see what further conclusions can be drawn. I continue to break out the votes in three groups - the five conservative candidates (Santorum, Gingrich, Perry, Bachmann and Cain), the two moderate candidates (Romney and Hunstman) and the libertarian (Paul) - for reasons explained in my last post. Also, the numbers through Super Tuesday have changed slightly from the last post, as more complete tallies in some states have become available. This time I'm including the Wyoming results in the totals, but not the tiny vote totals from the territories (the Northern Mariana Islands and U.S. Virgin Islands; no popular vote totals are available from Guam or American Samoa).
I. Popular Vote Totals To Date
Let's look at how the week of contests since Super Tuesday stacks up against the popular vote count up to then:
Let's take a different angle and break that out by month:
Read More »
Before opponents of Mitt Romney get too excited here, it's important to remember that three of the past week's four contests were on very unfavorable turf for Romney: the Kansas caucus came in a state where the GOP has a very strong pro-life conservative contingent (Sam Brownback is the governor, after all), and the primaries in Deep South states Alabama and Mississippi are tough sells for a Massachusetts moderate (although the Mississippi GOP is very establishment-minded; there are few red states in which the Tea Party has less influence). The next five weeks feature a battery of states, some of them quite rich in delegates and a number of them winner-take-all, where Republicans are more accustomed to nominating guys like Romney - Illinois, New York, Pennsylvania, Wisconsin, Maryland, Connecticut.
That said, especially after Romney's team made the mistake of talking up his chances in Mississippi (where he finished third), this has been a rough week for him in the popular vote, salvaged only by the continuing division among the conservative bloc. The conservatives drew at least 64% of the vote in all three states to less than 30% for the moderates, and Newt Gingrich alone ran almost even with Romney even when you include Hawaii, which Romney won. Month-to-month, Romney's share of the vote has been declining even as the field narrows, with the conservatives drawing a clear majority of the votes cast in March (aided as well by poor showings by Ron Paul in the Deep South) despite not even being on the ballot in Virginia. Neil has more.
None of this means that Romney will not be the nominee. Barack Obama lost the popular vote to Hillary Clinton after March 1 by 600,000 votes, and still won the most delegates; even if this race finally devolved into a 2-man race and Romney started losing head-to-head battles with Rick Santorum, he'd still probably take the nomination. And as of now, even if Romney can't win over a majority faction of the party, he has still outpolled any other one candidate.
There are, in my view, three groups of voters to watch to see if the primary breaks out of its current groove:
First, the Newt voters. Newt needed wins last night in Alabama and Mississippi to re-establish himself as at least a regional candidate. He didn't get them, and it's hard to see where else he'll find more favorable conditions. Like Erick, I think it's time for Newt to exit and let this be a 2-man showdown between Romney and Santorum (Paul doesn't need to get out, his voters aren't going anywhere else). Newt has enough delegates already to have a little leverage if there's a contested convention, but he has no path to the nomination in his own right, there's no way a deal made with Beltway Republicans is going to end with him in either position on the ticket, and unlike Paul he's not representing a wholly distinct faction of the party. If Newt leaves, or if his remaining supporters give up on him, I'd assume the majority will go to Santorum - but Romney would at least have the opportunity to convince some who are uncomfortable with Santorum, and that could change the popular vote dynamics.
Second are the Momentum voters, who haven't really appeared in significant numbers yet (although they do represent some part of why Romney has pushed up to the mid-30s from the 25% position he was stuck at in national polls for all of 2011). Momentum voters may not love or even like Romney, but they're voters who will get behind the frontrunner and who, in GOP primaries, traditionally step in to end the contest (except in 1976, when if anything they went increasingly to Reagan down the stretch - but then, he was Reagan. Over the 1968, 1976 and 1980 primaries combined, he drew 51.9% of the popular vote). Except so far, they haven't - I thought we were seeing momentum at last when exit polls suggested a Romney win in Mississippi, but it didn't happen.
Third are their opposite - the Bitter Enders, which is largely the class I'm in at this point. These are the voters who have come around to the view that Romney will be the nominee, but will nonetheless go to the polls to cast ballots against him to register their protest at having such a candidate head the party. Paradoxically, Romney's best bet for getting the Bitter Enders to stop voting against him is to play up the possibility that Santorum will win (forcing people to think twice about the consequences of voting for him), or at any rate force a contested convention that could handicap the ticket in the fall. Pretty much everyone who's still casting votes for Perry, Bachmann and Hunstman is in this category, and I'd wager a fair number of the Newt voters as well (I won't decide for a while yet who to actually vote for on April 24).
II. State By State
Here, I've updated the table of states (also included here are the two territories with popular votes released, the Northern Mariana Islands and U.S. Virgin Islands), ranked by the conservative bloc's share of the popular vote.
As you can see, the conservatives have now drawn a majority in 12 states (including cracking 60% of the vote in 6 states in or around the South), and a plurality in 4; the moderates have drawn a majority in 6 (including Virginia), and a plurality in 5 plus the Northern Mariana Islands; the libertarians drew a plurality only in the U.S. Virgin Islands, where the uncommitted vote was more than a third of the vote. The only state primaries where a majority voted for the moderate candidates, given a choice, have been Massachusetts and New Hampshire.
III. Primaries vs Caucuses
The four new contests don't do a lot to change the overall picture I noted before, in which Ron Paul is much stronger in caucuses, Newt Gingrich in primaries. But note that the conservative bloc has now passed 50% of the vote in primaries.
(I'll skip the turnout comparisons to 2008, as we're now far enough into the primary schedule that the comparisons are now truly apples to oranges. But I will note that Ron Paul continues to draw significantly larger numbers of voters than in 2008).
« Close It
Posted by Baseball Crank at 1:30 PM | Politics 2012 | Poll Analysis | Comments (5) | TrackBack (0)
March 7, 2012
POLITICS: Super Tuesday By The Numbers
The voting is over, and so for the most part is the counting. The delegate math, I leave to others; let's take a look at how the popular vote has shaped up over the course of this primary season and what conclusions we can draw. First, the overall popular vote before Super Tuesday, on Super Tuesday, and to date.* In addition to listing the candidates' individual vote totals, I've classified them in three groups: the five conservative candidates (Santorum, Gingrich, Perry, Bachmann and Cain), the two moderate candidates (Romney and Hunstman) and the libertarian (Paul). While there will undoubtedly be some grousing over the use of those labels, I think it's uncontroversial to note that Santorum, Gingrich, Perry, Bachmann and Cain all built their campaigns around appealing first to the conservative wing of the party and reaching out from there, while Romney and Huntsman took the opposite approach (and Paul, of course, is in his own category), so this turns out to be a reasonably useful descriptor of how the electorate has broken out between the voters responding to these different appeals. If anything, this overstates the moderate voting bloc, as Romney's "electability" argument, among other things (including religious loyalties among Mormon voters), has tended in exit polls to draw him some chunk of conservative support.
I. Popular Vote Totals To Date
There are three obvious conclusions here. One, Romney is steadily outpolling any one of his individual rivals, cementing his frontrunner status. Two, his frontrunner status derives entirely from the division among his opponents: the conservatives have consistently outpolled the moderates. And three, despite winning his home state of Massachusetts by a 60-point, 220,000 vote margin on Super Tuesday and despite none of the conservatives being on the ballot in Virginia, Romney's not getting any stronger - even with Perry and Bachmann out of the race and Cain not drawing a single recorded vote, the conservatives drew a majority of the votes on Tuesday. Thus, as Romney pulls away in the delegate race and thus advances closer to being the nominee, he does so over the sustained objections of a near-majority faction of the party. More optimistically, the strength of the conservative vote - even in a year when that vote is fractured and underfunded and the remaining conservative candidates are decidedly subpar - bodes well for conservative candidates who can unify that vote in the future.
Let's dig deeper below the fold:
Read More »
* - Excluding the Wyoming Caucus, for which the vote totals are minuscule and there's no comparable 2008 data. Late-arriving votes are still being tallied in some of the Super Tuesday states as well, but the numbers are pretty close to final everywhere. Also, vote totals (including minor candidates) were only available for some states, whereas in others I just added up the people I had data for, thus the totals don't completely match up.
II. State By State
The conservative bloc has won a majority in nine states, concentrated heavily but not entirely in the South and the caucus states (but including Ohio, where the conservatives drew 52.2% of the vote). Add to that three states where the conservative bloc formed a plurality, including Michigan. By contrast, the moderates have drawn a majority in five states - two New England states, two caucus states with large Mormon populations, and Virginia - and a plurality in five others, including two more New England states and Arizona, which also has a significant Mormon population. Only Florida, where Romney poured vast financial resources into the notorious 65-to-1 ad advantage over Newt Gingrich (the only opponent on the airwaves), did the moderates come close to a majority outside of Romney's most natural home turf.
III. Primaries vs Caucuses
I'm on record as far back as 2008 believing that caucuses should be abolished, and that the ability to win primaries is much more indicative of general election strength than winning more sparsely-attended and often unrepresentative caucuses. It's worth examining how the votes break down by type of election:
Mitt Romney, in 2008, was largely a creature of the caucuses; with his money and the unity and organization of his Mormon support, he won eight caucuses while winning only three primaries, in his home states of Massachusetts, Michigan and Utah. That's been inverted in 2012, although Romney fared much better in the Super Tuesday caucuses than he had before yesterday, undoubtedly owing in part to heavy Mormon support in Idaho and Alaska.
This time around, it's been Rick Santorum who relied heavily on the caucuses (where he could overcome his financial constraints by relying on cohesive religious conservative communities); although Santorum has broken out of that box in the Missouri, Michigan and Ohio primaries, only in Tennessee and Oklahoma has he won significantly contested primaries. Meanwhile, Newt Gingrich continues to draw nearly twice as much support in primaries as caucuses, reflecting a campaign that is not well-organized and draws on older voters unified by ideology rather than religion. But the real story of the caucuses is Ron Paul, who regularly draws twice as much support in caucus states as primary states, where the larger electorate more easily drowns out his band of committed activists.
IV. Turnout and 2008 vs 2012
Note here that Idaho switched from a primary in 2008 to a caucus in 2012, dramatically reducing turnout; not coincidentally, it's the only state so far where Ron Paul drew fewer votes than in 2008. Romney, meanwhile, has seen his vote share vs. 2008 dramatically lifted by his showing in states like Ohio and Virginia that didn't vote last time until after he had withdrawn from the race. Also note that Washington last time held both a caucus and a primary, which explains why the caucus was much more heavily attended this year. Ohio figures from 2008 are a little iffy because two sets of numbers were reported. What you can see from this chart is that, aside from states where he didn't run last time. Romney really has only grown his vote totals from four years ago in a couple of states where he went all-out spending money (or, as in South Carolina, where there was a particularly large, up-for-grabs McCain vote). The fact that Romney drew fewer votes than four years ago in ten different states, as the frontrunner against an arguably weaker field, does not inspire confidence.
Overall, turnout has not been impressive outside of South Carolina, Michigan and Ohio and a couple of the smaller electorates. Note that South Carolina was the one state where the conservative vote really came together behind a single, non-hometown candidate; like the fact that Newt's and Rick Perry's highest showings in the national polls were the highest of any of the candidates, this is indicative of the fact that there's a lot of untapped enthusiasm out there for a candidate who can unite and excite the various components of the conservative base of the party. Newt wasn't able to sustain that any more than Perry was, but there's no doubt that both pursued a path to the nomination that had a higher ceiling than that of an unexciting moderate or a religious conservative.
V. Where Do We Go From Here?
As I noted at the outset, Romney hasn't - yet, at least - shown the kind of growing share of the vote that would characterize a frontrunner who is sealing the deal. From here the race moves to Kansas (as well as Guam, the Virgin Islands and the Northern Marianas) on March 10, and Alabama, Mississippi and Hawaii (as well as American Samoa) on March 13; other than Hawaii, these are likely to be much more friendly territory for Santorum and Gingrich than Romney (Paul may do well in Kansas, which holds a caucus). Missouri holds its caucus on the 17th, Louisiana its primary on the 24th. On the other hand, we have two primaries that are likely to heavily favor Romney - Puerto Rico, where he's been endorsed by popular Governor Luis Fortuno, on March 18, and Illinois, which has a fairly liberal Republican electorate long accustomed to settling for moderates, on March 20. My guess is that we won't really begin to see momentum - Romney starting to finally pull away in the popular vote, as opposed to winning tight pluralities in competitive states while each candidate wins the states more naturally favorable to them - until we get into the winner-take-all votes in April, partcularly Wisconsin (which should naturally be competitive) and Pennsylvania (which should naturally heavily favor hometown favorite Santorum).
Is that a good thing? I think thus far, it's been important for the primary to keep going, because (1) until yesterday, there remained a path to victory for other candidates, and the voters need to know that the race won't be ended prematurely by backroom deals rather than by acts of the voters, (2) it forces Romney to work for conservative support and become a better candidate, and (3) it sends a message to the party establishment that Romney-style candidates who are long on money and short on reliable principles will not go down without an expensive and laborious fight.
Like Erick and Neil, I think Romney's narrow wins in Florida, Michigan and now Ohio have cemented him in a position where it's all but certain that he'll be the nominee, and as I said before, I think Michigan was the last hope for a new entrant in the race. For reasons (2) and (3) above, I don't see a lot of harm in continuing the campaign through the rest of March until we are 100% certain of that fact, as it gives additional states the chance to register their objections to the frontrunner and put Santorum through the final paces of seeing if he can find one last way to knock Romney off his pedestal. Newt, despite gaining some delegates yesterday and a moral victory in his home state of Georgia, no longer has much to accomplish in this race besides a protest candidacy.
There's a three-week break between the April 3 primaries in Wisconsin, Maryland and DC and the next set of primaries. My guess is that if Romney can win Wisconsin, it will be a lot harder for Santorum to justify continuing, and if he folds his tent, Newt and Paul will no longer have a race to show up for. Given how Wisconsin has been Ground Zero for many of the leading political battles of the past two years, that could be the real end of things, and if not, it will come with Pennsylvania, New York, Rhode Island, Delaware and Connecticut on April 24.
But whenever the verdict of the delegate race becomes final, the voters have already spoken, and their message is clear: this is still the conservatives' party, awaiting only the right leader to unite it.
« Close It
Posted by Baseball Crank at 3:30 PM | Politics 2012 | Poll Analysis | Comments (14) | TrackBack (0)
January 22, 2012
POLITICS: Three States Down, 47 To Go
The basic dynamics of the 2012 GOP nomination battle remain unchanged: the bulk of the GOP electorate doesn't want Mitt Romney, but isn't really sold on an alternative. Iowa's voters broke late to Rick Santorum as the conservative alternative; South Carolina's broke late, and much more decisively, to Newt Gingrich. It remains up to Newt now to prove he can hold together the conservatives going forward, as Santorum was not equipped or financed well enough to do.
It's worth noting here the raw numbers. While the categories don't perfectly describe the candidates or their supporters, it has been generally true that Romney and Jon Huntsman have appealed to the more moderate Republican primary voters; Gingrich, Santorum, Rick Perry and Michele Bachmann to the more conservative voters; and Ron Paul to the libertarian voters. What we see in the first three states is that in South Carolina, as in Iowa, the conservative vote was a majority:
Iowa: Conservatives 53%, Moderates 26%, Libertarians 21%
There will be other states - possibly including delegate-rich California and New York - that will more nearly resemble New Hampshire's profile; there will be states where Newt is not on the ballot (Virginia), where Romney has a home-field advantage (Michigan, Massachusetts, Utah), or where the confluence of caucuses and a large Mormon population favors Romney (Nevada). But at the end of the day, regardless of desperate efforts to prop up Santorum, it is hard to see any of those structural/organizational factors overcoming the core question: either Newt will unite the conservative vote, or Romney will have to earn a share of it away from him. Which has always been how we needed to pick a nominee. However you describe the GOP "Establishment," our nominee can and should only be one whom the primary voters - however reluctantly - have decided after reflection and stress-testing to nominate.
Florida won't be the last test of this, given Romney's money and organization advantages there, but it will be the first serious one. In Churchill's phrase, South Carolina was not the end, or the beginning of the end; it marked the end of the beginning.
-Efforts to prop up Santorum to help Romney, here and here
Posted by Baseball Crank at 10:35 PM | Politics 2012 | Poll Analysis | Comments (8) | TrackBack (0)
January 2, 2012
POLITICS: The Conservative Race In Iowa
There are 2,286 delegates awarded in the GOP primaries and caucuses; the nomination thus requires wrapping up 1,143 delegates. Between them, Iowa and New Hampshire award 10 delegates; South Carolina and Florida, the other two states voting later this month, award 75. By contrast, three states (California, Texas and New York) award a combined 422 delegates, more than a third of the total needed to win. So, the race is far from over after New Hampshire, and as long as there is credible opposition, it can go on for quite a while after South Carolina and Florida as well.
