Baseball Crank
Covering the Front and Back Pages of the Newspaper
June 15, 2005
BASEBALL: 30 Not Likely

David Pinto muses over whether Dontrelle Willis might have a shot at 30 wins. I looked back at the numbers, and this much is clear: Willis would have to do something totally unprecedented.

When Lefty Grove won 31 in 1931, he made 11 relief appearances in between his starts and threw 27 complete games. When Dizzy Dean won 30 in 1934, he made 17 relief appearances and 24 complete games. When Walter Johnson won 36 in 1913, he made 12 relief appearances and 29 complete games.

Those are the only three pitchers ever to win 30 games while starting fewer than 37 in a season. And of the 6 other times a pitcher has won 30 with between 37 and 39 starts, they've averaged 7 relief appearances and 33 complete games.

I just don't see a modern pitcher, making 35 or fewer starts and not relieving between starts, winning 30, especially without a lights-out closer or a wrecking crew offense, neither of which the Marlins have.

UPDATE: From Chris, in the comments, a Retrosheet breakdown of wins as a starter for the last four 20-game winners:

McLain '68 -- 31 as a starter, 0 as a reliever
Dean '34 -- 26 as a starter, 4 as a reliever
Grove '31 -- 27 as a starter, 4 as a reliever
Bagby '20 -- 25 as a starter, 6 as a reliever

Posted by Baseball Crank at 7:16 AM | Baseball 2005 | Comments (12) | TrackBack (0)
Comments

I think there will be another 30 game winner in our lifetime ... but he will win them all in relief. Elroy Face went 18-1 in a season where he made 57 relief appearances and pitched 93.3 innings for basically a .500 team. (He had 10 saves.) I could see a really effective, durable reliever getting 30 wins if he got into enough close games.

Posted by: Chris at June 15, 2005 10:22 AM

Now that they've yanked Leiter from the rotation, making it a 4-man rundown, will that help his chances (more starts) or hurt his chances (tired quicker)?

Willis is an energetic young man, so if anyone could be helped from a 4-man rotation, it might be him.

I'd love to just see a race. Like George Brett chasing .400 (full season), the McGuire/Sosa HR chase, etc. I think a 30-game win chase would be really fun to watch.

In any event, I think Willis might have a chance at 25 wins, which - in itself - is amazing.

Posted by: Garth at June 15, 2005 11:36 AM

Everything would have to go right for an entire season. Not likely. A batter hitting .400 is more likely. Wins are not completely controlled by a pitcher. If his offense doesn't score he can't win.

Posted by: LargeBill at June 15, 2005 12:37 PM

I should know how to look this up, but I don't. How did Denny McClain do it in 1968?

Posted by: keypusher at June 15, 2005 2:23 PM

41 starts (of which he completed 28).

I also don't think we'll see anyone win 20, let alone 30, in relief without a major change in reliever usage patterns. Modern (ie, post-Eck) closers rarely enter games without a lead, and thus rarely win as many as 7 or 8 games in a season.

Posted by: The Crank at June 15, 2005 2:38 PM

Right, Crank, the scenario I described would require a major change in reliever usage patterns, but I think that will happen sooner or later. Someone, at some point, is going to figure out that it is stupid to use a relief "ace" to close out 40-45 games where he enters with a 2-run lead in the 9th, and start using their relief stud when the game is tied in the 8th instead.

Posted by: Chris at June 15, 2005 3:12 PM

How many games in a given year does a team (A) have a tie in the last 2 innings and (B) wins the game?

Maybe 50-60 games out of the 162. (and the worse the teams are, the more that number goes down -- the Royals, for example, would probably only have 30 or so situations fitting that bill.) Now, out of those 50-60 opportunities, you would have your relief ace come in, and he would almost never yield a run.

It just seems to me that - at least for now - managers won't be clever enough to put their relief ace in while it's a tie; a team won't be good enough to blow everyone out, but will be good enough to win in the last few innings everytime; a pitcher would be able to pitch 1-2 innings everytime that situation comes up all year AND pitch consistantly well.

Posted by: Garth at June 15, 2005 5:26 PM

Bob Welch won 27 games in 1990 and had only 2 no-decisions, Clemems and Gooden won 24 each in 1986 and 1985 respectively and had 5 NDs apiece. Gooden had a 1.53 ERA. If those guys can't crack 30 it is almost impossible. It would require a level of perfection from a player and team that are pretty much unattainable. Willis should be psyched if he gets to 20.

I don't see a big change in relief use happening like that. Relievers don't pitch that many innings and managers, in general, are not going to put their best pitcher on the hill when the game is tied and they don't know how long he is going to have to go. A long-term healthy #1 guy in the pen is a rare thing. M. Rivera is, by the standards of the day, a freak in terms of quality and durability. Managers aren't going to risk a guy like that unless it is a must game and there are only a handful of those a year.

Posted by: jim at June 15, 2005 5:28 PM

Does anyone know how many relief wins Lefty Grove and the others had in their 30 win seasons? BaseballReference does not give that information.

Posted by: Steve at June 16, 2005 9:21 AM

http://www.retrosheet.org/boxesetc/VPHA01931.htm

Just compare the "winner" and "starter" columns. Of Grove's 31 wins, 27 came as a starter and four came as a reliever. I am not going to do it for everyone, but that's how you'd do it if you wanted to figure it out.

Posted by: Chris at June 16, 2005 9:36 AM

Okay, I'll do a few, but I'll let someone else tackle the deadball era guys.

McLain '68 -- 31 as a starter, 0 as a reliever
Dean '34 -- 26 as a starter, 4 as a reliever
Grove '31 -- 27 as a starter, 4 as a reliever
Bagby '20 -- 25 as a starter, 6 as a reliever

Maybe someday science will decide that the strain on a pitcher's arm is really an exponential function of the number of pitches he throws per start, and not a function of the number of starts. (The jury is still out on that question.) If that proved true, some club might go to a three-man rotation, give three guys 54 starts each, but never let them work past the 6th inning. You could win 30 if you got 54 starts. (Nobody has had that many starts since the 19th century, but that doesn't mean we will never see it.)

Posted by: Chris at June 16, 2005 10:00 AM

Looks like we can't do the same thing for the 30 game winners from the deadball era, if anyone wanted to try it. The Retrosheet game log only lists the starting pitchers for each game (and not the winning / losing pitchers) for 1917 and earlier.

Posted by: Chris at June 16, 2005 1:46 PM
Site Meter 250wde_2004WeblogAwards_BestSports.jpg