March 18, 2015

BP Unfiltered

More On This Week's Play Index

by Sam Miller

As Effectively Wild listeners know, each email episode includes a statistical query that I do using Baseball-Reference's Play Index tool. Sometimes the results are easier to see than to read, like this week's, so here's a companion article about the search. The question was: Who, since 2000, has thrown the fewest pitches per start in a season?

One answer is Chris Jakubauskas, who threw 12 pitches in his only start in 2010. The 12th, to Lance Berkman, ended up on Jakubauskas' face (then all the way back to the backstop), and he missed two months recovering. It's a stupid-scary highlight video and you shouldn't watch it.

That's not the spirit of the query, though, so the better answer is Jose Lima, who threw fewer than 71 pitches per start in 12 goes in 2002. While one way to get your pitch count high is to allow a bunch of baserunners, a way to keep your pitch count low is to allow all the baserunners:

• First inning: 1.084 OPS
• Second inning: 1.008
• Third inning: .670
• Fourth inning: 1.002
• Fifth inning: 1.210

He didn't often make it out of the fifth inning.

This is all going somewhere, promise. So, Lima demonstrates something that is quickly pretty clear looking at these things: Low pitch counts are one of those things that are good and admirable but ultimately correlate to sucking. Sort of the inverse of the runners left on base thing. Announcers will never tire of knocking the team that leaves a lot of runners on base. The team that leaves the most runners on base usually wins. Similarly, announcers will praise the pitcher with a low pitch count, as they should. But the pitcher who throws the fewest pitches usually loses. (I'm not sure these examples are actually analogous.)

This is what that looks like, in bubble form:

(r = -.44)

The exceptions are easily explained. Justin Verlander's bubble is easy to spot, high and to the right--clearly, his horse reputation (and, perhaps, his still-decent FIP) earned him long leashes even as he allowed more runs than anybody on his staff. Michael Pineda's is easy, too, as he wasn't allowed to throw 100 pitches in a game until September (or 110 in a game at all) despite the AL's only sub-2 ERA. In both cases, they're exceptions that prove the rule: If Verlander had been better, he would have thrown even more pitches per start, quite possibly leading the league, as he did in 2010, 2011, and 2012. If Pineda had been worse, he probably would have thrown fewer. Kershaw, the other sub-2 ERA, might also have been further to the right except that baseball games end after nine innings. One suspects he had more than 97 pitches in him on Aug. 16th, but the game was over.

Which takes us to the exception that disproves the rule, and the destination I've been working toward: Greg Maddux. If you were to look at the 200 lowest Pitch/Start seasons since 2000, the list that Jose Lima "tops," you'd find mostly terrible ERAs, a couple young kids treated cautiously, and a lot of Greg Maddux. Five of the top 75 seasons belong to him, and in most he was quite good. (All five of those seasons are in the top 35 ERAs of the group.) And this distinction between Greg Maddux and everybody else who finds their way onto this list is best summed up thusly:

Pitcher	Season	P/ST	ERA
Brian Matusz	2011	78.8	10.69
Greg Maddux	2002	78.8	2.62

Each pitcher took 79 baseballs out to the mound with him per game. Maddux used those 79 baseballs to face nearly four more batters than Matusz. (Maddux threw nearly one fewer pitch for every two batters than the average MLB pitcher that season.) With those 79 baseballs, Maddux got five more outs than Matusz. And with those 79 baseballs, Matusz allowed three more runs than Maddux did.

Earlier in this article, I used the phrase "the exception that proves the rule." That's a sneakily tricky expression to use correctly, and when I use it I have to stop and think it through to make sure I did it right. Over the years, writers have periodically referred to Greg Maddux as the exception that proves the rule, because he was so often exceptional. But in most cases, I would argue he was just the exception, the exception that in most cases disproved a rule. Brian Matusz is the rule. Brian Matusz is the rule that proves the rule, and Maddux is the exception that you just can only stare at in wonder.

Sam Miller is an author of Baseball Prospectus. Follow @SamMillerBB
Click here to see Sam's other articles. You can contact Sam by clicking here