That said, the early states are traditionally a test of strength that helps winnow the field to the more serious contenders, especially those with the fundraising ability and appeal beyond a narrow niche to make a serious effort to win the nomination. But three of the seven candidates now in the race are pretty much guaranteed to go beyond Iowa. First, Mitt Romney: Romney would like to win Iowa, and could be embarrassed if he finishes third (lower is very unlikely), but no matter what happens, Romney's money, his appeal to the moderate wing of the party, and his establishment support will carry him to New Hampshire, where he is heavily favored to win easily. Second, Ron Paul: Paul could do well in Iowa as a protest vote if there are a lot of independents and Democrats re-registering tomorrow on caucus day, but his hard core of support and idosyncratic appeal guarantee that he will be in the race as long as there's a race, regardless of how he does in any contest, yet with no chance of ever winning. And third, Jon Huntsman: Huntsman has placed all his chips on New Hampshire and already plans on finishing a distant seventh in Iowa. The only effect Iowa has on Huntsman is indirect: if Romney looks weak coming out of Iowa, Huntsman can ratchet up his efforts to convince New Hampshire moderates that Romney is fatally flawed.
Where Iowa could matter a lot, however, is in sorting out the four candidates running as the field's conservatives: Rick Perry, Newt Gingrich, Rick Santorum and Michele Bachmann. (Let's leave aside for the moment the arguments over who can claim the term "conservative"; clearly this is the role in the field all four are pursuing). They represent a caucus-within-a-caucus, and even though they are likely to be separated 1-4 by a relatively small number of votes, their order of finish could have an outsized impact on the race, eliminating anywhere from 1-3 of them from the field.
Read More »
The reason for this is the basic dynamic of this race: after six years of running for President, the polling and anecdotal evidence is quite clear that Mitt Romney is the preferred choice of only about a quarter of the GOP electorate, and at least two-thirds would clearly prefer a more reliably conservative candidate. Paul, for a variety of reasons, can't present that alternative, and Huntsman won't. But just as clearly, Romney can win the nomination by a strategy of divide and conquer: keep the conservative wing of the party from putting its votes and money behind a single alternative. His campaign has pursued this strategy craftily, focusing fire on whoever looked likely to dominate the conservative vote at any given time, and more recently by his allies talking up Santorum, an alternative Romney clearly feels he can defeat. It's a strategy that has relieved Romney of the need to make any sort of positive argument on anything other than the faults of various conservative Republicans. But it will be his undoing if, as the race proceeds, he fails to prevent the majority faction within the party from uniting around a standard-bearer.
Of the four conservatives, Bachmann is now the most vulnerable to a poor showing in Iowa, assuming that she's running out of self-interest and not - as some have suggested - as a stalking horse for Romney. Bachmann was born in Iowa, hails from a neighboring state, won the Iowa Straw Poll, has spent a bunch of time in the state, and led the RCP polling average in the state most of the summer with support as high as 27% (higher than Romney has ever polled in Iowa; he's never cracked 23% after winning 25.2% in 2008), and her social-conservative, evangelical background and message should resonate well in Iowa. Yet, she's collapsed to single digits and sixth place in most polls, has suffered key organizational losses, she doesn't have a ton of money in the bank, and - not to be overlooked - unlike the other candidates besides Paul, she actually has to run for re-election in a not-entirely-safe district this fall. If Bachmann finishes fourth of the four, it is hard to see how she justifies staying in the race.
At the opposite end of the scale are the two guys Romney fears: Gingrich and Perry. Newt surged to national and Iowa poll leadership in December on the basis of his massive name recognition and excellent debate performances, and his role as a former Speaker of the House and leader of the 1994 conservative revolution still give him a lot of credibility and goodwill on the Right. Newt could, despite his many vulnerabilities, sustain a campaign with sufficient funding and earned media exposure to beat Romney if he could unite the Right; while polling is somewhat stale at present, he was last seen with significant leads on Romney in South Carolina and Florida, both states in which the other candidates haven't polled in single digits. Ideally, Newt wanted to beat Romney in Iowa, so he could build the argument that he was the man to stop Mitt. But Newt's support, much of it cannibalized from the early December collapse of the Herman Cain campaign, is not deeply rooted, and the not-Romney aspect of that support could desert him quickly if he's shown to be unable to either outpoll Romney or unite the Right. Newt's poll support in Iowa has dropped in the RCP average from 31 to 13 in a little over two weeks of ceaseless negative TV ads from Romney allies, he's currently polling fourth, and it's questionable if Newt has the organization on the ground to capitalize even on that much support. If he finishes far behind Romney and behind Santorum - worse yet, behind Perry as well - Newt's supporters in the southern states may start taking a harder look at alternatives. It's hard to see Newt leaving the race without being forced out, but then he's clearly bitter at Romney right now over the negative barrage; were I Perry (whose last book Newt wrote a forward for), I'd work overtime in the aftermath of a poor Iowa showing by Newt to try to convince Newt to step aside and focus the field.
Then there's Perry. Let me go out on a limb: if Perry finishes third in Iowa, he'll be the nominee. He's the guy best suited by money, organization and resume to capitalize on a strong Iowa showing, which is why Romney's media allies have been talking up Santorum's momentum instead. I don't expect Perry to finish third; he's polling fifth, and is probably most likely to finish fourth behind Romney, Paul and Santorum. Perry can afford that, if it's a respectable fourth: if Newt and Bachmann end up out of the race, Perry can make a solid argument that he's still the only credible alternative to Romney, and his style is clearly more suited to running in southern states like South Carolina and Florida. Perry's debate stumbles buried him for a while, but more than one candidate in this race has gotten a second look as the wheel continues to turn; but he needs to show that his hard work in Iowa of late has yielded some sort of progress. A fifth place showing behind both Newt and Santorum will put him on the ropes - not out just yet, perhaps, but with a much more complicated road to climbing over both to win South Carolina.
Which brings us to Santorum, the spoiler, only finally drawing attention (and scrutiny). Santorum has almost no campaign outside of Iowa, where he's spent vastly more time than anyone else in the race, doing endless, weary retail events touting his social conservatism. It's much harder to envision Santorum scaling up to a national race against Romney than it is with Newt, and just as Newt bears the scars of the GOP's failures in the 1996-2000 period, Santorum bears those of 2006, a more recent loss when he - as a member of the Senate GOP leadership - lost the Senate majority, lost his seat by 18 points to a colorless opponent, lost the support of party conservatives over his endorsement of Arlen Specter, lost the party's credibility on spending, and became a lightning rod for gay activists over his various foot-in-mouth moments on social issues. Santorum is an ex-Senator with no executive experience, and Senators are famously terrible presidential candidates, as we saw in basically every primary and/or general election since 1964 (think of McCain, Kerry, Goldwater, McGovern, Dole, Hillary, Kennedy, Bradley, Biden, Tsongas, Muskie, Edwards, Gramm, Dodd, Byrd, Gore, Brownback, Baker, Bayh, Glenn, Harkin, Hatch, Hollings, Hart, Kerrey, Lugar, Specter, Bentsen, Church, Cranston, Bob Smith, and Scoop Jackson) - the only way a Senator can win a presidential election is against another Senator, as Obama did by beating Hillary and McCain. While there may not be time to ventilate all of Santorum's problems, the greatest of which is his legacy as a loser in 2006, there is little doubt that Romney could and would destroy him once he's no longer useful in denying oxygen to more capable adversaries. But a top-3 showing in Iowa makes it impossible for Santorum to go away before South Carolina.
The clearest outcome in the conservative primary in Iowa would be for Perry or Newt to win it. The second clearest would be Santorum first and Perry second, which largely deflates Newt and takes out Bachmann. The worst plausible case is Santorum-Newt-Perry-Bachmann, which probably eliminates Bachmann but leaves Newt and Perry both wounded and regrouping for a messy South Carolina showdown.
We'll finally know more tomorrow.
« Close It
Posted by Baseball Crank at 5:19 PM | Politics 2012 | Poll Analysis | Comments (31) | TrackBack (0)
September 14, 2011
POLITICS: The Broader Meaning of NY-9
Last night's victory for Republican Bob Turner in NY's 9th Congressional District was not as resounding as the 22-point blowout in Nevada's 2d District, a district that is much more likely to play a role in contested Presidential and Senate races next year. And it shouldn't be oversold, for some of the reasons Nate Silver identifies. But Sean Trende's analysis is nonetheless a must-read regarding the broader trends it represents.
Also on a demographic note, this Atlantic article (aside from the error of forgetting - as I noted in the comments - that Rick Perry's 2006 race was not his most recent election) is a good roundup of why Perry is the GOP candidate who offers the best hope of capturing a competitive share of the Hispanic vote, as he has traditionally done in Texas.
UPDATE: Closing comments due to a spambot invasion, which tends to happen when the blog isn't updated frequently enough.
Posted by Baseball Crank at 9:00 PM | Politics 2011 | Politics 2012 | Poll Analysis | Comments (5) | TrackBack (0)
December 21, 2010
POLITICS: It's Cens-mas!
The Census Bureau today released the official reapportionment figures from the 2010 Census, which will determine (1) what states gain and lose House seats and thus will be prime targets for redistricting and (2) what states correspondingly gain and lose votes in the Electoral College for 2012.
By and large, the news was good for the GOP. For the immediate impact, I'll focus on the Electoral College, although it's worth noting how many of the redistricting states - especially the two biggest gainers, Texas (+4) and Florida (+2), and one of the two biggest losers, Ohio (-2) - are now under heavy GOP control (and the GOP just recently took control of the NY State Senate, assuring a place at the table in the other state losing more than one seat, as NY is also -2).
Read More »
Here are the other states gaining or losing one seat. Gaining a seat:
Arizona, Georgia, Nevada, South Carolina, Utah, and Washington
Thus, the GOP picks up six seats in three of the most reliably red states in the union (UT, SC & TX), five seats in three states that have been solidly if not overwhelmingly Republican (FL, GA, AZ), one in a state that has leaned narrowly Republican but went Democratic in 2008 and at the Senate level in 2010 (NV), and one state that has been reliably but narrowly Democrat (WA). Note who is missing: California failed to gain seats for the first time since 1930. Also two states where Democrats had made recent strides - North Carolina and Colorado - failed to gain seats as some early projections had shown, while Minnesota narrowly avoided losing one.
Now, the losers, besides New York and Ohio:
Illinois, Iowa, Louisiana, Massachusetts, Michigan, Missouri, New Jersey, and Pennsylvania
Again, mostly good news. Despite the recent revival of the GOP in the Northeast (see Scott Brown, Pat Toomey, Tom Corbett, Chris Christie, Paul LePage, and GOP gains of five House seats in NY alone in 2010) and Michigan and Illinois (Rick Snyder and Mark Kirk), the loss of seven seats across NY, PA, NJ, MA, MI & IL is overall good news for Republicans, especially at the presidential level. Note that NY & PA have both lost seats every census since 1930; NY is now in parity with Florida with 29 electoral votes each. IA is more of a swing state (one of just three states that flipped between 2000 and 2004, when the red-blue divide was static) and especially friendly to ethanol-industry-owned Barack Obama, MO more red and LA now reliably so at the presidential level, but on balance this is not such bad news for the GOP.
This sets the following as the states with 10 or more electoral votes:
Sean Trende pulls the regional patterns underlying the 9.7% population growth since 2000:
The official population of the U.S. as of April 1, 2010 was 308,745,538, up from 281,421,906 in 2010. The Northeast grew 3.2 percent, the Midwest grew 3.9 percent, the South grew 14.3 percent and the West grew by 13.8 percent. Overall, it was the slowest growth in the country since the 1930s.
And the Electoral College consequences:
If the 2008 election had been held under these census numbers, President Obama would have defeated John McCain 359 to 179 - essentially flipping Iowa into the Republican column before the election begins. For 2004, the numbers are starker still: Bush's 286-251 victory would become a 292-246 win, meaning that even if Kerry had won Ohio, he still would have lost (in 2004, flipping Ohio would have been sufficient to give Kerry the win).
These numbers are an opportunity for Republicans, but also a challenge, since the main areas undergoing population growth are (with a few exceptions like Utah) growing mainly through a growing Latino population. Which explains why Democrats are so eager to try to divide off Latino voters to vote as a homogenous race-conscious bloc the way African-Americans do; it's their only path to offset their shrinking deep-blue-state power base. There's no future in being the party of the Northeast and the West Coast, and Democrats know this; liberal pundits and left-wing bloggers are quite open about the extent to which they bank on racial demographics as their salvation, and those demographics only benefit them if they can maintain very high rates of racial division in the voting patterns of Latino and African-American voters, far higher than you would find among white voters. Republicans don't have to win the Latino vote outright to fend off that challenge, they only need individual Latinos to remain open enough to both sides that Republicans can persuade a decent percentage to vote GOP, as the Democrats still do among white voters. Frankly, it's not a coincidence that the two states with the most population growth have had, for the past decade and a half, GOP governors starting with George W. and Jeb Bush who worked hard to cater to Latino voters and declined to join in the harshest anti-immigration (even anti-illegal-immigration) rhetoric or policies.
Let's look deeper at the map through the eyes of a Democrat, and you can see why the campaign of relentless racial division touted on a daily basis by left-leaning commentators and pursued by President Obama in some of his ugliest moments of the 2010 stretch run - calling Republicans the "enemies" of Latinos who had to be "punished,", asserting that Republicans were "counting on ...black folks staying home" - will only be exacerbated as Obama's natural strategy for 2012, given how his performance in office has lost him much of the support among white voters (and some among Latino voters) that he enjoyed in 2008. Let's say the GOP candidate wins the following states, most of which are natural GOP states even though some went for Obama in 2008:
AL, AK, AZ, AR, FL, GA, ID, IN, KS, KY, LA, MS, MO, MT, NE, ND, OH, OK, SC, SD, TN, TX, UT, WV & WY
That's 238 electoral votes; 270 are needed to win, and conventional wisdom would tell you that a Republican who wins Ohio and Florida is sitting pretty. Meanwhile, Obama takes:
CA, CT, DE, DC, HI, IL, ME, MD, MA, NJ, NY, OR, RI, VT, WA
That's just 14 states plus DC, but carrying 186 electoral votes. The Republican is up by 52, but there's still 114 up for grabs.
Now, let's divide the remaining states in three groups.
One is the states with largely white populations but that have tended to be favorable turf for Democrats in the past: MN, WI, IA, & NH. That's 30 votes. Give those to the Republican - an awfully generous assumption, but we're looking at how Obama and his team will view the states they risk writing off by a strategy of racial division - and you've still only got 268 votes for the Republican, two shy of victory, 248 if you can't wrest Minnesota and Wisconsin from the Democrats.
Now, on the other side, look at the three swing states with large Latino populations that have represented the decisive force in the past few election cycles, as left-leaning commentators have often touted: CO, NM, & NV. That's 20 votes. If Obama can sweep all three, he's up to 206.
That leaves us with PA, MI, VA, NC: four states with large African-American voting blocs, two traditionally Democrat - but trending Republican in 2010 (PA & MI) - and two traditionally Republican - and heavily GOP in 2009-10 (VA & NC) - but won by Obama in 2008. PA & MI are still a significant vote (36 EVs), and with union-driven GOTV, Obama should be competitive; if he takes those by pushing African-American turnout over expectations, he trails 268-242, with the 28 electors of VA & NC to decide the prize. And if there's large enough African-American turnout there as well to tip those two states over to Obama, he's at 270 and re-elected, without winning Florida, Ohio or Indiana, without winning Minnesota, Iowa, New Hampshire or Wisconsin. Divide et impera.
That's all educated speculation, for now, and the 2009-2010 elections certainly suggest that it's far from a sure bet strategy for Obama. But given his history and the open thinking of people on his side, this is the map to victory I expect him to pursue, one that places virtually its entire emphasis on race as the trump card. The largest challenge for the GOP will be to play on the issues that favor Republicans - basically, everything but racial division and possibly gay rights - and avoid getting trapped in the racially polarized map that Obama is likely to consider his most plausible path to re-election.
« Close It
Posted by Baseball Crank at 12:29 PM | Politics 2010 | Politics 2012 | Poll Analysis | Comments (5) | TrackBack (0)
November 5, 2010
POLITICS: Apples and Oranges
One of the favorite sports of poll junkies after an election is to grade the pollsters, and that process is in full swing already, and should be. Neil Stevens, however, has an excellent post cautioning against putting too much stock in Nate Silver's latest effort to attach Rasmussen. The biggest specific problem he identifies is that Rasmussen offers two separate types of polls - its own polls and the POR polls, which are done at the request of paying clients under their own terms, among other things using a different margin of error - and Silver's analysis lumps the two together as if they're the same thing.
Polling involves a certain amount of art as well as science; evaluating the accuracy of polls after the fact, however, ought to be a task that can be done through a consistent and transparent methodology, for example comparing pollsters' accuracy at similar distances from Election Day. It doesn't appear that Silver's critiques are using a sufficiently objective methodology to be trustworthy guides to making sense of the pollsters.
Posted by Baseball Crank at 5:10 PM | Politics 2010 | Poll Analysis | Comments (1) | TrackBack (0)
POLITICS: Independents' Day
At ground level, Republicans win elections by doing one or more of five things:
1) Get more Republicans to vote;
There's been a lot of talk about the "enthusiasm gap" in turnout between Republican and Democrat voters, about how the Democrats registered a lot of voters to vote in the "historic" 2008 election who may not be likely to vote again, and about how the developments of the past two years have driven more people to register as Republicans. I won't attempt to evaluate those arguments here. But let us focus on one simple point, #5 on the list above: Republicans won so many elections on Tuesday because they benefitted from an enormous swing in independent voters from the Democrats to the GOP.
