Baseball Crank
Covering the Front and Back Pages of the Newspaper
November 04, 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).

The conventional wisdom was that increased turnout would help Democrats. If this were true, one would expect that, at least in contested states, the marginal voters would break for Kerry - i.e., that when you subtract out the 2000 returns from each side, what's left should lean Kerry. This would be true unless Bush moved so many Gore voters to his column (above and beyond the number of 2000 Bush voters who abandoned him) to negate the benefits of new Kerry-leaning voters. (My own suspicion is that, in general, the people who voted last time and switched sides were close to a wash, although they likely broke for Bush in some places like NJ where he lost decisively last time but closed the gap significantly).

But if you run the calculations of marginal votes, what you get is Bush majorities in the marginal numbers in a lot of places. In some states by big margins - in Connecticut, for example, Bush wins about 88% of the marginal vote. Ohio was an exception, but Bush takes 48% there, enough to hold a state he won by a few points last time. Of course, in New Hampshire, which he lost, he drops to 43%.

I'll run a state-by-state table of these later on when we're closer to having final tallies (including absentees) to provide a good comparison. But let's at least run the table on the national popular vote. Here's the equation I used:

((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:

Bush 2000Bush 2004Bush +Gore 2000Kerry 2004Kerry +Bush Share of Increase
50,456,00259,117,5238,661,52150,999,89755,557,5844,557,68765.5%

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 08:38 AM | Politics 2004 | Comments (23) | TrackBack (0)
Comments

Of course, isn't it possible that Bush convinced long-time voters to swing in his direction to such a degree that even the more liberal nature of the new voters wasn't enough to stem the tide?

Posted by: Jared at November 4, 2004 09:34 AM

I voted for Bush (on the Conservative Line) more out of fear of Kerry than being happy with Bush. It doesn't really matter in NY since it is a foredrawn conclusion that Kerry would carry the state. Bush has almost totally ignored the immigration issue and that could cause serious problems in years to come for the Republican party.

Posted by: Louis HR Muller at November 4, 2004 09:41 AM

Roberto at DynamoBuzz reported that New Jersey saw an increase of about 300,000 voters in 2004 over 2000. Almost all of those votes went for Bush.

Posted by: Eugene at November 4, 2004 09:53 AM

I suspect that alot of Bush voters in even in non-competitive states (both Red and Blue) were motivated by knowledge that the overall popular vote count does shape the environment in which legal challenges could be mounted against the results in particular states.

Gore's +500K in the overall popular vote in 2000 was a big factor in his ability to keep the Fla challenge going from a PR standpoint.

Posted by: Tom at November 4, 2004 10:35 AM

I suspect that alot of Bush voters, even in non-competitive states (both Red and Blue), were motivated by knowledge that the overall popular vote count shapes the environment for possible legal challenges against specific state results.

Gore's +500K in the overall popular vote in 2000 was a big factor in his ability to keep the Fla challenge going from a PR standpoint.

Posted by: Tom at November 4, 2004 10:37 AM

Moral values were an important factor in this campaign. (For anyone with a modicum of sanity the Democratic Party these days is scarry.) I compare what we saw to the surge of new movie-goers for _The Passion of the Christ_.

Posted by: Meta-jester at November 4, 2004 08:58 PM

Why wouldn't it be possible for these "new voters" to be "old voters" who have voted at various times in the past, but who simply didn't vote in the last election? Or voted in another state? I firmly believe that truly "new" -- as in "young" -- voters broke heavily for Kerry, until someone shows me some evidence that isn't the case. There are a large number of people who stayed home last time around, only to be motivated to go to the polls by the WOT, or Kerry's nuances, or the gay marriage ballot measures, or a zillion other things. Just a thought.

Posted by: daleb7 at November 4, 2004 10:52 PM

you've got to adjust for dems who voted bush, and GOPers who voted kerry.

this would reduce the new voter share for bush - marginally - since bush got more dems than kerry got GOPers.

Posted by: reliapundit at November 4, 2004 10:53 PM

"I firmly believe that truly "new" -- as in "young" -- voters broke heavily for Kerry,..." I can't find the link, but in a small college in New Hampshire, 90% of the faculty voted for Kerry, 68% of the students voted for Bush.

