Comments on: An army of Mark McGwires

By: Josh

Josh — Tue, 27 Oct 2009 15:39:45 +0000

You would think. More than anything else, this has highlighted for me how bad Oliver Perez’s 2009 was. Position players pitching (in the smaller sample size) allowed 7.70 BB/9 and 1.42 HR/9. Perez, in 2009, allowed 7.91 BB/9 and 1.64 HR/9. Perez obviously has an edge, in FIP terms, because he strikes guys out, but still… ouch.

By: Ted Berg

Ted Berg — Mon, 26 Oct 2009 23:12:18 +0000

That does sound more reasonable. Basically, we just need some figure to show that McGwire would suck hardcore as a pitcher, since he probably would.

By: Ted Berg

Ted Berg — Mon, 26 Oct 2009 23:10:24 +0000

Done.

By: Ted Berg

Ted Berg — Mon, 26 Oct 2009 23:10:12 +0000

It does feel like McGwire would have to be worse than Ollie’s 2009, but to McGwire’s credit, Ollie was absolutely abysmal this year and it was in an awfully small sample. You’d have to hope that, just by throwing the ball over the plate and letting the other McGwires play defense behind him, McGwire could pitch to that level.

By: Josh

Josh — Mon, 26 Oct 2009 22:35:09 +0000

OK, here’s my first tweak. If you restrict the data set further, to the 70 most recent players (to better reflect the current offense/defense balance in baseball; the previous data set went all the way back into the 1940s), you get a composite FIP of 7.07. That yields -0.128 wins above replacement per nine innings, which is -20.754 WAR pitching every inning of the whole season, or -4.15 WAR for each of the five pitchers (each of whom pitches nine innings every five games). That sounds more like it, doesn’t it? although now our sample size is only ~100 IP by position players over the past ~20 years.

By: Josh

Josh — Mon, 26 Oct 2009 22:12:51 +0000

Crap, I forgot to close my link tag. Any way to fix that?

By: Josh

Josh — Mon, 26 Oct 2009 22:12:15 +0000

Going on with this. I took a look at an explanation on Fangraphs for calculating WAR for a pitcher. I’m just going to stick with the 6.16 FIP I came up with, because park-normalizing it doesn’t make sense, and season-normalizing it is something I’m not sure how I would go about. League-average runs per game in 2009 was 4.61 per B-R.com.

Now, this hypothetical all-McGwire pitching staff (I’ll call it MP for short) doesn’t actually exist, so it has no effect on the win value of runs in 2009 MLB. By the estimation in the Fangraphs article, that value is (4.61 + 2) * 1.5 = 9.915 runs per win. MP’s FIP of 6.16 is 6.16 – 4.61 = 1.55 runs per inning below league average. 1.55 / 9.915 = 0.156. Subtracting that from .500 is .344. A replacement-level starting pitcher is .380, again according to the linked article, not that much worse than MP. Subtract .380 from .344, and you get -0.036 wins below replacement per nine innings.

Based on the conditions of this thought experiment, MP is effectively the only pitcher on this team of McGwires. In other words, MP is pitching every inning of every game. (Sometimes 8 innings rather than 9, but I’m just going with 9 for simplicity.) -0.036 per nine innings (per game, in effect) is -5.832 wins. If you divide MP into a five-man rotation, each of which pitches nine innings every fifth day, that’s -1.17 WAR each.

Is that right? Oliver Perez was -0.8 WAR this year, and if you were to extrapolate his performance out to the 291 innings that each fifth of MP pitches, that’s -3.62 WAR. Was Perez really significantly worse than a composite of position players? Well, MP’s FIP is 6.16 and Ollie’s was 6.40 in 2009, so I guess maybe he was.

Unfortunately, in the end this leads me to believe that the initial data set I used was probably not indicative of what Mark McGwire would actually do if you threw him out there on the mound. I can’t REALLY believe that a position player (pitching in all game situations, as opposed to just the limited set of game situations represented by the data set I used) would actually be a better pitcher than Oliver Perez, as crappy as his 2009 was. In lieu of anything better, though, I guess one could go ahead and use -5.832 WAR for the whole pitching staff of this team, at least as a jumping-off point.

By: Josh

Josh — Mon, 26 Oct 2009 20:49:57 +0000

Well there you go. You answered my question before I even finished asking it.

By: Josh

Josh — Mon, 26 Oct 2009 20:49:14 +0000

Yep, I found that much on Retrosheet (other than the last sentence about Baumholtz's recollection of the play), except that the Retrosheet box score says this happened in the first inning. What neither source provides any explanation of is why Eddie Stanky brought Musial in from the outfield to pitch to Baumholtz. The Cards and Cubs were both out of the playoff hunt at that point, but it's still a strange thing to do; was it just for sh@#s and giggles?

By: Ted Berg

Ted Berg — Mon, 26 Oct 2009 20:45:09 +0000

The Wikipedia claims it was a publicity stunt:

The only major league pitching appearance of Musial’s career occurred as a publicity stunt during the last Cardinals’ home game of the 1952 season.[88][89] Cardinals manager Eddie Stanky had a reluctant Musial pitch to Frank Baumholtz, the runner-up to Musial for the best batting average in the National League that season.[88] With Baumholtz batting right-handed for the first time in his career, Musial’s first pitch was hit so hard it ricocheted off the shin of third baseman Solly Hemus and into the left field corner.[88] The play was ruled an error, and Musial was embarrassed enough by his complicity in the gimmick to avoid pitching again for the remainder of his career.[88]