Read More »
The Wall Street Journal covers the dramatic overall swing from 2006 and 2008, as reflected in exit polls, concluding from an analysis of multiple exit polls that:
-Republicans won independents 55-40 across House races (+15);
-Democrats had won independents 57-39 in 2006 (+18);
-Obama had won independents 52-44 in 2008 (+8).
The results, of course, were dramatic. While not every race is settled, last I checked, Republicans seem to have gone 239-187 (56%) in House races (with 9 races still contested), and - assuming no further counting or recounting disturbs the current vote counts - 24-13 (65%) in Senate races, and 23-13-1 (62%) in Governor's races (at least Connecticut and Minnesota remain contested, although Minnesota looks like it's fairly likely to go Democratic and the Democrat currently leads in Connecticut), on top of making colossal gains in lower-profile state legislative races, gaining some 680 seats, winning control of 18 new chambers and at long last returning to the levels of Republican influence in state legislatures that were last held before Herbert Hoover.
Now admittedly, exit polls are an inexact art at best, but they're the best we have, and unlike pre-election polls - or the exits that get leaked before the results are out - exit polls released after the votes are counted at least are constrained to reach the same results as the final polling. To see how the exits played out in specific races, I decided to drill down on a race-by-race basis, using CNN's exit polls for Senate and Governor races (for comparison to the WSJ's multi-poll average, CNN showed Republicans winning independents in House races by a 56-38 margin (+18), so CNN's numbers may be slightly more favorable than the overall average). CNN didn't poll every contested race; it skipped the Alaska Senate race and less competitive open-seat Senate races in Indiana, North Dakota and Kansas, as well as interesting and hotly contested Governor's races in New Mexico, New Hampshire, Minnesota, and Rhode Island. That said, CNN polled 25 of the 37 Senate races and 19 of the 37 Governor's races, and the results are interesting enough for what they are.
Let's take a look first at the Senate races, with the Republican candidates ranked by how well they showed with independents, as well as the share that independents made up of each state's electorate (some states have a lot more independents than others, usually depending on whether one or other party has a ton of registered voters in the state):
(If you're wondering, Charlie Crist won only 35% of independents, a terrible showing for a guy whose entire rationale as a candidate was his supposed appeal to independents).
There are a number of interesting takeaways from this. First of all, other than totally hapless candidates like the opponents of Patrick Leahy and Daniel Inouye, virtually every Republican Senate candidate was competitive among independent voters this year, and Republicans won independents in 19 of the 25 races, including five candidates who lost anyway. The top of the chart includes some people in uncompetitive races (Grassley), some people who have always appealed to independents (McCain), and some people who ran against especially disreputable opponents (DeMint and Kirk - the next time some liberal gives you a hard time about the people who supported Christine O'Donnell for Senate, remind them that 71% of South Carolina Democrats voted for Alvin Greene for Senate. Alvin Greene.) Poor Blanche Lincoln got slaughtered, losing independents by 43 points as a two-term incumbent. But perhaps the most striking candidate to win over 60% of the independent vote is Rob Portman, the former Bush budget director running in perennial swing-state Ohio; while Portman was a good candidate running against a hapless opponent, it's a mark of the poisonous climate for Democrats that independents broke 66-27 for an establishment Republican. Also stunning is Kelly Ayotte winning New Hampshire's sizeable independent bloc by 26 points, running for an open seat in a New England state that voted for Kerry and Obama.
The conventional wisdom is that Tea Party Republicans couldn't reach independent voters, notwithstanding of course the fact that much of the Tea Party movement's success is in reaching conservative but unaffiliated voters who don't necessarily self-identify as Republicans or even vote every year. The CNN exits suggest that, while some Tea Party-backed candidates ran better than others, virtually all of them drew more independent support than their opponents. Rand Paul won independents by 16 points, Ken Buck by 16, Marco Rubio by 16, Ron Johnson by 13. Even weaker GOP candidates like Sharron Angle, Christine O'Donnell, John Raese, Carly Fiorina, Jim Huffman (Ron Wyden's opponent in Oregon) and Linda McMahon won a bigger share of independents than did liberal veterans like Harry Reid, Barbara Boxer, Russ Feingold and Patty Murray. While Angle and O'Donnell didn't do well enough among independents to offset partisan advantages in their states - and Angle clearly ran far weaker among independents than Nevada GOP Gubernatorial candidate Brian Sandoval - the numbers belie the idea that they were toxic to independent voters. (Interesting side note: Angle won college educated voters, but lost non-college-educated voters by 10 points). As for the Democrats, the only Democratic candidate in a significantly contested race to win a majority of independents was Joe Manchin, who turned around his campaign with the famous ad in which he literally put a bullet through Obama's legislative agenda. That speaks volumes about the mood of independent voters this fall.
The other take-home lesson is that a better get out the vote (GOTV) operation might have made all the difference for Ken Buck and Dino Rossi, both of whom at this writing seem to have lost nailbiters. Rossi won Washington's huge independent faction by a blowout 18-point margin and won 96% of Republicans, but appears to have lost due to the Democrats' large registration and turnout advantage. Buck similarly won independents by a comfortable 16-point margin, although unlike Rossi he took some losses in his own caucus (winning Republicans 89-10). The common denominator is that they lacked organizational support. The RNC, crippled by Michael Steele's feckless financial mismanagement, was unable to put its formerly formidable GOTV operation in the field this year at all; the RGA appears to have stepped into the breach in many states that had contested governor's races (it spent a reported $102 million under Haley Barbour's direction), but Washington had no governor's race (Rossi lost the last two of those, too, also narrowly) and Colorado's GOP slate imploded despite a receptive electorate that swung the state's General Assembly to GOP control. Money spent by the various national organs on other races might have swung those two races.
Now, let's turn to the governors:
(Tom Tancredo, as the de facto Republican candidate after Dan Maes' collapse, won 48% of independents, running ahead of both Maes and Democratic winner John Hickenlooper).
Given that governors' races are less tied to Washington partisanship and ideology, we see a slightly less uniform trend here and fewer blowouts in favor of the GOP candidate, but again, Republicans across the board ran exceptionally well with independents. Note that Bill Brady, who narrowly lost the Illinois Governor's race to Rod Blagojevich's running mate, ran as well with independents as almost any GOP gubernatorial candidate in the country; he just ran ever so slightly behind Mark Kirk, and the Democrats' registration advantage in Illinois is too much to overcome without a huge win among independents. Brady, Tom Foley in Connecticut and Chris Dudley in Oregon stand as the three Republicans who won a majority of independents and (assuming the "counting" process goes poorly for Foley) still lost. (Dudley, by the way, had the most staggering gender gap in exit polls I think I've ever seen - he won men by a landslide 60-36 margin and lost women by an even more decisive 62-36).
Four take-home lessons should be obvious:
1) Republicans and Democrats alike need to persuade independents in order to win any significant number of elections, and the mood of independent voters will be just as important in 2012 as it was in 2006, 2008 and 2010;
2) GOTV and expanding the base matter too, and it's still possible on occasion for Republicans, at least, to lose races where they win independents decisively;
3) The broader trend among independents will be driven by the national mood and show up across the board wherever there's a competitive race, but individual candidates matter, as weak ones will fail to draw enough strength from the national tilt to win unless they're running in very safe states;
4) Good conservative candidates can and do win over independents just as well as good moderate candidates. Bold-colors conservatives like Rubio and Pat Toomey, establishment conservatives like Portman and John Kasich and moderates like Ayotte all did well with swing-state independents. Poor candidates fell closer to breaking even. Conservatives shouldn't stop running good conservative candidates simply out of fear that they can't win - but we should still prefer a strong moderate to a fatally defective conservative, if our goal is to win elections.
« Close It
Posted by Baseball Crank at 12:40 PM | Politics 2010 | Poll Analysis | Comments (8) | TrackBack (0)
September 13, 2010
POLITICS: What's At Stake In Delaware
Tomorrow, the voters in seven states and the District of Columbia go to the polls to conclude the primary election season. The most closely-watched race on the ballot will be the race between long-time at-large liberal Republican Congressman Mike Castle and Tea Party-backed conservative insurgent Christine O'Donnell for the Republican nomination for the open Senate seat previously vacated by Joe Biden. Because the election is a special election, the winner of the seat will be seated immediately (as with the Illinois Senate race) and serve until the general election in 2014.
There have been a long series of contested primaries in the GOP this year, albeit not all of which ended up getting resolved at the ballot box. Just in the Senate races we've had victories of one sort or another for the conservative insurgents in Florida, Pennsylvania, Utah, Alaska, Colorado, Nevada, and Kentucky, victories for the moderate establishment candidate in Indiana, Arizona, California, Washington, and (likely on Tuesday) New Hampshire, victories for conservative establishment figures in Ohio and Missouri, and less clear-cut ideological battles in Connecticut, Kansas, Arkansas, Louisiana and Wisconsin (New York's another story entirely). In only one state (Illinois) has a liberal Republican won a race against any sort of opposition, and prominent and experienced liberal Republicans have gotten crushed in Connecticut (Rob Simmons) and California (Tom Campbell). Notably, a few of the moderates who did fend off a conservative challenger (e.g., John McCain and Carly Fiorina) did so with the help of an endorsement by Sarah Palin, the lightning rod of this primary election season.
The Castle-O'Donnell race has become perhaps the most divisive primary of this cycle within the conservative movement, for reasons I'll explain in a moment. There are a couple of important questions at stake that are worth considering, which really go to the heart of what kind of party the GOP should be; but it's equally important to recall that we are compelled to face those questions only because of the particular weaknesses of these two candidates and the conditions in Delaware. The result is that there are good arguments on both sides of this one. As I'll explain, I come out on the side of backing Castle, but the case for backing O'Donnell can't be dismissed out of hand and deserves serious reflection.
Read More »
Let's start by remembering a crucial principle that I have reiterated at every possible opportunity, and that is almost invariably the first casualty of commentary on primary fights: the individual circumstances of individual races matter. Ideas don't run for office, people do. The dynamics of different races in different years in different jurisdictions with different electorates are different. Arguments that win in one race may be foolish in another. We all need to put down the rhetorical blunderbusses on this race and look at the facts in making our judgments.
Why Conservatives Should Support Castle
At first glance, the argument for Castle is a microcosm of the argument for moderate or liberal Republicans in a lot of races: he can win and O'Donnell can't, or at any rate he's a fairly sure winner and O'Donnell, in a general election, would be a very long shot. This is not, however, simply a repetition of the idea that moderates are always more electable, a view that is often flat wrong (the people who want us to run moderates are often the same ones who criticize the party for not having ideas...but of course, not having ideas is generally the defining characteristic of moderates, see the John McCain and Bob Dole presidential campaigns). It's not even just the assumption that conservatives can't ever win in Delaware, although it would be nice for people commenting on this race to understand that Delaware really is a different animal from Colorado or Nevada or Utah or Alaska. As Martin Knight noted:
The race to replace Mike Castle in the House has the Democrat leading either of the two likely Republicans by approximately 16 points. This is the exact same electorate that will be faced with a choice between Coons and O'Donnell in the Senate race - the same electorate that is already showing that they prefer Coons to O'Donnell by 11 points and prefer Castle to Coons by a similar 11 points.
Castle, of course, has won many statewide races in Delaware, and is the easy favorite if he wins the nomination. The last guy more conservative than Castle to win statewide in Delaware was Bill Roth, and Bill Roth was nobody's idea of a conservative; the last actual conservative elected statewide was Pete du Pont, and that's more than two decades ago, plus electing a du Pont in Delaware is pretty much the opposite of populism. Even a passing familiarity with the state's economy and political culture should tell you why a populist candidate like O'Donnell is a tougher sell in Delaware than even in, say, Massachusetts or Connecticut.
And we're not talking about a generic conservative candidate like O'Donnell, but O'Donnell herself. O'Donnell's not a veteran state legislator like Scott Brown or Sharron Angle, or a former federal judge like Joe Miller, but a repeat candidate with no real experience at any level of government. I'm all for citizen legislators, but my point is that she really has nothing to fall back on to convince skeptical voters that she's a serious person. And O'Donnell has had a series of problems the past few weeks, from a trainwreck radio interview where she was caught fibbing about having won some counties against Joe Biden in 2008, to a disastrous effort on her behalf to claim that Castle was gay, to a series of hit pieces on her financial and other woes (Robert Costa at NRO sums up O'Donnell's troubles, and see also John McCormack at the Weekly Standard). Undoubtedly some of this is dirty pool - the Castle camp's last-minute FEC complaint smells like this - but we're 51 days from the general election, and it's not helpful to have a candidate who is taking on water as rapidly as O'Donnell, and far too late in the day to find a replacement candidate.
The picture, looking at the polls, is predictably grim for O'Donnell; as Neil Stevens has catalogued here and here, polling shows fairly convincingly that O'Donnell isn't even as popular with Republicans in the general election as Castle, even before you get to independents and Democrats. The PPP poll showing O'Donnell pulling to a within-the-margin-of-error 47-44 lead among likely primary voters may establish that she's electable in a primary that doesn't even draw all Republican voters to the polls, but it certainly doesn't suggest that she's a strong candidate:
GOP voters are pretty sharply divided about O'Donnell as well. 45% have a favorable opinion of her with 41% seeing her unfavorably. Only 50% of primary voters think she's fit to hold public office but she does much better than Castle on the ideology front- 53% think she’s about right.
If you believe in supporting the most conservative viable candidate, there's still a rather persuasive argument that you have to back Castle because O'Donnell's just not viable, and Castle is - if nothing else - more conservative than Chris Coons. Even some arguments for O'Donnell, like Erick's view that O'Donnell shouldn't be a priority to Jeffrey Lord's argument that losing elections today helps build to victories tomorrow, assume that O'Donnell is basically a sacrificial lamb. Democratic observers seem inclined to give Delaware a pass if Castle wins, but possibly jump into the race if facing O'Donnell.
Castle recently matched her promise to stop the lame-duck agenda. I asked his staff for a statement from Castle on the lame-duck session and they provided this very strong statement from the congressman: "The only business that should be conducted during a lame-duck session of Congress is keeping the government running until the newly elected legislators are sworn in. I do not agree with those who say this period of time should be used for passing controversial legislation and would not play a role in helping to circumvent the will of American voters."
I accept that people experienced in dealing with Republicans of Castle's ilk may not place a lot of value on that promise, but in weighing the likelihoods here, I expect that it's probable that Castle at least would not break a promise that unambiguous that quickly.
As for what Castle would do in office beyond a lame-duck session, I have no illusions that he'd be in the GOP foxhole more than half the time at best, and far less than that on fights where conservatives are taking on the GOP establishment, as opposed to siding with the powerful corporate interests that hold so much sway in Delaware. Geraghty's piece is worth reading for the details of how Castle got a career ACU rating of 52 and a rating of 56 in 2009 (up from 28 in 2008, natch), even if you gag on the rather risible suggestion that Castle is any but the most once-in-a-blue-moon accidental kind of friend to pro-lifers. Either way, however, Castle's status as a fair-weather friend to the GOP and only very rare fellow traveler with conservatives is better than the Democratic candidate, Chris Coons, a man who once wrote a college newspaper article entitled "Chris Coons: The Making of a Bearded Marxist" about his political conversion after a trip to Kenya. Whatever your view of the 2006 Rhode Island Senate primary race between Lincoln Chaffee and Steve Laffey, the outcome from a Coons victory would be similar to the one in that race, ensconsing a liberal Democrat who could well be in office for decades. If you're serious about beating liberal Democrats and driving their terrible ideas from the public square, wanting to keep Chris Coons out of the Senate ought to be a thing worth doing.
The real rub is the 51st vote to control the chamber and pass a budget. It's unlikely that the GOP will get close to needing a 41st vote to hold filibusters, or having a 60th vote to break them without bipartisan support, between this November and the 2014 election, when Castle's term would be up. Moreover, if (as many of us now expect) the GOP gains control of the House this fall, there will be little threat in 2011-12 that the Senate will be voting on a whole lot of liberal legislation, and thus few "bipartisan compromises" on which a Castle defection might be damaging. But with GOP gains in the Senate looking like a good shot at +7 or +8 and a non-trivial chance at the +10 needed to swing the chamber, the decision on whether to throw away the Delaware seat could be a consequential one. And I trust that, at least between now and 2012 (when the GOP could conceivably take the White House, but could also get more reinforcements in the Senate), Castle won't have any reason to bolt the party Jim Jeffords-style, not having done so yet (not that I'd put it past him once Republicans are governing again).
There's also a benefit to demoralizing and defunding the opposition. The Democrats, lacking control of the Senate, will lose the jobs of a lot of staffers; they'll lose access to corporate and lobbying money that (as we all know in the real world) flows to the party in power. They'll lose the ability to recruit candidates on the promise of joining the majority (recall that in both 2012 and 2014 the Democrats will be defending twice as many Senate seats as the GOP, so unless we see a repeat of the 2006 and 2008 Democratic waves, it's not likely they'll gain seats and highly likely they'll lose more). The GOP will gain these things. If you think that shifting more power and advantage to the GOP over the Democrats is a good thing, you would presumably support Mike Castle.
Now, if you assume GOP control of the House and Democratic control of the White House, control of the Senate is not quite as urgent in other ways; the GOP has a firm foothold in the legislative process and a platform from which to launch oversight investigations, etc. But not every issue goes through the House; the Senate has unique jurisdiction over matters like treaties and, of crucial importance, judicial nominees, and the agenda-setting power of a majority in those areas can be profound. We know from experience, as well, that even liberal Republicans like Olympia Snowe, Susan Collins and (once upon a time) Arlen Specter can be counted on to take the conservative side in some fights, witness the confirmations of Chief Justice Roberts and Justice Alito (by contrast, liberal Republicans will never be useful in a fight against any but the worst judicial nominees from the Democratic side; that's life).