Posted by: ic at November 4, 2004 11:08 PM

It's actually better for Bush if you take the 2000 Nader voters minus the 2004 Nader voters and add them to Kerry. Nader had 2,882,955 votes in 2000 and only about 399,742 in 2004. It is probably reasonable to assume that about 2 Million of Kerry's 2004 votes came from this pool. I'm letting another 400K just drift away. That means Kerry's net gain is only 2.5 million vs Bush's 8.6 million. That runs at about 77%.

Of course Buchanan's 448,895 might have drifted to Bush but that still means a 2.5 to 8.2 or 75% ratio. Either way your point is well taken.

Posted by: OldManRick at November 4, 2004 11:10 PM

First time 30 year old voter. I voted for Bush, and I am not an evangelical. Look at the numbers for Minnesota and you find that Kerry won the state by little over 50%. And this is a state that voted for Mondale, for crissake.

Posted by: calebcharles at November 5, 2004 01:03 AM

Interestingly, almost all of Bush's gains came in urban areas, not the rural areas we think of as his base of support: here for details.

Posted by: coyote at November 5, 2004 03:10 AM

54 years old. Second time voter [2000 for Bush was first time]. Do not attend church except on very 'special occasions' [a memorial service or a wedding]. I'm still consider myself a Christian. And I am retired Navy. Security and defense were my issues.

My close friend is in her late 50's and this was first time I convinced her to go vote. She said she hadn't voted since the late 70's. Knew it should not make much of a difference in Louisiana as far as Bush carrying the state, but I wanted us to help add to the NATIONAL TOTAL. Was sick of hearing Gore won the popular vote. And was hoping to push David Vitter over 50% for the Senate. YES!!

NEW VOTERS:

If the demograhic part of the exit polls were anywhere near correct, 18-24 year olds were still just less than 10% of the vote. With overall increase in voting, that means the 18-24 increase was about 1.25 million voters. There was still [by my latest count] an incease of 13.84 million voters for Bush-Kerry over Bush-Gore. Means 12.6 million more 25+ year olds voted for Bush-Kerry. I'm sure some of those people had voted before, but they are 'new' voters from last election.

I split overlay of 2004 less 2000 by states Bush carried by less than 6%, states Kerry carried by less than 6%, and then states Bush and Kerry carried by 6% or more.

Looks like it was fairly even in battle ground states as far as new voters. Appears less democrats were motivated in the other states to bother taking time to vote for Kerry. I was highly motivated to be part of Popular Vote total.

Bush won by

Posted by: steve houpt at November 5, 2004 04:17 AM

54 years old. Second time voter [2000 for Bush was first time]. Do not attend church except on very 'special occasions' [a memorial service or a wedding]. I'm still consider myself a Christian. I am retired Navy. Security and defense were my issues.

My close friend is in her late 50's and this was first time I convinced her to go vote. She said she hadn't voted since the late 70's. Knew it should not make much of a difference in Louisiana as far as Bush carrying the state, but I wanted us to help add to the NATIONAL TOTAL. Was sick of hearing Gore won the popular vote. And was hoping to push David Vitter over 50% for the Senate. YES!!

NEW VOTERS:

If the demograhic part of the exit polls were anywhere near correct, 18-24 year olds were still just less than 10% of the vote. With overall increase in voting, that means the 18-24 increase was about 1.25 million voters. There was still [by my latest count] an incease of 13.84 million voters for Bush-Kerry over Bush-Gore. Means 12.6 million more 25+ year olds voted for Bush-Kerry. I'm sure some of those people had voted before, but they are 'new' voters from last election.

I split overlay of 2004 less 2000 by states Bush carried by less than 6%, states Kerry carried by less than 6%, and then states Bush and Kerry carried by 6% or more.

Looks like it was fairly even in battle ground states as far as new voters. Appears less democrats were motivated in the other states to bother taking time to vote for Kerry. I was highly motivated to be part of Popular Vote total.

Bush won by

Posted by: steve houpt at November 5, 2004 04:19 AM

Iím 54 years old. Second time voter [2000 for Bush was first time]. Do not attend church except on very 'special occasions' [a memorial service or a wedding]. I'm still consider myself a Christian. I am retired Navy. Security, defense and the way the Senate treated judicial nominees the last 4 years were my main issues.