All of this assumes that conservatives, having no realistic alternative but to work within the Republican party, despite not having undisputed control of the party, have a stake in the party's electoral success, and that gaining control of the Senate by means of Mike Castle would not hamper either the party's success or the movement's power within it. Which leads us to...
Why Some Conservatives Have Good Reasons For Supporting O'Donnell
The fact that there are good and practical arguments - with which I largely, if reluctantly, agree - for supporting Castle does not mean the arguments to support O'Donnell are frivolous or unworthy of serious consideration.
I'd rather see the Democrat get elected than see Mike Castle get elected. Seriously, I know many of you disagree with me, but if the majority depends on Mike Castle, to hell with the majority.
I would rather 50 seats without Mike Castle than 51 seats with Mike Castle. The push to support Mike Castle by "conservative" groups, pundits, and others says more about the selling out of the conservative movement to the GOP than anything else. It happened in the Bush years and many conservatives were so thoroughly co-opted by the GOP Establishment they might as well be cut off from the conservative movement permanently.
Here's the point of departure: there is a serious argument, and one I have particularly credited in races like the Pennsylvania and Florida Senate races, that even if they're electable, Republicans can't afford to elect big-government moderates and liberals, because both as a matter of philosophical conservatism and, ultimately, long-term political self-interest, the GOP can't afford to be in a position (as it was by 2008) where voters feel they simply can't tell the difference between the two parties in terms of spending, regulation and the size of government. The argument goes that guys like Castle will badly dilute the GOP brand every time they either (1) give "bipartisan" cover to some Democratic nonsense or, worse yet, (2) water down the GOP's own agenda as a condition for getting it passed. This is an especially big problem when you're dealing with the last vote in a majority. And as a practical matter, because Castle knows that the likely outcome of a primary challenge to him (assuming he intends to serve more than a single term, which is debatable) is a Democratic victory, he'll actually be less persuadable by grassroots pressure from the Right than a guy like Ben Nelson, who will vote for Democratic control of the chamber but needs to win Republican-leaning independents in Nebraska to keep his job. Hogan adds detail to the behind-the-scenes dynamic of adding a guy to the caucus who is neither conservative nor trustworthy:
Liberal and establishment Republicans have corrupted the entire Republican Conference in the Senate. Other than a very few stalwarts, most Republican Senators (whether they are supposedly "moderate" or "conservative") simply do not IN ANY WAY stand for the principle of limited government - they like to spend money, they want to be seen as "reasonable" more than principled, and they get together in their little Senate club and drench themselves in the kool-aid of "getting things done" for the American people (which is code for sticking it to the people). That remains true today. They look at the Tea Party movement and what is happening in primaries around the country with dismay and disdain. They want it to go away.
The related argument is, basically, that Mitch McConnell and his leadership team simply aren't ready yet to absorb the full lesson of the 2006 and 2008 elections and the Tea Party movement; that while John Boehner may be ready to take a harder line on spending and size of government in the House, giving McConnell the majority back - with all its perquisites - would reward him before he's learned any lesson.
These are serious points, but I think ultimately they misread the times and the extent to which different strategies are needed in different situations. Republicans in 2002-06 grew a majority, it is true, that depended on way too many people who were not small-government conservatives. But we need to recall three things. One, the war was such a dominant issue in those years that the grassroots of the party simply wasn't looking after these issues (and that includes bloggers like me, I admit). Two, and relatedly, we had leadership from the White House that de-prioritized and at times affirmatively trashed small government. Bush wasn't nearly as bad as Obama, and the flaws in his record on spending and regulation are sometimes exaggerated, but there's no serious argument that he was a small-government guy, and we knew that even before he began running in 1999. And three, we had a lot of misalignment, i.e., non-conservatives elected in jurisdictions where a conservative should have a fighting chance. We're at work fixing that problem in places far more hospitable to conservatives than Delaware.
When we turn to 2009-10, Republicans have needed, as an outnumbered and beleaguered minority facing an unprecedented effort to permanently expand the federal government, to insist on a greater level than before of party discipline and ideological purity so as to get the message out to a frustrated public that both sides were not equally guilty of the things that have so inflamed public sentiment about an out of control government. My own guess is that a guy like Castle doubled his ACU rating from 2008 to 2009 not because he was thinking about an election before the same voters who have been electing him for years, but because he saw the ground shift to the point where even he had to vote against some really alarming things, like the Obamacare and stimulus bills.
But these things will not always be the case. The goal of anyone who cares about conservative ideas becoming public policy should be to build a working majority. A working majority needs a core, and we've been building that in the many states where a conservative can get elected. But it can also make use of marginal members who don't vote with the core all that much, but who win elections that would otherwise be won by determined opponents. In the long run, I'd rather have a Senate with 45 or 51 conservatives and 10 moderate/liberal Republicans than a Senate in which we can only win those elections a conservative can win, a Senate in which we give ourselves no margin for error in having the votes to make policy (consider how few conservatives you can think of who never dissent from the pure conservative position on any issue). That may be a frustratingly incomplete result, but frustratingly incomplete results are what the Senate was designed for. A movement serious about governing must try to win the marginal districts like Delaware, even if that means electing marginal candidates.
As for the concern about McConnell, I fear that some of the critics are letting their focus on McConnell distract them from what's really going on. The GOP will return just 32 incumbents (including Scott Brown) in 2011, so if a majority is possible it will consist of an unprecedented 19 freshman Senators, many of them ready and able to join the "blow stuff up" caucus of Senators like Jim DeMint and Tom Coburn who are serious about fighting tooth and nail against government excess. You don't need that many of those people in the Senate to make a lot of havoc. Also, McConnell's close associates among the moderates in leadership have been systematically eliminated - you can bet that we're going to have a leadership team that has more voices at the table for conservatism in 2011, given the number of new Senators who will have overcome determined opposition by the NRSC and the leadership to win election, and those people will have a very powerful argument that if they're ignored, yet more heads will roll in future primaries in states more conservative than Delaware. Those new conservative Senators will also bring to town a lot more new conservative staffers. Majority leaders by nature have no choice but to ride the tiger of their own caucus, and while a caucus with Castle, Kirk, Snowe and Collins will pull McConnell to the left, he'll have an awful lot more people pulling him to the right no matter what his own inclinations.
Then there's the strategic argument, which is basically that leaving the Democrats with an impotent majority in the Senate is better than having a Congress fully under GOP control, because Obama will be able to repeat Bill Clinton's record of demonizing the GOP's obstructionism. I hear this, but I don't buy it. One, the public will already know that Obama is engaging in knock-down-drag-out budget fights with John Boehner and the House GOP; Obama's preparing the ground for that already. Two, the public isn't always that attentive; voters seemed barely to notice in 2008 that the Democrats were already running both Houses of Congress. Three, Obama's not Bill Clinton. Clinton had failed to do half as much damage to small government in his first two years in office, and had a lot more skill and experience in presenting himself as a man capable of learning lessons about the size of government. Clinton could plausibly portray himself as a reasonable, compromising fellow spurned by extremist Republicans because he was actually, if grudgingly, signing on to some significant conservative policy priorities, notably welfare reform. Obama will do nothing of the sort. And as I said before, in terms of short-term electoral strategy, I think the GOP is better served institutionally by defeating the Democrats, demoralizing them and stripping their power base than by banking on winning the rest of the majority later.
Erick has suggested another concern whispered of in Delaware circles, that "[i]f Mike Castle becomes the next United States Senator from Delaware he is going to get sworn in, serve a bit, then become a Democrat, resign, and let Beau Biden get an appointment." This may sound less fanciful if you've noticed three things: one, that when Castle finished his term as Governor of Delaware he basically cut a deal where he'd run for Congress and the state's Democratic Congressman, Tom Carper, would run for Castle's job; two, that Castle never tried to run against Joe Biden; and three, that Castle's allies in the Delaware GOP haven't run a candidate against Beau Biden for the State Attorney General job.
All that said, it seems to me much more likely, given how Castle and Delaware politics work, that what would actually happen is that the 71 year old Castle serves out his four years and then steps aside, at the point at which Biden would be deemed ready to run after four years as AG. If the idea is that Castle is going to all this trouble to keep the seat warm for the Bidens as part of a clubby backroom deal, it doesn't make much sense; Biden could have been the nominee this year if he'd wanted to, or the incumbent Ted Kaufman - a Biden crony - could have run again.
Then there's Jeff Lord's argument that essentially casts O'Donnell's race as a Goldwater-style sowing of the seeds for future conservative victories in Delaware. The problem with this idea is pretty much the same as the problem with arguing for a moderate presidential nominee for the GOP after the debacle of 2008: it's already been done. O'Donnell ran for the Senate just two years ago, and got 35% of the vote. Conservatives have had our say already in Delaware, and hopefully someday we'll find another great candidate who can try again there, as Pete du Pont did. Supporting Castle can itself be reconsidered under the conditions of 2014. But losing with O'Donnell will accomplish nothing.
I know part of this whole thing is about sending an intraparty message that guys like Castle aren't acceptable, but have we somehow not sent that message already by knocking off Bob Bennett, Lisa Murkowski, Charlie Crist, Arlen Specter, Trey Grayson, Sue Lowden, and Jane Norton?
At the end of the day, yes: Mike Castle would be a weak, unreliable Senator who will do little for the conservative movement, and perhaps marginally weaken the movement within the GOP. But a Castle victory will also deal a blow to the movement's bitter enemies in the Democratic party, whereas an O'Donnell win is highly likely to increase the power of the Democrats. You want to choose likely defeat with a conservative, do what I'm doing and vote for David Malpass in New York tomorrow. But while there are real and serious arguments on both sides, I can't let an opportunity to defeat the Democrats and all they stand for pass by, not this year. If I was in Delaware, I'd vote for Castle.
« Close It
Posted by Baseball Crank at 12:30 PM | Politics 2010 | Poll Analysis | Comments (42) | TrackBack (0)
August 3, 2010
POLITICS: The Not-So Popular Party
Much of the behavior of Democratic Senate candidates can be explained by one simple fact: very few of them are going to get 50% of the vote this fall. Even the candidates who have a good chance to win are going to struggle to get to 50%. Let's take a quick look at the RealClearPolitics polling averages for the 22 Senate races that RCP lists as being in play, or where there's no average the last listed (generally Rasmussen) poll, just focusing on the Democratic candidate - as you will see, in only 4 of those 22 races is the Democrat polling 50/50 or better, and in only 7 of the 22 is the Democrat even polling above 43%:
These are the few popular Democrats. Wyden, Blumenthal and Manchin are all popular figures (an incumbent Senator, AG and Governor, respectively). The first two are solid favorites; Manchin could face a tough race if the GOP can tie him to the Obama Administration, which is deeply unpopular in West Virginia, but for now he's in a good place. Gillibrand has suffered soft approval ratings but for now has little known opposition.
Three incumbents in jeopardy (an incumbent polling below 50 is always considered in danger) and facing vigorous challenges, but all stand a fighting chance. Harry Reid, while he's revived from being down below 40%, simply can't crack 50% - neither can his son, running for governor - which is why his strategy is almost entirely built around the national media battering Sharron Angle and letting third-party candidates siphon off enough votes to let him win below 50 as he did in 1998.
Only Feingold and Bennet are incumbents here, and so a number of these races still have a lot of undecideds. (In Colorado, the Senate primary is something of a proxy war, with President Obama actively backing Bennet after failing to bribe Romanoff to drop out of the race, while former President Clinton is backing Romanoff). But again, voters don't seem too enamored of any of them, which is why the Democrats will be running an almost entirely negative fall campaign focused on driving small numbers of voters from the GOP to third parties.
Not all of these races are uncompetitive - incumbent Richard Burr is listed at a weak 46% in North Carolina, Mark Kirk is even below Giannoulias in Illinois, and of course in Florida the establishment Democrats are abandoning their own candidates to line up behind the incumbent, nominally Republican governor (a man the DNC stood ready to demonize if McCain had chosen him as a running mate in 2008), Charlie Crist. But once again, the divide-and-conquer strategy is basically the only way they can play this hand.
If the Democrats manage to hold onto the Senate, it won't be because the people of more than a tiny number of the states voting this November have actually given their performance a thumbs-up.
Posted by Baseball Crank at 5:02 PM | Politics 2010 | Poll Analysis | Comments (6) | TrackBack (0)
March 15, 2010
Good friend and RedState colleague Neil Stevens has launched Unlikely Voter, a new poll-analysis site. Go check it ouand get in on the ground floor of what is sure to be a busy site this election season. The initial explanatory post looks at the Specter/Sestak Senate primary race in Pennsylvania.
Neil is RS' resident tech/math guy, and aims to provide some mathematical rigor to the space already inhabited by RealClearPolitics' multi-poll averaging and Nate Silver's fivethirtyeight.com; while both of those sites are useful, RCP is an apples-and-oranges snapshot rather than an analysis site, and Silver's site, while superficially impressive, is too often driven by advocacy and in some cases apparent vendettas against particular polling firms, and tends at times to overstate the degree of certainty in its models, which tend to assume that all trends will continue indefinitely (like when Silver constructed a polling model predicting approval of same-sex marriage in 2009 by the following states: Vermont, New Hampshire, Massachusetts, Maine, Rhode Island, Connecticut, Nevada, Washington, Alaska, New York, Oregon) or tend to predict things like legislative votes that can't reliably be predicted with mathematical models.
We really have never had a satisfactory replacement in this space for Gerry Daly's site, and hopefully Neil will fill that gap.
November 12, 2009
POLITICS: Latest Connecticut Poll: Good News For Simmons, Bad News For Dodd, Obamacare
The latest Quinnipiac poll of Connecticut voters is out, and while it is (standard disclaimer) only one poll, it shows bad news for Chris Dodd, good news for his strongest challenger, Rob Simmons, and bad news for President Obama's health care plan.
Here's the topline result on Simmons vs Dodd:
Former Connecticut Congressman Rob Simmons has an early lead in the Republican primary race for the 2010 U.S. Senate contest and runs better than any other challenger against Sen. Christopher Dodd, topping the Democratic incumbent 49 - 38 percent...
The poll shows Dodd with a favorable/unfavorable rating of -15 (38-53) among men and -25 (34-59) among Independents, and a re-elect number of -24 (34-58) among men and -32 (30-62) among Independents, the latter mirroring the showing of Jon Corzine and Creigh Deeds among Independents.
It's still somewhat early to judge whether any of the other Republicans in the race would be electable against Dodd; clearly, Simmons, as a moderate former Congressman, has a very real shot of winning this race, as he's polling basically where Chris Christie was polling at this stage against Corzine. And bear in mind, this was a poll of registered, not likely voters; the likely-voter screens almost always help the GOP candidate, especially since 2010 will be an off-year election in which polls are consistently showing that voters on the Right are far more motivated and energized. Here's the poll's sample:
From November 3 - 8, Quinnipiac University surveyed 1,236 Connecticut registered voters with a margin of error of +/- 2.8 percentage points. The survey includes 474 Democrats with a margin of error of +/- 4.5 percentage points and 332 Republicans with a margin of error of +/- 5.4 percentage points.
I don't have offhand the overall registration breakdowns for CT. The sample here is 38.3% Democrats, 34.8% Independents and 26.9% Republicans, as compared to 2008 exit polls showing an electorate 43% Democrats, 31% Independents and 27% Republicans. So, Republicans aren't oversampled here; whether the poll oversampled Independents at the expense of Democrats depends on whether you think 2008 turnout is representative of what the electorate will be going forward.
Anyway, time will tell as to whether the other GOP candidates can credibly challenge Simmons. McMahon is clearly well-funded, and her pro wrestling background suggests some familiarity with the kind of populist appeal that made Jesse Ventura a governor, but Ironman at Next Right, a close observer of the CT political scene, thinks she is a poor stump speaker and too close with Rahm and Ari Emanuel and Lowell Weicker to be trusted, including a $10,000 donation to the DCCC in the fall of 2006 while it was pouring money into CT to help defeat Simmons and Nancy Johnson (McMahon herself didn't vote in that election). $3 million in state tax credits for WWE and a heavy WWE lobbying presence in the state capitol are also not the kind of resume lines that are likely to help a populist campaign against the goodies-collecting Dodd. All of which adds up to more reason why McMahon will have a long way to go to convince GOP voters that she's a better option against Dodd than Simmons.
As for Connecticut's other Senate seat, up again in 2012 and presently held by an incumbent from the Connecticut for Lieberman party, Jay Cost has argued that the 2006 race shows that Lieberman needs to win over Republicans and conservative-leaning independents to keep his job, and that this helps explain his opposition to Obamacare:
18% of all voters [in 2006] were self-identified Republicans who voted for Lieberman. 14% of all voters were self-identified conservatives who voted for Lieberman. Simply put, Lieberman won that 2006 race in large part because conservative Republicans voted for him, not Schlesinger.
The Q poll strongly supports Cost's thesis - Lieberman's poll profile is essentially that of a liberal Republican at this point, and Connecticut voters are far more skeptical of the Democratic health care plan than they are of Obama in general:
By a 51 - 25 percent margin, Connecticut voters say Sen. Joseph Lieberman's views on issues are closer to the Republican Party than to the Democratic Party. There is agreement on this among voters in all parties.
Note also that the poll shows that voters trust a Republican over Dodd on the health care issue, 43-37. And this is a liberal northeastern state; today's Q poll in Ohio, which shows some encouraging news for Rob Portman, has voters disapproving of Obama's health care plan by 55-36 and Obama's approval rating running lower than the Democratic Senate candidates.
As a final footnote, recanvassing shows that Bill Owens - who ran against the House health care bill, although he then voted for it as soon as he was sworn in - has lost a significant part of his margin of victory over Doug Hoffman (who also ran against the House bill) in NY-23. Even assuming that the net result of the recanvassing doesn't lead to any efforts to challenge the legitimacy of Owens' election, the dwindling margin of victory undermines efforts to make much hay of Hoffman's loss, and offers yet another data point - from the Northeast, no less - to suggest that support for Democrats and their health care plan is faltering almost everywhere.
If Connecticut is turning into dangerous turf for liberal Democrats and their big government schemes, that should be a sign to encourage opponents of big government everywhere to get in the game.
Posted by Baseball Crank at 12:11 PM | Politics 2009 | Poll Analysis | Comments (7) | TrackBack (0)
October 22, 2008
POLITICS: Nobody Knows Nothing
I'm not in the group that says McCain is secretly winning and the polls are a gigantic false-flag psy-ops program designed to discourage GOP turnout. We know, after all, that whatever the biases of various people and institutions involved, the final polls in 2004, properly understood, were highly accurate, and the 2006 polls were mostly so as well (2002, less so). Neither am I in the group that attributes potential poll inaccuracy entirely to Obama's race - while that may well be a factor, I think there are fair arguments that run deeper to polling methodology, and I also think Obama's inexperience, the absence of a candidate from the incumbent administration, the massive new-voter operation by the Obama camp, the Palin wildcard (this has to be the first time ever that a VP pick drew nearly the same convention speech audience as the POTUS nominees and the VP debate outdrew the POTUS debates, and now she's delivered the biggest ratings for SNL in 14 years) and the sudden, late external shock of the credit crisis are all reasons why public sentiment may be more volatile and harder to get a fix on than usual.
Polls are not votes. They are evidence. The likely answer from the evidence we have remains that McCain is losing and likely to lose; I'm not going to cocoon myself or anyone else from that (there's a reason why candidates who say "the only poll that matters is the one on Election Day" usually end Election Day with a concession speech). But there is more than enough uncertainty out there that I endorse wholeheartedly the view that the last thing Republicans and other McCain supporters should do is get discouraged and throw in the towel before every stop is pulled out to win this thing. You gotta be in it to win it.
UPDATE: Geraghty notes here and here the panicked frenzy of attacks he gets from the Left whenever he suggests that some polls are showing a closer race than the conventional wisdom (or the bulk of polls, for that matter). I chalk this up partly to the Online Left's longstanding view that the winner of any argument is the person who can demonstrate the greatest degree of anger, but it's certainly a curious phenomenon coming from people who would seem to have every reason to be confident and no particular reason to take time from their day getting angry at a conservative pundit for showing a glimmer of optimism. Unless you do buy into the view that such people really are banking very heavily on a demoralized opposition.
SECOND UPDATE: It appears that the AP poll showing Obama up only by one has a 4-point advantage for the Democrats in the party-ID breakdown. For anybody who has followed the polls this year, that's the single biggest question: when you factor in GOTV and whether the likely-voter screens and all have or have not accurately predicted who will vote, will the party ID numbers look like 2006, when a terrible climate for Republicans still produced just a 3-point advantage in party ID for Democrats in the exit polls? Or will it be more like a double-digit advantage in party ID, figures we have not seen since the 1970s? Note that the Geraghty posts I linked to up top show very few examples of dramatic changes in party ID year to year. Even in Jay Cost's chart of registered voter ID, the biggest swings are about 7 or 8 points in some years, 1984 and 1994 for the GOP and 1996 for the Democrats. It may well be that 2008 really will show a historic realignment away from the position the GOP held in 2006 (which was already awful, the worst Republican year in a decade), but just bear in mind that it has to be for the bulk of this year's polls to be accurate.
Posted by Baseball Crank at 12:44 PM | Politics 2008 | Poll Analysis | Comments (14) | TrackBack (0)
October 7, 2008
POLITICS: Yes, We...Can?
So far as I can tell, nobody in the history of modern polling has won a presidential election from as big a hole as John McCain now stands in, at last check a national polling advantage in the neighborhood of 5 points for Obama. Now, if you are a betting man, surely you like your odds on Obama. But does that mean that the race is over? Perhaps, but not necessarily. While the circumstances are of course different, we have seen two past Republican campaigns, neither of them headed by the most dynamic of campaigners, provide examples of strong closing-month performances.
The most obvious recent example was 1996. The Gallup poll, which admittedly is one of the more volatile polls (Obama presently leads it by 8) on October 6/7, 1996 showed Bill Clinton with a commanding 22 point lead, 56-34 over Bob Dole with 5 points for Ross Perot (the first of two debates was on October 6). Four days later, after the first debate and the Vice Presidential debate, that lead was 57-34 (Clinton +23). In an October 14-15 poll, conducted on the eve of the second, October 16 debate, Dole pulled much closer (48-39, Clinton +9), but as late as October 20-21 the poll showed Clinton up 19, 52-33 with 8 for Perot. Dole then began his serious charge, pulling above 40% for the first time on November 4-5, to finish at Clinton +11 (52-41-7), and ended up at Clinton +8 on Election Day, 49-41. Dole thus ended up shaving as much as 15 points off Clinton's lead in less than a month.
Then there's 1976. Jimmy Carter had, of course, famously led by 34 in one midsummer poll...in a poll conducted September 24-27 (the first debate was September 23), Carter led 51-40 (+11), but in one conducted September 27-October 4, that lead dropped to +2, 47-45. Carter widened his lead to +6 on October 8-11 after the famous "Democrat wars" gaffe by Bob Dole in the October 6 VP debate, led +6 (47-41) on October 15-18 (the second debate, with Ford's Poland gaffe, was October 15), was still at +5 on October 22-25 (the third debate was October 22), but an October 28-30 poll for the first time showed a Ford lead, 47-46. On Election Day, Carter won 50-48.
Polling today is more sophisticated, of course, and there are other distinguishing factors as well. On the one hand, the 1996 election had a third party candidate who surged up to double digits in late October, and Dole was running so far behind a still-strong GOP Congressional brand (Republicans held both Houses of Congress through that race) that a good deal of his late surge was just natural Republicans coming home. Some of the same was true of Ford's surge. On the other hand, the 1996 race should have been much less volatile than this one - it matched a 3-decade Senate veteran with a sitting president in a time of peace and prosperity - yet the polls showed significant movement late in the game. 1976 was more similar to the present race, as it pitted a moderate Republican running in a time when the GOP brand was as destitute as it has been since the New Deal, matched against a relatively green and unknown opponent. And of course, this year's race involves not only an unprecedentedly inexperienced and far-left presidential candidate and times of economic uncertainty and foreign war but also the triple complicating factors of no incumbent, Obama's race, and McCain's age coupled with Palin being not a whole lot more experienced than Obama. Those are all reasons why we might expect more, rather than less, real underlying volatility in voter preferences in addition to the possibility that the polls themselves are having trouble measuring the race. And at the end of the day, while it may at first glance seem harder to push upward in the polls against the headwind of a bandwagon once the media has (correctly) called the race for the frontrunner, as in 1996, there is a difference in the degree of difficulty between pulling up close to 50 and breaking through it.
Again: none of this should be reason for Republicans to celebrate - as I said, nobody in a hole like this has actually won a race. But history tells us that voter preferences can still shift in the last month, and if Obama's lead now is accurately reflected by the RCP average of +5.3, it is still very much worthwhile for McCain-Palin and their supporters to fight on to the end.
Posted by Baseball Crank at 9:30 AM | Politics 2008 | Poll Analysis | Comments (31) | TrackBack (0)
September 16, 2008
The further we get into the fall, the more meaningful the state-by-state polls become. But it's nonetheless useful to bear in mind the hard numbers from past years to keep a realistic view of what the range of possibilties are in any given state. A few months back, I had gone through the Federal Election Commission website and put together a spreadsheet, which I'm only getting back around to now, tallying up all the votes for federal office (President, Senate, House) in the last four election cycles (2000, 2002, 2004, 2006) comprising two presidential elections, four House elections, and a full cycle and a third of Senate races. The chart below lays out the results.
Now, let's be clear: while the underlying numbers are actual votes cast, basically what I'm doing here is using a metric, not a statistic; I'm combining different types of votes over time in a way that's not scientific, but rather an effort to take disparate pieces of data and make them digestible. Obviously, there are a host of reasons why this isn't science: turnout is much larger in presidential years, some incumbents in the Senate and House run unopposed (although this is itself usually a sign of strength), a third of the Senate seats are counted twice here, gerrymandering affects House races, and of course, there's no fixed way to measure the relative probative value of 2006 results vs. 2000 results in measuring 2008's political terrain. That said, using three levels of balloting over four election cycles does help give us a large enough sample size to get a look at the real, underlying partisan makeup of particular states, and limit the distorting effects of individual personalities.
Here's the methodology. I present two sets of numbers: "raw" numbers that treat each of the four elections alike, and "weighted" numbers that give a larger weight to more recent results. For the raw numbers, I tallied up all votes cast for each of the two major parties (ignoring third party votes, for simplicity's sake) in presidential, Senate or House races in 2000, 2002, 2004 or 2006. For the Weighted totals, I weighted the votes by year as follows:
i.e., a vote for a House candidate in 2006 was worth twice the weight of a vote for the same candidate in 2002, and four times the weight of a vote for that candidate in 2000.
The final two columns attempt to combine the electoral vote weight of each state with its partisan composition in order to put the closeness of the state in the context of the reward for presidential candidates of swinging it, dividing the number of electoral votes by the square of the margin separating the two parties (the sum is then divided by 100 just for ease of the reader). The equation is:
Without further ado, here is the chart:
Read More »
A couple of things jump off the list:
(1) Yes, as we all know, Colorado and Pennsylvania are two of the really critical battlegrounds in 2008, and not only at the presidential level.
(2) Some states are genuinely competitive yet never become swing states at the presidential level. Maine has two Republican Senators because they are much more liberal than any national Republican; South Dakota has a Democratic Senator and at-Large Congresswoman because they are much more conservative than any national Democrat, and aren't running for Commander-in-Chief.
« Close It
Posted by Baseball Crank at 9:24 AM | Politics 2008 | Poll Analysis | Comments (5) | TrackBack (0)
September 11, 2008
POLITICS: Narrowing The Field
With 54 days until the election and four debates still to go, a lot can happen; the presidential race could still end up getting badly away from either McCain or Obama. But now that we have the benefit of polling done entirely after the two VPs were picked and the two conventions held, it's possible to get a sense of what the playing field really looks like. On a national level, the race is still close, but looks much better for McCain, who leads by 2.5 in the RCP poll average; of the 9 polls listed, McCain leads in 6, Obama one, and two are tied, with all showing fewer undecideds than existed a month ago but only one poll giving either candidate 50% (the USA/Gallup poll showing a 54-44 McCain lead among likely voters - a result that would mean the race is effectively over if it was repeated in multiple polls, but which is apparently a serious outlier).
The race, however, will be conducted on a state-by-state basis, which sends us back to the Electoral College. You can run the polls yourself, but below the fold I will walk through what my gut is telling me after looking at those polls. The bottom line is that for all the talk of how Obama and McCain were map-changing candidates, this race now looks like it will go down to the wire in just a handful of crucial battleground states, with most of the Bush-Gore/Bush-Kerry red-blue patterns holding steady (the persistence of these patterns being good news for Republicans after the 2010 census, but that's another day's argument).
Read More »
The RCP map shows Obama up 217-216, with nine states up for grabs
I generally think that's correct as far as the states that RCP has moved into each candidate's column - those are states that are not going to be in play unless you get a big national movement. For example:
1. McCain largely burned his bridges long ago in Iowa, a 2004 Bush state, by his principled opposition to ethanol subsidies; he skipped the 2000 Iowa caucuses and finished a distant third there in 2008. Obama, by contrast, is one of the ethanol industry's largest recipients of cash and (perhaps not coincidentally) a booster of subsidies. Iowa launched Obama, and is likely to stay in his column along with politically similar Minnesota and Wisconsin, even though all three will end up being fairly close.
2. A surge of African-American voter turnout will make North Carolina closer, but I expect McCain to hold his turf there.
3. Florida may be close in the polls, but really it's been a steady Republican state. There were two anomalous factors that conspired to make it close in 2000: the popularity of Joe Lieberman on the ticket with older Jewish voters transplanted from the Northeast, and the early call by Fox and other networks for Gore that sent Republicans home in the panhandle before the polls had closed. Absent those factors, 2002, 2004 and even 2006 were all good Republican years in Florida.
When push comes to shove, I also expect Indiana - a rock-ribbed Republican state even in the Clinton years - and most likely Ohio and Virginia to stay home with the Republicans, close though all three will be, and while Obama has struggled in Pennsylvania (recall the "bitter" comment and his thumping in the primary there), I suspect that traditional party loyalty and Ed Rendell's machine will put him over the top at the end. As you will see as you walk through these maps, it's all but impossible for McCain to win without both Ohio and Virginia (and, obviously, Florida) - he's unlikely to swipe Michigan or Pennsylvania unless he's winning Ohio - and even more implausible for Obama to win without Pennsylvania. These are bedrock states of each party's path to victory.
If you add in those states to each side, that gives us a map with only five swing states, and a 260-238 McCain lead:
McCain has a lot of atmospheric help going for him in Michigan: Democratic mismanagement of the state government and the economy; the arrest and resignation of Detroit's Democratic mayor, who supported Obama; lingering bitterness over Obama's effort to avoid seating the Michigan delegation during his battle with Hillary. On the other hand, McCain got beat in the Michigan primary himself after his comment about jobs not leaving and not coming back was taken as a sign of undue pessimism about the economy. Fundamentally, though, Michigan is a Democratic state, albeit narrowly; it's still an uphill battle. If we assume Obama holds onto it, we get McCain 260, Obama 255, with only Colorado and Nevada (won by Bush twice), New Mexico (won by Bush and Gore), and New Hampshire (won by Bush and Kerry):
All four of these states seem legitimately too close to call if you've been reading the polls. There are individual factors at work. McCain's been popular forever in New Hampshire, site of his crucial primary victories in 2000 and 2008 (both of which depended on his popularity with independents) and the home of countless McCain town halls and bus rides over the years, whereas Obama fared poorly in the primary there. In Nevada, the Yucca Mountain nuclear waste disposal site is a flashpoint; McCain's support of the site and ambitious plan to build more nuclear plants, compared to Obama's unwillingness to embrace either, put McCain in a bind there. Colorado's been trending Democratic due to a large influx of Latinos as well as liberal Californians. Then again, the distinctly Western flavor of the McCain-Palin ticket could prove appealing over prolonged exposure in CO, NV and NM.
This is where things get really hairy, because there are three different combinations (McCain wins CO, or NM + NH, or NV + NH) that get us a 269-269 tie. I think we can all agree that this would be a terrible outcome for the nation, and would cripple the next president's ability to govern, just as the recount made it impossible for Bush, even before he took office, to even approach the "uniter, not a divider" tag he'd campaigned under; the Democrats were permanently estranged from him before Day One.
The first problem, if there's a 269-269 tie, is the "faithless elector" problem, i.e., some elector bolting sides to break the tie, a result that would create an enormous outcry. If we get past that, the 12th Amendment explains what happens:
The person having the greatest number of votes for President, shall be the President, if such number be a majority of the whole number of Electors appointed; and if no person have such majority, then from the persons having the highest numbers not exceeding three on the list of those voted for as President, the House of Representatives shall choose immediately, by ballot, the President, the votes shall be taken by states, the representation from each state having one vote; a quorum for this purpose shall consist of a member or members from two-thirds of the states, and a majority of all the states shall be necessary to a choice. And if the House of Representatives shall not choose a President whenever the right of choice shall devolve upon them, before the fourth day of March next following, then the Vice-President shall act as President, as in the case of the death or other constitutional disability of the President.--The person having the greatest number of votes as Vice-President, shall be the Vice-President, if such number be a majority of the whole number of Electors appointed, and if no person have a majority, then from the two highest numbers on the list, the Senate shall choose the Vice-President; a quorum for the purpose shall consist of two-thirds of the whole number of Senators, and a majority of the whole number shall be necessary to a choice.
We'll skip over here the part where some bizarre deadlock prevents the House from deciding, and Dick Cheney ends up the President...basically, the vote would come down to which party controls the most state delegations in the House, and at least at present, that's the Democrats by a margin of 2 or 3 states as of last count - I believe the most recent special elections swung them another state. I'm sure we'll all count more closely if it happens. Of course, that's the current Congress; I'm a little less clear on whether the current Congress or the one elected in November would tally the electoral votes...but assuming it's still the Democrats, this probably means Obama wins if it's 269-269.