My close friend is in her late 50's and this was first time I convinced her to go vote. She said she hadn't voted since the late 70's. Knew it should not make much of a difference in Louisiana as far as Bush carrying the state, but I wanted us to help add to the NATIONAL TOTAL. Was sick of hearing Gore won the popular vote. And was hoping to push David Vitter over 50% for the Senate. YES!!

NEW VOTERS:

If the demographics part of the exit polls were anywhere near correct, 18-24 year olds were still just less than 10% of the vote. With overall increase in voting, that means the 18-24 increase was about 1.25 million voters. There was still [by my latest count] an increase of 13.84 million voters for Bush-Kerry over Bush-Gore. Means 12.6 million more 25+ year olds voted for Bush-Kerry. I'm sure some of those people had voted before, but they are 'new' voters from last election.

I split overlay of 2004 less 2000 by states Bush carried by less than 6%, states Kerry carried by less than 6%, and then states Bush and Kerry carried by 6% or more.

Looks like it was fairly even in battle ground states as far as new voters. Appears less democrats were motivated in the other states to bother taking time to vote for Kerry. I was highly motivated to be part of Popular Vote total.

Bush won by Bush +1.92M - Kerry +1.57M [Bush 55%]
Bush overall increase in these states from 2000 was 26.1%.

Kerry won by Bush +1.47M - Kerry +1.46M [Bush 50.1%]
Bush increase in these states from 2000 was 19.4%.

Bush won by 6% or more [25 states]
Bush +4.27M - Kerry +1.84M [Bush 69.9%]
Bush increase in these states from 2000 was 20.3%.

Kerry won by 6% or more [13 states + DC]
Bush +1.30M - Kerry +9,509 [Bush 99.3%]
Bush increase from 2000 was 9.0%.

Take NJ, WA, CA, NY & RI away [I'll explain]

8 states + DC [less above states] Kerry won by +6%
Then Bush +855K - Kerry + 653K [Bush 56.7%]
Bush increase in these states from 2000 was 17.2%.

Why I dropped those 5 states to compare.

1. New Jersey - 230,978 more votes in 2004. Bush was +303,321, Kerry +10,470 [Nader and others minus 82,813]. Obviously there were voters in NJ that switched from Gore to Bush. My guess, the 9/11 effect was close to home.

2. New York - 65,037 more votes in 2004. Bush was +379,083, Kerry minus 136,638 [Nader and others minus 177,408]. More Gore to Bush. More 9/11 VERY close to home.

3. Rhode Island - 7,304 more votes. Bush +30,790, Kerry minus 2,437, Nader and others minus 21,049. More Gore to Bush ?? 9/11 effect.

4. California - Baseball crank mentioned above. Currently showing 1 million less votes in 2004. ?? Absentee ballots? Bush minus 164K, Kerry minus 434K, Nader and others minus 407K.

5. Washington state - Based on CNN data, I show 283K less votes in 2004. Both candidates less votes. ?? Absentee ballots?

NOTE: Bush took a beating in Vermont [Howard Dean ??] Bush +935, Kerry +34,599. Nader and others minus 16,952.

Posted by: steve houpt at November 5, 2004 04:23 AM

Iím 54 years old. Second time voter [2000 for Bush was first time]. Do not attend church except on very 'special occasions' [a memorial service or a wedding]. I'm still consider myself a Christian. I am retired Navy. Security, defense and the way the Senate treated judicial nominees the last 4 years were my main issues.

My close friend is in her late 50's and this was first time I convinced her to go vote. She said she hadn't voted since the late 70's. Knew it should not make much of a difference in Louisiana as far as Bush carrying the state, but I wanted us to help add to the NATIONAL TOTAL. Was sick of hearing Gore won the popular vote. And was hoping to push David Vitter over 50% for the Senate. YES!!

NEW VOTERS:

If the demographics part of the exit polls were anywhere near correct, 18-24 year olds were still just less than 10% of the vote. With overall increase in voting, that means the 18-24 increase was about 1.25 million voters. There was still [by my latest count] an increase of 13.84 million voters for Bush-Kerry over Bush-Gore. Means 12.6 million more 25+ year olds voted for Bush-Kerry. I'm sure some of those people had voted before, but they are 'new' voters from last election.