Then we get the VP, picked by the Senate. The Senate currently has 49 Democrats, 49 Republicans, one Socialist, one member of the Connecticut for Lieberman party who has endorsed McCain-Palin, and a tiebreaking vote by Dick Cheney. My guess is that given Senatorial courtesies and the like, especially with Biden as the opponent, McCain would probably dissuade Republicans from putting up a fight to saddle Obama with Palin as his VP, another outcome that would be highly unstable and bad for the country.
Well, that was a long digression into the parade of horribles, but the bottom line is, McCain needs 270 to win, Obama probably needs 269. And at this writing, the single state most likely to swing that difference is Colorado. The odds are pretty good that the margin of victory will be one state, maybe two, that are decided by just a percentage point or two.
Fasten your seatbelts.
« Close It
Posted by Baseball Crank at 7:38 PM | Politics 2008 | Poll Analysis | Comments (17) | TrackBack (0)
July 29, 2008
POLITICS: Polls in Perspective
We've had wild-swinging polls lately - national polls showing Obama +9 and McCain +4 - but the important thing is not to panic. Polls go up, they go down, and at any rate while the national polls can give you some idea of the direction and momentum in the race, in the end the only polls that matter are on a state-by-state basis, and those really don't get hugely meaningful until after the conventions.
Anyway, this post has an excellent look back at past July 4 Gallup polls. It's a useful reminder that these races can move in a number of different directions, and the one thing they almost never do is finish in November right where they were in July.
I think the race right now is where the bulk of the polls seem to say it is: Obama's winning, but not by a big margin, and he's a long way from putting McCain away, especially in the key swing states. Obviously, you can find people to argue that there's an underlying dynamic underway that will sweep one or the other of them away, and maybe that will look clear in retrospect. But for now, it's still a race. Which is why the Veepstakes buzz (at present, leaning towards Tim Pawlenty and Mitt Romney on the R side, Tim Kaine on the D side) is being scrutinized so heavily for its impact on particular states and blocs.
Posted by Baseball Crank at 12:33 PM | Politics 2008 | Poll Analysis | Comments (7) | TrackBack (0)
June 4, 2008
POLITICS: Obamomentum - End-of-the-Primaries Edition
Now that the Democratic primaries are finally over, let's take one last look at the charts I have been running for some time now (see here, here and here) of the Democratic presidential primary popular vote totals for the months of March, April, May and now June. (Source: RCP, except I used CNN's updated figures for Montana) - "margin," of course, is Obama's margin of victory/defeat in each primary:
In other words, Obama ends the last 3 months of the primary season more than 600,000 votes in the hole, losing the popular vote decisively to Hillary over a stretch of 16 primaries in which 13 million votes were cast. In percentage terms, Hillary's 4.52% margin of victory for that period is larger than the general election margins of Bush over Kerry in 2004, Carter over Ford in 1976, Nixon over Humphery in 1968, Truman over Dewey in 1948, and just a point smaller than that of Clinton over Bush in 1992. He lost six different primaries by margins of 100,000 or more votes. All this during the time period when he should have been sealing the deal with Democratic voters after having taken what looked at the time like a decisive, momentum-tipping lead in mid-February. The final insult was losing South Dakota, a state he was widely projected to win and in which he led decisively in the few polls taken until the last day or two before the election, and which cast its ballots while the vultures were visibly circling Hillary's campaign.*
It's Obama's weakness in that period even within his own party that has to be troubling to Democrats pondering his chances in November. Recall that Obama faced virtually no serious scrutiny until he pulled within 20 points of Hillary in the national popular vote in mid-December in the immediate aftermath of Oprah's campaign appearances on his behalf (he didn't pull within single digits until after he won Iowa); it was only after he was christened the clear frontrunner that he started to take serious fire, beginning in late February and early March with Hillary's "3 a.m." ad, the Goolsbee/NAFTA flap, and of course the Rev. Wright story, and continuing with the accumulation his radical left-wing associations, his endless stream of verbal flubs, and his ever-growing list of friends, mentors and staffers cast under the bus. Jay Cost has a great series of posts (start here) on the demographic breakdown of how and where Hillary beat Obama; consider this, among his many charts, looking at the states Bush won in 2004 that the Democrats would have some hope (and, obviously, need) to pick off in 2008:
That there is not much of a winning coalition in most parts of this country.
Let's also wrap up my look at turnout, using the same baseline as before (the number of votes in 2006 for House Democrats in the state) - this time, I'll just run the chart just for the same time period (the full chart and explanation of sources is here), but leaving out Puerto Rico, where I used a different and ultimately inaccurate baseline to capture the relatively disappointing turnout:
Turnout started to tail off after the early May primaries. In general, through the primaries and caucuses (especially caucuses), Obama tended to do better in the lower turnout states, but we see here that at the very end, he did better in Montana, which had fairly robust turnout, than in South Dakota, where even adjusting for the fact that Stephanie Herseth's 2006 victory may set an unrealistically high bar for turnout, the turnout was quite low.
Read More »
« Close It
Posted by Baseball Crank at 9:29 AM | Politics 2008 | Poll Analysis | Comments (9) | TrackBack (0)
June 1, 2008
POLITICS: Puerto Rico Polling
Just a note: I'll update the Obamomentum charts after Tuesday's balloting. From CNN's report on Hillary's victory in today's Puerto Rico primary, it appears that turnout was much, much lower than my back-of-the-envelope projection from the 2004 general gubernatorial election, but Hillary's margin of victory was still 142,000 votes due to her overwhelming 68%-32% margin. In other words, this is the sixth time in three months that Obama has lost a primary by 100,000+ votes.
UPDATE: Patrick Ruffini notes the wide margin by which Hillary's victory exceeded late polls. There are a number of possible explanations for this - perhaps the polling was done poorly, perhaps people who say they will vote for Obama are just disproportionately unlikely to show up to vote, perhaps people even in Puerto Rico are unwilling to admit on the telephone that they are not voting for Obama - but it does rather undermine the idea that Obama's persistent underachieving of the polls and exit polls is just a feature of racist white people.
Posted by Baseball Crank at 11:47 PM | Politics 2008 | Poll Analysis | Comments (1) | TrackBack (0)
May 21, 2008
POLITICS: Obamomentum, Kentucky and Oregon Edition
Well, it's time once again (see here and here) to update the chart with what should now be the complete Democratic presidential primary popular vote totals for the months of March, April and May - I delayed in getting this posted because it took until very late today to get all the votes tallied from Oregon, and in fact I'm running this with 99% of precincts there reporting because who knows how long that last 1% will take. Of course, a handful of late votes are still trickling in from IN & NC, too.
(Source). In other words, even before we get to Puerto Rico, Obama is nearly half a million votes in the hole since the events of late February and early March (i.e., the 3am ad, the Goolsbee/NAFTA flap, and of course the Rev. Wright story). Now, let's look at turnout, using the same baseline as before (the number of votes in 2006 for House Democrats in the state) - this time, I'll just run the chart just for the same time period (the full chart and explanation of sources is here):
You will note that turnout was down from the boom in the early May primaries, but at least in Kentucky, the voters still turned out in quite large numbers in the face of the Obama camp's argument that there was nothing left to vote on. Obama had more success in Oregon, where turnout was much lower.
Let's update last week's projections by bringing the turnout back down to 110% for South Dakota and Montana relative to the 2006 figures, while still using the conservative 60% estimate for Puerto Rico. I still don't have new polls for Puerto Rico, South Dakota or Montana - I'm still using the one poll each from the first two and the imputed poll results from the matchup with McCain for the third. Here's where that would get us:
How much guesswork is in this last chart? A lot. We really are operating in the dark as to these last three primaries. Conservative South Dakota blogger Ken Blanchard argued last week that the poll in his state is too optimistic for Obama:
The last poll taken in South Dakota showed Obama 10 points ahead of Senator Clinton. But those results were announced back in April, and a lot of muddy water has gone down the Big Jim River since then. If I had to guess, I would guess that Obama figures to lose South Dakota. He is showing up to show that he isn't just writing off the rural states. If I were him, I would skip the Watertown trip, and instead visit Cabelas in Mitchell. A few photos of him buying a pair of hip waders, and sipping coffee with the locals would be just the thing. It is the Cabelas demographic he is having trouble with.
We shall see; one would hope someone would bother running one last poll. For Obama, though, the best news is that there are only three more of these to go.
Posted by Baseball Crank at 8:39 PM | Politics 2008 | Poll Analysis | Comments (2) | TrackBack (0)
May 14, 2008
POLITICS: Obamomentum, West Virginia Edition
It's not every day you see the presumptive nominee lose a presidential primary in a swing state by 41 points (by contrast, despite a persistent protest vote faction, McCain hasn't actually lost a primary since Kansas and Louisiana on February 9), but that's exactly what happened to Barack Obama last night in West Virginia, and suggests pretty strongly why his campaign seems to be writing off the state for November.
Anyway, let's update the chart I've been running (last installment here) showing the popular vote trend since Obama's armor started to crack at the beginning of March. Here's the current chart:
Attentive readers will note that some of the earlier states' vote totals have changed slightly; I'm using the RCP figures, which presumably got some late corrections on TX, PA, IN and NC. RCP also has the overall numbers, which mostly narrowly favor Obama, though it depends on what states you count; in any event, recall that Obama didn't even crack 30% in national polls against Hillary until after he won Iowa, and didn't catch her until early February, after trailing by double digits for most of 2007; much of the Obama vote in the first two months was basically the honeymoon of a challenger who had yet to be seriously vetted. So while looking at the last two and a half months doesn't cover the whole race, it likely tells us a lot more than looking at things that happened a political lifetime ago. If you knew nothing else of this race, you'd certainly look at those trendlines and think, especially after last night, that Hillary was really pulling away by now.
Where from here? Another interesting thing is that my rough-estimate projection last week vastly underestimated Hillary's West Virginia margin of victory, which was twice what my mostly unscientific model had projected. Partly that's because I was working off one poll with a lot of undecideds, but partly it's because turnout was larger than the baseline I was using, the number of votes for House Democrats in the off-year 2006 election, which I took as a reasonable proxy for the number of people available to vote in a Democratic primary. I'll attach at the bottom of this post a chart showing how that figure has served as a predictor of turnout, but the take-home from that exercise is that West Virginia Democrats came out in very large numbers despite their betters in the media telling them the race was over, already, man - not quite as large numbers as we saw in Indiana, North Carolina, Mississippi and Texas, but larger proportional turnout by this metric than in Pennsylvania, Ohio, Wisconsin, or Virginia. The voters apparently think there's still something to vote on.
If we update last week's projections, then, with more recent polling and an assumption of 125% rather than 100% turnout relative to the 2006 figures, what do we get? I'm using now the RCP averages for Oregon and Kentucky, but I don't have new polls for Puerto Rico, South Dakota or Montana, and I'm still using the conservative 60% estimate for Puerto Rico:
In other words, if form holds - even using the conservative projection of the vote total in Puerto Rico - Obama could end up well over half a million votes under water for the last three months of the primaries. We can only speculate as to why Obama has been struggling like this - whether it's a sign of Hillary's strength, the nature of the later primary states, a temporary weakness or permanent damage to Obama - and whether it will carry over in the general election campaign. But I can say this: if Obama was a fatally damaged general election candidate by this point, this is pretty much how you would expect him to be doing in the late primaries.
Read More »
Here's the turnout chart, which is interesting in its own right:
The primary vote total is just Obama's and Clinton's votes, so it may underreport turnout in a few of the early races when some people were still casting votes for Edwards (in Michigan, where Obama didn't bother getting on the ballot, I just added Hillary plus the "uncommitted" vote). The states marked "NA" are caucus states that have not released vote totals; the states with turnout below 20% are mostly caucuses as well, most of which Obama won. We can only speculate whether primaries in those states would have produced (1) more votes for Obama or (2) a different outcome. At any rate, you can see that turnout by this metric had largely hovered around 100% until the more recent primaries that saw a surge of interest - another reason why the March-April-May results should be more worrisome to Obama backers, as despite his claims at bringing in new voters, he actually tends to do worse the larger the turnout.
« Close It
Posted by Baseball Crank at 12:01 PM | Politics 2008 | Poll Analysis | Comments (9) | TrackBack (0)
May 7, 2008
POLITICS: Obamomentum, Revisited
Most anyone watching the primaries had expected all along that Obama would win North Carolina - where the Democratic primary electorate is dominated by African-Americans and college towns - and Hillary would win the more conservative white Democrats in Indiana last night, but Hillary's relatively narrow margin of victory in Indiana and the simple fact that Obama notched a victory in a state of significant size after a string of losses both add up to an undeniably good night for Obama. Let's update the chart I ran previously of the popular vote since the beginning of March:
As you can see, over this period - covering the time after the genuine cracks in Obama's previously untouched public brand image had appeared - Obama is still behind in the popular vote, and with only Oregon on May 20 as a likely source for significant number of votes for Obama, that's not going to change.
That's even before you deal with the exit polls - I'll leave the dissection of those to others, but it seems pretty clear that Obama is getting crushed among white and Latino voters, and you can't win much of anything in these United States without those two groups. It's also before you deal with the popular vote for January and February, which is harder to measure because you get into the question of how to estimate the caucus popular votes (in some states, these were not recorded) or whether to count Florida and Michigan:
Here's the remaining schedule, with a chart showing the most recent poll I could find - I used Rasmussen for West Virginia, Kentucky and Oregon, a mid-April Dakota Wesleyan poll for South Dakota, a mid-April Puerto Rico poll, and, lacking a head-to-head poll, I used Rasmussen's general election numbers for Montana, which show Obama polling much better, but with basically similar numbers to the South Dakota poll (but note that unlike earlier Obama mountain-state victories these are primaries, not caucuses). I then projected the number of voters - for the states, I used the number of ballots cast for Democrats in the House in 2006*, since this seems to have been a fairly reliable proxy for the number of ballots cast in the primaries in Pennsylvania, Indiana and North Carolina; Puerto Rico is more challenging, but to be conservative I just assumed a turnout of 1.2 million voters, which is roughly 60% of the 2004 gubernatorial general election turnout (in which above 80% of registered voters voted); as Ben Domenech has noted, given Puerto Rico's traditionally high voter turnout and the realization that this may be a unique opportunity to affect the mainland presidential election, if Hillary's still battling at this point the turnout could be much closer to the general election figures:
Obviously, these are very rough estimates, especially since some of these polls have upwards of 20% of the electorate undecided, but you get the general idea. Much will depend on the turnout, especially in Puerto Rico, but I think it's a safe bet that when all is said and done, Obama will be down somewhere in the neighborhood of 400,000 votes for the period covering the last three months of the primary campaign. Heck of a way to launch a general election campaign.
Read More »
* - For the mathematically curious:
I'm using FEC sources for this except for the Puerto Rico numbers linked above, and RCP for the popular vote chart.
« Close It
Posted by Baseball Crank at 9:22 AM | Politics 2008 | Poll Analysis | Comments (28) | TrackBack (0)
April 19, 2007
POLITICS: Hillary and McCain
RCP looks at the gulf between Hillary's strong position in the polls and her weakness among the lefty "netroots," in large part driven by her stance on the war. In a lot of ways this mirrors John McCain, who is still polling OK but is almost friendless in the blogosphere. In McCain's case that may be due in part to the fact that
1. Bloggers are independent writers about politics
and thus he may be disproportionately unpopular with bloggers. I think it's at least partly the case that there will be some convergence in each camp, although the presence of Obama does create a serious threat to Hillary that no other candidate could have mounted.
Posted by Baseball Crank at 6:28 PM | Politics 2008 | Poll Analysis | Comments (1) | TrackBack (0)
June 2, 2005
POLITICS: 2006: The Terrain
There's been a lot of talk, more than usual this early in the election cycle, about the 2006 Senate races and the odds of either party picking up seats and changing the dynamics in a Senate now perennially deadlocked over judicial nominations and other business. In fact, much partisan strategy over these battles will, as always, be shaped by the prospects for the next election - where the parties hope to gain, where they fear to lose, and whether they expect to be dealing from a stronger or weaker hand come January 2007. With that in mind, let's take a look, using some hard numbers, at the political terrain for the 2006 Senate races.
There are polls, of course, but polls this early are volatile. Before we get to the polling data, there are two main pieces of hard data - actual votes - that we can use to evaluate the political climate in a state entering the beginning of a Senate race. The first is the red/blue issue: when people were paying greatest attention, which party did they side with? The polarizing nature of the 2004 election, with a conservative Republican and a liberal Democrat, sharpened that distinction. The second is the history of this Senate seat: how did the incumbent do in his/her last election? This second item is of particular importance where the incumbent is running again, although you do have to bear in mind that you are dealing with election results from six years ago, before 9/11, the Iraq War, the Florida Recount, Enron, judicial filibusters, Terri Schiavo, blogs, etc., etc., etc. Rather than rest on one or two of these data points, let's combine the two. I present a ranking of the Senate seats to be contested in 2006, from most to least likely to change parties, based on adding (1) the incumbent party's percentage of the vote in the last race for this seat (S%) to (2) the incumbent party's percentage of the vote in the 2004 presidential election (P%) (all numbers from FEC sources here, here and here):
Observant readers will note that I'm missing a state, Vermont. The problem is that Jim Jeffords ran there as a Republican in 2000, so it's hard to make anything of his 65.56%-25.42% thumping of his Democratic opponent. Kerry won 58.94% of the vote in Vermont, so if you double that and throw out the Jeffords anomaly, the D% should probably be 117.88, ranking the state near Maryland as an open seat the Democrats ought to be able to defend.
These notes are important. John Kyl is in a very strong position, but he ran unopposed in 2000; he's not quite as bulletproof as he looks. The Democrats may seem weak in several spots because they ran the table in close Senate races in 2000, but several of those candidates knocked off incumbents last time around, and will start in a stronger position this time around with the headwind of incumbency at their backs rather than in their faces. I figured "divided opposition" where the two main candidates pulled below 96%, leaving a number of voters on the table, but since Ted Kennedy beat his opponent 72.69%-12.86% in 2000, that doesn't amount to much.
I'd hesitate to say what threshhold indicates a realistic chance of a seat changing hands, but obviously anyone below 100 has to be viewed as an opportunity for the other side, and anyone above about 110 is - other than open seats - an extremely tough race. You can see that most of the most competitive races, based on this criteria, involve Democratic-held seats.
Of course, all of this is prologue; the 2006 races will be fought, like every election, with a new backdrop of issues and partisan mood and momentum, which so far seems to be favoring the Democrats. The number of genuinely competitive races is bound to be reduced if credible challengers can't be located, as was the case in 2004 in Nevada, for example, where Harry Reid was vulnerable but the GOP couldn't get a serious challenger. But the numbers above at least provide a solid guide to where the needle stands entering those races, and how far it has to move to save or defeat the incumbents listed above.
UPDATE: Don't miss Gerry Daly's effort to tweak some of the variables here to create a more accurate measure. I don't necessarily agree that the other Senate seat is all that instructive, as demonstrated by longstanding in-state splits: D'Amato and Moynihan, Grassley and Harkin, Domenici and Bingaman, etc. and the fact that some states just get in the habit of re-upping incumbents. For example, the persistence of Robert Byrd does nothing to help Jay Rockefeller. On the other hand, for similar reasons, I'm inclined to agree with a commenter at RedState that the last Senate election isn't that useful in evaluating open seats, at least not in the case of someone like Frist or Sarbanes who ran as an incumbent six years ago.
Read More »
UPDATE #2: Here's the table re-done to just double the presidential vote in the four states where there's an open seat:
« Close It
Posted by Baseball Crank at 6:45 AM | Politics 2005 | Poll Analysis | Comments (5) | TrackBack (5)
November 16, 2004
POLITICS: Exiting The Democrats
You have to take the national exit polls with a grain of salt, but it appears that this poll weights out to the correct result, and if so, a few things jump off the page:
1. Bush won white voters 58-41. He won white males by 25 points and white women by 11. Now, I know white people aren't exactly a cohesive group, and that there's something vaguely distasteful, even, about speaking of a "white vote". But if you're not even competitive with a demographic that constitutes 77% of the electorate, you got problems. Similarly, 81% of the electorate consists of Christians, and while the poll doesn't combine Protestant and Catholic, if my (rusty) algebra is correct, Christians voted for Bush by a margin of 57-42. At the cross-section of the two majority groups, 61% of the electorate is white Christians, and they broke 63-36 for Bush. Again, you can't afford to lose by that kind of margin with a majority voting bloc.
2. 49% of voters trusted Bush and not Kerry to fight terrorism, and those voters broke for Bush 97-3, such a decisive margin as to suggest that this issue was a deal-breaker for nearly half of all voters. In short, all else aside, Kerry was about 99% defeated just by the lack of voter trust in him as a war leader. This is supported by the fact that voters who trusted both candidates on terrorism broke for Kerry 75-24, while voters who trusted both candidates on the economy broke for Bush 61-38.
Posted by Baseball Crank at 7:41 AM | Politics 2004 | Poll Analysis | Comments (1) | TrackBack (0)
November 7, 2004
POLITICS: Where Bush's Swing Voters Came From
In this post, I examined the national popular vote and concluded that, comparing of the increased number of Bush voters from 2000 (about 8.66 million) and the increased number of Kerry voters as compared to Gore voters in 2000 (about 4.56 million), one of two things had happened - either:
1. Bush had won the votes of 65.5% of "new voters," defined as people who - regardless of whether they had voted in past elections - didn't vote for either Bush or Gore in 2000; or
2. Bush had won less than 65.5% of such voters but had stolen away so many Gore voters (even over and above Nader voters who switched to Kerry) that he could approximate the same effect.
As more poll data comes in, I'm more convinced now by some of the commenters to the prior post who argued that it was more the latter than the former, and that the Gore voter switch is particularly pronounced when you consider the likelihood that most of Nader's voters from 2000 went over to Kerry. (I heard someone on TV claim that exit polls showed Bush won 10% of Gore voters). This is a conclusion that should cause ABC's The Note great embarrassment for its now-famous declaration, back on August 11, that "we still can't find a single American who voted for Al Gore in 2000 who is planning to vote for George Bush in 2004."
I calculated the 65.5% "marginal votes" figure by applying the following formula to the national popular vote:
((Bush 2004 votes) - (Bush 2000 votes))/(((Kerry 2004 votes) - (Gore 2000 votes)) + ((Bush 2004 votes) - (Bush 2000 votes)))
As noted, Bush won an additional 8.66 million Republican votes, whereas Kerry won something on the order of 4.56 million additional Democratic votes. I computed these figures by ignoring third-party candidates, figuring that people Kerry won over who had voted Nader last time are, in many ways, equivalent to bringing new people into the process, and by comparing the official FEC tabulations from 2000 and the latest running tallies so far. I would caution that the 2004 figures are still moving targets; returns are coming in daily. The 65.5% figure, for example, is down to 64.5% as of Friday, and may go up or down as more absentee and provisional ballots are tabulated in various states.
Anyway, I thought I'd take a state-by-state look to see where it was, precisely, that all of those 8.66 million new Bush voters came from. The numbers that follow were computed Friday, November 5, following the call of Iowa, the last contested state, for President Bush. It's a particularly interesting question for me, as a New York City Republican listening to my fellow New Yorkers rage at what they saw as the provincialism of the red-staters who gave Bush his victory (See here and here for examples): where was it that all these extra Bush votes came from? What state led the charge to Bush?
That's right, New York. The single largest percentage of marginal voters swinging to Bush came among the benighted, provincial, knuckle-draggin', Bible-thumpin', troglodytes of the Empire State itself. New York was one of only three states in the Union (along with Rhode Island and Alabama) to see an increase in Bush votes and a decrease in Kerry votes as compared to Gore, and the only one in which the decrease was significant. Bush gained nearly 400,000 additional votes in New York while Kerry lost more than 120,000 - a swing of nearly half a million votes. That swing, by the way, all but eliminated Gore's 540,000 advantage in the national popular vote all by itself. Before New Yorkers fume at Bush voters in the South and the Great Plains states they should look around at their neighbors and ask themselves how many of them have been strangely quiet about this election.
It wasn't just New York, of course; the fourth-largest marginal swing was New Jersey, and Bush won over 80% of the marginal votes in Connecticut. Can you say, "September 11"? And, come to think of it - when you combine those states with the nearly 1 million new Bush votes in Florida - there may have been another factor at work in 2000, much noted in the media at the time and much ignored in the media this time: Joe-mentum. Without the presence of the first Jew on a national ticket, Kerry may not have had the same oomph in states with a large Jewish population ("Where have you gone, Joe Lieberman, your party turns its lonely eyes to you . . . ") Of course, these are basically Democratic states, so Bush still didn't win them. But he won over a lot of people here in the past four years, and that showed in the final tallies.
I list the states in order of the percentage of the marginal vote won by Bush:
Read More »
A few thoughts:
*I had to tweak the formula a bit to adjust for the two states with declining turnout and Alaska (where I used Kerry as the numerator).
*California's turnout was probably down for at least two reasons: Bush had campaigned there in 2000, creating an impression of being in play, and ignored the state this time; and voters may have been a bit burned out after the 2003 governor's race. But I also suspect that the numbers will come up a bit when the final tallies crawl in. The declining California turnout compared to booming turnout in the dark-red states had a lot to do with Bush's national popular vote gains.
*Alaska's swing towards Kerry I would attribute mostly to voter anger at Frank and Lisa Murkowski (anger Republican voters felt safe venting, given how safe the state was for Bush).
*You can see that Bush lost ground in Ohio and New Hampshire and Colorado, but only enough to lose him New Hampshire.
*I have no idea what was up in Montana.
*Unsurprisingly, the real battleground states were closer to 50% on the marginal turnout scale.
*Bush winning 55% of the marginal turnout in Illinois seems particularly impressive; not only did he cede the state to Kerry, but Election Day saw juiced-up state Democrats and depressed state Republicans due to Barack Obama's landslide victory over Alan Keyes.
*Note that booming turnout in states like Texas, Florida and Georgia may also be a sign of population growth in those states.
« Close It
Posted by Baseball Crank at 11:18 PM | Politics 2004 | Poll Analysis | Comments (8) | TrackBack (1)
November 4, 2004
POLITICS: Marginal Votes For Bush
Here's something I think is really, really interesting, as long as you understand that the methodology isn't so much science as a rough way of measuring the impact of something that might be more accurately measured if you had accurate exit polls. Turnout was up across the country, such that Bush got more votes everywhere than he did in 2000, and Kerry got more votes everywhere than Gore did in 2000 (except California in each case, as far as I can tell, although there may be a bunch of absentee ballots yet to count).
((Bush 2004 votes) - (Bush 2000 votes))/(((Kerry 2004 votes) - (Gore 2000 votes)) + ((Bush 2004 votes) - (Bush 2000 votes)))
For these purposes, I ignored third-party candidates, since people Kerry won over who had voted Nader last time are, in many ways, equivalent to bringing new people into the process. Looking at the official FEC tabulations from 2000 and the latest tallies so far, I get the following:
When you put the numbers in that context, you see that Bush was actually hugely more successful at the margins in his combination of bringing new voters to the polls and convincing more people to switch to him than away from him. Remember that next time you hear that high turnout always and everywhere favors the Democrats.
Posted by Baseball Crank at 8:38 AM | Politics 2004 | Poll Analysis | Comments (23) | TrackBack (0)
POLITICS: Believe The Polls
By now you've heard a lot about how the partial exit polls that leaked out during the day on Election Day across the internet were skewed to an almost absurd pro-Kerry extent, and you've seen how pro-Democrat pollster John Zogby's final results were the same way just before the election (he projected more than 300 electoral votes for Kerry).
But the state-by-state polls actually weren't all that far off if you knew how to read them. Personally, I was relying on two reliable sources down the stretch: Daly Thoughts and RealClearPolitics, both of which came out with the same Election Day prediction of 296 electoral votes for Bush. Assuming that nothing overturns Bush's lead in Iowa, which looks like the last state not definitively called, Dales and RCP will have each gotten 49 of 50 states right, missing only Wisconsin, which Kerry held on to by the narrowest of margins.
In fact, RCP's national poll average showed a fairly steady lead for Bush throughout the fall, so anyone who put their faith in the RCP guys knew what was likely to happen. Media reports to the contrary were mostly based on cherry-picking pro-Kerry polls and/or on the assumption that new voter turnout would moot all the old polling models. Dales in particular should be explaining over the next few weeks why that was a bad idea (Kaus got in the best cheap shot yet: "Bush 51, Kerry 48: Pollster Ruy Teixeira demands that these raw numbers be weighted to reflect party I.D.!")
Mark Steyn often argues that liberal media bias is a Republican's best friend, as Election Day is the only time that Democrats are forced out of the self-serving illusions given them by the media and compelled to face reality. On this one, he seems to have been right; the evidence was there in the polls, but people who were reading Zogby and the various media outlets that trumpeted a late Kerry surge missed it. Glad I was reading guys who could tell me the score.
Posted by Baseball Crank at 8:19 AM | Politics 2004 | Poll Analysis | Comments (1) | TrackBack (0)
POLITICS: The Message
We'll see more from exit polls and the like, although one of the ironies of this election is that the exit polls were so wrong about the result, yet they will still be used to break out who voted for who and why. Makes you wonder.
Anyway, here's my best guess on the message of this election as it pertains to the issues (more later on the candidates and the campaigns):
1. The War on Terror: Polls regularly showed that people are split on the war in Iraq, with many Americans having misgivings on the reasons for going there and our progress in winning the war. Ultimately, nothing succeeds like success; I'm confident that in time, we'll have enough tangible progress to get more public acceptance.
But Democratic hopes that unease about the war would sink Bush turned out to be largely unfounded. Even if people weren't so sure they supported the Iraq war, it was clear throughout the campaign that they trusted Bush and his team to carry the broader war through to victory, or at least as far as they could get in four more years. To some people, that may sound irrational: if you don't trust Bush on Iraq, why trust him at all? But most people, I think, understand that the president knows more than they do about any particular foreign controversy; they are perfectly capable of doubting the Iraq war based on what they know, and yet resting comfortably with the more general sense that Bush has proven himself to be a guy who's not going to take potential threats sitting down.
One of the criticisms that has sometimes been made by Democrats is that Bush politicized the war. If they mean simply that Bush sought political profit from his leadership in wartime and his handling of the time of crisis after September 11, well, that's politics; do these people not remember Oklahoma City, or the 1944 election for that matter? But that's not it; what really rankles is not that Bush scored political points off of having handled some uncontroversial things well. What rankles is that Bush found electoral advantages in 2002 and 2004 from the Democrats' own differences of opinion with his policies. As if it was noble of Democrats to attack the president's policies at all turns in the harshest of possible terms and seek to undermine them in Congress, and yet somehow improper for the president to point out these differences to the American people and ask them to decide which side of these various controversies they trusted.
This is the great dilemma for Democrats. Democrats have a set of beliefs about domestic politics (more later on this), and many of them feel cheated in some sense that foreign policy swamped those issues in the campaign. But at the same time, a large segment of Democrats remain harshly critical of the president's foreign policies. A Tony Blair/Joe Lieberman-type Democrat who doesn't put daylight between himself and Republicans on foreign policy and national security issues would make it nearly impossible to politicize those issues and remove deep divisions in our politics. If Democrats are going to bemoan the prominence of national security in our politics, they need to decide: are they willing to go along with Republican policies and attitudes that are popular, at least in broad outline, with the public? If they are, the security issue can be neutralized. If not, then they will have to accept the natural consequences of their own ideas.
2. The Economy: Some Republicans will argue that the president's economic policies have been blessed by the electorate. I'm not sure I'd go that far. Polls seemed to indicate, again, a generally divided view, with Kerry sometimes having advantages on the economy. But it is clear that voters found Bush's economic management at least sufficiently unobjectionable that bread-and-butter issues didn't overwhelm the rest of his message, even in hard-hit places like Ohio and Michigan (Bush did better in Michigan than in 2004). And, of course, there's no question that Bush's fealty to his tax cut pledges helped him hold his base, and that - as in 2004 - a number of House and Senate races went Republican after being fought on economic issues.
3. Social Issues and the Courts: Here, I believe there is a mandate, if one that Republicans need to interpret carefully. Republicans up and down the ticket did exceptionally well with rural and other socially conservative voters, and Karl Rove's prediction that he could bring out millions of evangelical Christian voters who didn't vote in 2000 proved prophetic. Polls regularly showed that voters preferred Bush over Kerry in picking judges, and it's now already conventional wisdom that the same-sex marriage issue played disastrously for Democrats in the heartland. With the Senate now up to 55 Republicans, Bush will be amply justified in appointing conservative judges and in pushing to get through the appellate judges who are already stalled. If Bush is really devious, he could respond to the next Supreme Court vacancy by appointing Miguel Estrada and daring Democrats to complain about his lack of judicial experience after they spent years keeping him off the bench.
But the posture of the same-sex marriage issue should also serve as a reminder: America is a progressive country and a conservative country, and politicians forget one of the two parts of that formula at their peril. Progressive, in the sense that there is a broad, general acceptance of social change. People may fight about particular changes in our society and grumble and groan about the decay of everything, but at a fundamental level, the public is willing to accept that attitudes about race, gender roles, sexual behavior and the like do change over time, and the society changes accordingly. Certainly, efforts to use government to forcibly hold back such changes in attitude almost always result in political setbacks. Bill Bennett had this to say yesterday:
With all due respect to Bennett - much as I'm sure he and I agree on many values issues - that's not going to work. But if it's important to recognize the progressive nature of social change, it's at least equally important to recognize the conservative impulse as well: people who may be willing to be persuaded to change their minds about things - or who may give way in time to people with different opinions - may not be so enthused about court decisions that take away from the people the development of that process and tie it up in a constitutional straitjacket. In some cases, that straitjacket can actually reverse the direction of the progressive impulse (as any social change can be reversed over time if attitudes change); pro-lifers are optimistic that, if anything, the absolutism of pro-abortion groups like NARAL and their allies in the courts have succeeded in provoking a general trend towards more rather than less disapproval of abortion. If such a trend grows visibly over time, eventually there will not be popular support for candidates like Kerry who swear to appoint judges with a pro-Roe v. Wade litmus test. This election could wind up being seen in retrospect as such a turning point, as Bush (like Reagan) got a larger share of the popular vote than avowedly pro-abortion candidates like Kerry, Gore, Clinton, Dukakis and Mondale ever did.
People like Kevin Drum keep telling us that times are a-changing and eventually, issues that favor conservatives will go away. But this dichotomy will never go away, no matter what the particular issue. Liberals are forever trying to use the courts to short-cut or entirely avoid the process of persuading people on social issues, and that will continue to be a self-defeating tendency no matter what the specific issue at hand. As long as conservatives focus their energies on appointing judges who will leave most such issues in the hands of the people and don't try to make major social changes of their own before their time, social issues will remain a bulwark of conservatism.