Have some data on 2004 election. Will post separately.

Posted by: steve houpt at November 5, 2004 04:27 AM

Iím 54 years old. Second time voter [2000 for Bush was first time]. Do not attend church except on very 'special occasions' [a memorial service or a wedding]. I'm still consider myself a Christian. I am retired Navy. Security, defense and the way the Senate treated judicial nominees the last 4 years were my main issues.

My close friend is in her late 50's and this was first time I convinced her to go vote. She said she hadn't voted since the late 70's. Knew it should not make much of a difference in Louisiana as far as Bush carrying the state, but I wanted us to help add to the NATIONAL TOTAL. Was sick of hearing Gore won the popular vote. And was hoping to push David Vitter over 50% for the Senate. YES!!

NEW VOTERS:

If the demographics part of the exit polls were anywhere near correct, 18-24 year olds were still just less than 10% of the vote. With overall increase in voting, that means the 18-24 increase was about 1.25 million voters. There was still [by my latest count] an increase of 13.84 million voters for Bush-Kerry over Bush-Gore. Means 12.6 million more 25+ year olds voted for Bush-Kerry. I'm sure some of those people had voted before, but they are 'new' voters from last election.

Have some data on 2004 election. Will post separately.

Posted by: steve houpt at November 5, 2004 04:28 AM

Sorry for multiple posts.

The 4:23 AM post looks complete and correct. I was getting website not responding. Tried a few times and then was trying to split it up.

Posted by: steve houpt at November 5, 2004 04:34 AM

If the exit polls are reliable, first-time voters broke for Kerry but only marginally - 53%-46%

See http://jimmysweblog.net/2004/11/quick-analysis-of-2004-exit-polls.html which is based on http://www.cnn.com/ELECTION/2004/pages/results/states/US/P/00/epolls.0.html.

Posted by: Jimmy at November 5, 2004 06:20 AM

The exit polls are not reliable.

Isn't that obvious?

Kerry supporters intimidated or just annoyed Bush voters enough for them to not be vocal at the polls.

You can rely on the real fact that votes went up a lot, and Bush did a lot better than Clinton or his father ever did. And that other races also favored Republicans in epic proportion.

We are not a divided country. Or wed are at least less divided than we were under Clinton, Carter, Kennedy, as the GOP is far more successful now than either party has been since FDR.

The only people who think we are divided are those who really want us to be so that their urban pockets of fear party is on even keel.

Sorry, in a democracy, the winner gets to decide everything. And we won everything more than we won everything last time. A huge turn has occured since 1994, and it has only increased and increased in intensity.

Posted by: Dustin at November 5, 2004 10:58 AM

The flaw in the logic here is there is no way to measure how many people voted for Gore in 2000 and Bush this time. I've talked to five or six people over the last year who voted for Gore and came to regret it (myself included) and because of our support of the war, planned to (and I presume did, as I did) vote for Bush this time. I don't think there are an insubstantial number of marginally conservative people like myself who thought Gore the better choice in 2000 but either preferred Bush now or distrusted and disliked Kerry and thought him far too liberal to support.

Posted by: Howard Owens at November 5, 2004 12:27 PM

Scratch out that old rule of thumb the pundits were hawking on election day, the one that says increased turnout is bad for the incumbent.

How about retiring this one too: In the runup to the election the poll numbers were high for those,

"Who were not happy with the direction the country was going in."

I must confess that I was with the MSM bias on this one and interpreted it as a strike against Bush. Then I heard a woman call in on CSPAN who said she's not happy about the porno-ization of TV culture for her kids. Then someone else called in and said the direction of the country was wrong because they haven't built a domestic oil refinery in 20 years. So "not happy with the direction" doesn't necessarily mean dump the incumbent.

Posted by: JLP at November 5, 2004 12:49 PM

Or it is just the effectiveness of secret software produced by Diebold and Sequoia? Of course the faith-based will always believe that black box voting is faultless.

Posted by: buma at November 5, 2004 07:24 PM
Post a comment









Remember personal info?






Site Meter