November 3, 2004
POLITICS: 2004: The Morning After
I stayed up until Edwards spoke at 2:30 (after being announced as "the next vice president of the United States"), so I'm just too spent this morning to do the full what-it-all-means post, or even to fully absorb the meaning of Kerry's refusal so far to concede. My gut tells me that Kerry's refusal to call it last night was only fair, given the traumatic 2000 experience for his party and given how close this one was in the Electoral College, although it's rather sad to see the tradition of Election Night concession speeches fade away. But I would hope he buries the hatchet by the end of today; fishing for an extra 500 votes when you have a popular vote plurality at your back is one thing, but going to war for 146,000 votes is quite another, and with Bush having won a decisive majority of the national popular vote, I suspect the public would run out of patience for a fight that lasts more than another day or so. The Democrats never got closure on the last election because the leader of their party never looked them in the eye and said, "we lost fair and square, it's over" the way the loser of every election had before. Kerry surely must be able to appreciate, particularly with the passions that election and the war have stirred up, why it will be crucially important to the peace of the nation going forward to do that soon.
My feeling this morning is mostly one of overwhelming relief. We got through the election without a terrorist attack, meaning the last thing Al Qaeda might have been holding back something for has passed. Not that they are done, but there was no other reason to wait other than lack of capacity to strike. And the election went well. The Commander-in-Chief will stay at the helm, and we will have the opportunity to carry his strategy through for another four years. The Senate will be more Republican, as we steel for a likely Supreme Court battle and maybe several.
For historical perspective, not only has Bush won a majority of the popular vote for the first time since 1988, but his 51% of the vote is larger than any Democrat has won, other than FDR (who did it four times) and LBJ in 1964, since the Republican party ran its first national election in 1856 (Jimmy Carter in 1976 is the only other Democrat to muster a majority in that period, and then it was 50% in the wake of Watergate). The Republican party remains a majority party at the national level, having won popular majorities now 7 times to the Democrats' two since 1945. It is, of course, particularly satisfying, on an emotional level, to see Bush win a larger share of the vote than Clinton ever did.
On the coverage last night, I was flipping channels continuously. CBS was actually the fastest network to call states early, but only FOX and NBC called Ohio for Bush, and at last check nobody was willing to say 270; it's safe to say that some of the networks just couldn't quite bring themselves to call a winner until the other side had conceded defeat. I do think FOX had the best coverage, for two reasons. First, FOX had the best ticker, packing in useful information on popular vote totals along with the percentages and share reporting for all the major races. Some of the others left out the raw numbers. Second, FOX had the incomparable Michael Barone, whose encyclopedic understanding of every battleground state down to the precinct level gave FOX viewers a decisive informational advantage in digesting the returns from hotly contested states like Ohio, Wisconsin and Florida.
Furthest-out line of the night, besides some of Dan Rather's Ratherisms, had to be Joe Scarborough discussing why statewide and nationwide elected officials like hurricanes in Florida.
Anyway, I'm tired and I need to get back to work. I'll be back on my usual early-morning blogging schedule wrapping up the election the next two days, and then I'll be resuming baseball coverage next week. I'll also be taking down some of the election-related bells and whistles on the site over the next several days.
Posted by Baseball Crank at 9:43 AM | Politics 2004 | Poll Analysis | Comments (2) | TrackBack (0)
November 1, 2004
POLITICS: Strong Incumbents, Strong Challengers
Looking at the RealClearPolitics 3-way polling averages, 11 out of 12 have Bush with between 47 and 51% of the vote, and 8 of 12 have Kerry with between 47 and 49% of the vote. The latest Rasmussen tracking polls are consistent as well, showing Bush leading 47.9-47.1, 48.1-47.1, and 48.8-47.4 over the past three days (the most recent listed last). Which means, essentially, that we have both an incumbent and a challenger who have a fairly solid base of support entering the last two days of the campaign. I think most of us will agree that it is highly likely that Bush will poll at least 47% on Election Day, and equally highly likely that Kerry will poll at least 46% and probably at least 47% - thus, at least a decently close election remains likely, although we could still have a decisive popular-vote majority and/or an Electoral College landslide.
Recognizing the limits of historical analogies, what can we determine from this? I decided to take a look at the final election results for elections dating back to 1824, when they started keeping records of the popular vote. There have been 25 elections in that period in which an incumbent has stood for re-election; 16 have been re-elected, 9 have been voted out of office.
Strong Incumbents, Weak Incumbents
What's interesting - and, in fact, what shows the limitations of historical analogies - is how few incumbents have lost races without a complete collapse in their support. Besides Ford, the other four incumbents to lose since 1900 got completely abandoned at the polls: Carter in 1980 got 41%, Hoover in 1928 got 39.6%, Bush Sr. in 1992 got 37.4%, and Taft in 1912 got 23.2% and finished third. Besides Cleveland and van Buren, the other two 19th century incumbents to lose also showed weakly, in both cases against candidates who beat them in the popular vote four years earlier: Benjamin Harrison drew just 43% in his 1892 rematch with Cleveland, and John Quincy Adams drew just 43.6% in his 1828 rematch with Andrew Jackson.
The average margin of victory for successful incumbents? 54.9 to 41.1 overall and 54.9 to 40.9 since 1900. The average margin of victory for successful challengers? 49.5 to 41.2 overall, and 48.5 to 37.8 since 1900.
Strong Challengers, Weak Challengers
Strong Incumbents, Strong Challengers
If you look at matchups of a strong challenger against a strong incumbent, there's only two historical precedents, both of them bad for the incumbent: 1888 and 1976.
In other words: tomorrow, history leaves us on our own. It's our job to make it.
September 26, 2004
POLITICS: The Microbial Theory
Let's consider exactly how bad things look right now for John Kerry in the Electoral College, by looking over RealClearPolitics' state-by-state battleground poll averages. Bush, of course, starts with a historical advantage: he needs 269 electors to tie, 270 to win, and if he holds the 2000 "red states," he gets 278. On the RCP scoreboard, Bush gets 291 if you count the states where his average margin is at least 3 points over Kerry.
With Ohio drifting away from Kerry and Wisconsin looking firmly planted in the Bush camp, Kerry's hopes are now totally dependent upon wresting Florida from Bush, while holding on to big battlegrounds like Pennsylvania (Kerry by 1.7), Minnesota (tied), Oregon (Kerry +0.7), and New Jersey (Kerry +1.4) (Michigan, at Kerry +5 now looks fairly safe for Kerry barring another big shift in the dynamics of the race).
But, leaving aside the issue of Maine and possibly Colorado splitting their electoral votes, consider this outcome - even if Florida gets away from Bush, he could still win with the following states:
Read More »
That's 273 electors, with Bush needing only to hold one red state that's close (New Hampshire, at +1.7), pick up one blue state that's close (New Mexico, Kerry +0.5), one that's trending Bush (Iowa, Bush +3.6), and one that's not close (Wisconsin, Bush +6.5), and he wins even if Florida (Bush +3.4) slides away and he can't replace it with another big state.
None of which means the race is over. But it means that a strategy of targeting states is almost entirely doomed for Kerry (although he does need to keep defending states like New Mexico). Kerry needs to change the race's overall dynamic. The debates are all but his last chance to do that.
« Close It
September 17, 2004
POLITICS: 54-40, or Fight?
Bush leads 54-40 in a Gallup poll due out this morning, raising further questions about the sometimes wide variance in polling. Still, I'd be surprised if many presidential candidates have won after trailing by double digits in a Gallup poll as late as the middle of September. The electoral math is getting grim for Kerry; if Bush wins Florida and Ohio, it's very hard for Kerry to win, and Bush is looking stronger in Pennsylvania and Wisconsin, which would really lock down the electoral college.
But don't get cocky; Dales still sees a lot of states in play.
Posted by Baseball Crank at 7:03 AM | Politics 2004 | Poll Analysis | Comments (1) | TrackBack (0)
May 22, 2004
POLITICS: Careful About Those Battlegrounds
The Washington Post reminds us that trends can be moving in different ways in different states, changing which states are "battlegrounds." Another reason why, if you aren't already, you should be reading Gerry Dales' Electoral College Breakdown on a regular basis.
March 31, 2004
POLITICS: Poll Taxing
Via Andrew Sullivan, the early results show a strong pro-Bush swing in the battleground states following his initial TV ad campaign. Perhaps most notable is this result:
14. If John Kerry were to win the election in November, do you think your federal income taxes would go up, or not?
That's never a good omen for a presidential candidate.
February 5, 2004
POLITICS: Charting The Battleground States
Let's have some fun with numbers . . . as primary season winds down and we look ahead to the likely Bush-Kerry matchup, it's important to bear in mind a lesson that the 2000 election drove home: presidential elections are won and lost in the Electoral College. (Which is, among other things, why national polls are of limited usefulness; it's the individual states that matter). So I thought I'd look at which states are likely to be "in play."
There are two variables: how many electoral votes a state has to offer, and how likely it is that the state could go to either candidate. The first is a fixed number; we know it in advance. (Daily Kos, which has some of the best horse-race coverage around, has a great calculator that lets you compute the electoral numbers by coloring various states red and blue). For the second, a good starting point is the 2000 election results.
I decided to take a whack at combining the two. I started by dividing a state's electoral votes by the percentage point difference between Bush and Gore, but that gave too much weight to the larger states, so I settled on dividing the electoral votes by the percentage point difference squared. (For ease of comprehension, I multiplied the percentages by 10 - thus, a 12-point difference was rendered as 1.2 before squaring it, a six-point difference as .6). This isn't a scientific sample, just a way of quantifying what we already intuitively know. Here's my ranking of the most-hotly-contested states (Under "Margin," I listed a negative margin for states won by Gore):
Read More »
If you cut the level of significance off somewhat arbitrarily at a factor of 20, you get 18 battleground states, with Florida naturally the biggest. Some thoughts:
*You can see why Karl Rove should be smiling. The least competitive states include a bevy of states that form Bush's base, while four of the top five and six of the top nine are Gore states from four years ago. On top of that, number 11 is Gore's home state, where Kerry is likely to be less competitive, while Kerry hails from a region of the country where most of the states are already solidly in his column anyway. And Florida seems likely to be more Bush-friendly this time, given how easily Jeb cruised to re-election there in 2002 and the absence of Lieberman on the Democratic ticket. There's even hope that Bush may find a more favorable hearing in New York due to the focus on terrorism and in California if Arnold Schwarzenegger is still popular in November, but as you can see, New York in particular isn't high on anyone's priority list.
*Then again, there are countervailing worries for the Bush camp. Kerry seems well-situated to pick off neighboring New Hampshire. Bush's biggest worry has to be elector-rich Ohio; the state has suffered exceptional job losses and Ohio Republicans have had a particularly bad record on tax and spending issues of late. And several of the Gore states at the top of the list (Iowa, Minnesota, Wisconsin, Oregon) were fertile grounds for Nader in 2000; that won't help Bush this year unless there's a lingering animosity towards Kerry's vote in favor of the Iraq war. Nevada could also be in jeopardy for Bush; while he's kept a surprising number of his campaign promises, Nevadans are none too happy about Bush backtracking on hopes that he would oppose the unpopular Yucca Mountain nuclear dump.
In the end, the best source will be state-by-state polls. But those aren't always easy for us amateurs to come by, at least not ones that are reasonably fresh and all taken at the same point in time. For now, though, we know which states to look most closely at.
« Close It
Posted by Baseball Crank at 6:55 AM | Politics 2004 | Poll Analysis | Comments (1) | TrackBack (1)
July 22, 2003
POLITICS: California Polling
Bush is losing ground in the polls in California. This underlines two things:
1. As I've been saying for some time, Bush has a better shot of reviving in NY (where the war on terror is especially close to home) than in CA. I'll believe a Republican winning in California when I see it.
2. Bush has nothing to lose from a recall of Gray Davis, and much to gain; if things just fester in California, voters won't be itching to reward any incumbents.
On a related note, CalPundit (actually doing some California punditry in a break from his all-Niger routine) has a hilarious story of Democrats plotting to force a budget impasse in California for partisan advantage-- in front of an open mike.
Just imagine if Newt Gingrich got caught saying some of the stuff in this article.
February 22, 2003
POLITICS: California and New York
There are few hardier perennials in the world of conservative journals of opinion than the article assuring us that, really, this time, Republicans are gonna start winning in California. It's right up there with "any day now, African-American voters are gonna wake up and realize that the Democrats take them for granted!" (The Wall Street Journal's John Fund is a master at both of these genres). Hugh Hewitt had a recent species of this in the Weekly Standard: Barbara Boxer's a loony leftist! Bush is gonna win the state! Hey, Cubs fans find a way to have hope each spring, so I guess California Republicans can too. Me, I'll believe it when I see it.
Here in heavily Democractic New York, though, I think Bush really can make inroads in 2004. My reasoning is simple:
1. Upstate New York has traditionally been good territory for the GOP, and Republicans have also proven competitive in the suburbs and on Long Island. In short, if Bush can neutralize Democrats' huge advantages in the City (in 2000, he lost Manhattan by more than a million votes), he's definitely in the game.
2. Voters in New York City have proven their willingness to vote for Republicans -- albeit more liberal ones than Bush -- when they feel their physical safety is at stake. Hence, we've had Republican mayors for the past decade.
3. Nobody cares more about progress in the War on Terror than New Yorkers. We're the City With The Big Bullseye, and everybody knows it. We were the opening battleground of this war. If Bush can convince people that he has made real progress on ths front by the fall of 2004 -- no major domestic terror strikes, Saddam gone, perhaps a new regime in Iran, maybe Osama's head on a spike -- he can be very competitive in the City, and maybe win the state.
It will all turn on the war -- but then, if the war is seen as going badly, Bush will be packing his bags in 2004 rather than counting electoral votes anyway.
November 8, 2002
POLITICS: Handicapping the 2004 Senate Race: The GOP Seats
Let's pick up where we left off yesterday, with the early prognosis for the 2004 Senate races. Once again, you can check the 1998 results for these seats here. This will probably about conclude this week's bout of election-related blogorrhea.
1. RETIREMENTS: Just one, a prominent one in a state that could go either way (I haven't checked lately, but last I heard the governors' race was still unsettled): John McCain in Arizona. You have to count that as a possible Democratic pickup.
2. THE NEWCOMER: Frank Murkowski in Alaska is supposed to be up for re-election, but he was just elected governor, and his appointed replacement will stand for re-election in 2004. That's a guarantee that there will at least be a contested election, but Alaska has been GOP territory for some time, and the ANWR controversy has exacerbated that. The main risk may well be that if Murkowski is a disaster as the governor, voters might take it out on his hand-picked replacement.
3. SAFE SEATS IN SAFE STATES: Republicans have a bunch of these, guys who are well set in their seats and won handily last time - Don Nickles in Oklahoma, Richard Shelby in Alabama, Judd Gregg in New Hampshire, Mike Crapo in Idaho, Sam Brownback in Kansas, Robert Bennett in Utah. I may also put Ben Nighthorse Campbell of Colorado along with fellow former Democrat Shelby in this category.
4. SWING STATES: Republicans Charles Grassley in Iowa, Arlen Specter in Pennsylvania and George Voinovich in Ohio seem safe - Specter faced a still challenge after the Clarence Thomas hearings, but he's tacked far enough to the left that it's tough for Democrats to outflank him now - but all three are in states that will be heated battlegrounds for the presidential race, and any one of them could find himself facing a formidable opponent. Iowa, for example, has a popular Democratic governor, Tom Vilsack, who won re-election handily Tuesday and, with Tom Harkin in the other Senate seat, has nowhere else to go but challenge Grassley or run for the White House (although Vilsack is also sometimes mentioned as vice presidential material, the Democrats' trend in the last few elections has been to look for veep candidates who will generate some national buzz). Put Campbell in this category - Colorado has been a swing state in recent presidential years, although it went GOP in a big way on Tuesday - if not in the one above.
5. ENDANGERED SPECIES: Three GOP incumbents won with less than 53% in 1998. One, Kit Bond of Missouri, is in a swing state, but I suspect he will do fine even if he's facing Dick Gephardt. Righty Jim Bunning of Kentucky will probably be helped hugely by the presidential race - Bunning barely won re-election with 49.75% of the vote last time, but Kentucky's a pretty conservative state these days nationally.
That leaves us the most endangered Republican of all, Peter Fitzgerald of Illinois, who slid into office on the trail of slime left by outgoing Senator Carol Mosely Braun. With Illinois' Democratic governor and Clinton aide Rahm Emanuel both newly elected, I can't immediately think who will be the favorite to take on Fitzgerald (the way things are going these days, I'd have to guess Paul Simon). But the Democrats will push very hard for this seat and have to be favored to reclaim it.
THE PROGNOSIS: I'd rate the Democrats as having two good shots at pickups here - Illinois and Arizona - with maybe about three other races winding up contested. With the Democrats themselves defending about three or four hot races, they will need a great string of luck to pick up two seats and go to 51 - and that's assuming they hang on to Louisiana and South Dakota this month. In other words, if the GOP can squeeze out one more seat in the Louisiana race or from South Dakota in the event of a recount, the odds will get prohibitively high against Democrats recapturing the Senate before 2006.