clock menu more-arrow no yes mobile

Filed under:

perspective in team BABIP

i think this dance routine needs more jazz hands.

alt. caption: more like i-a-notta touch-a you, amirite?
i think this dance routine needs more jazz hands. alt. caption: more like i-a-notta touch-a you, amirite?

trying to find some notes of optimism as we look up at a suddenly hard-to-narrow 4-game gap between Milwaukee and St. Louis, i ran across this:

august team BABIP -

STL - .306 (14th in the league) [LD - 18.6% (22nd in league)]

RECORD: 7-4 (WP: .636)

MIL - .342 (2nd in the league) [LD - 23.6% (3rd in league)]

RECORD: 8-2 (WP: .800)

PIT - .322 (8th in the league) [LD - 23.5% (4th in league)]

RECORD: 2-9 (WP: .182)

july team BABIP -

STL - .287 (23rd) [21.5% LD (6th)]

RECORD: 13-13 (.500)

MIL - .291 (21st) [20.4% LD (12th)]

RECORD: 16-11 (.593)

PIT - .310 (10th) [24.3% (1st)]

RECORD: 13-13 (.500)

june team BABIP -

STL - .266 (25th) [20.2% (12th)]

RECORD: 11-15 (.423)

MIL - .295 (15th) [19.2% (18th)]

RECORD: 14-13 (.519)

PIT - .309 (9th) [21.0% (9th)]

RECORD: 16-11 (.593)

now, i know that BABIP is not sheer luck. i've included the team LD rates, in part to show that the team BABIP varies independent of the LD rate, with the Cardinals' BABIP actually getting better as the team LD rate gets worse in August. i am also not trying to account for other factors, like the strength of the defense of the teams played in those months, or the park factors for the stadiums in which the teams played.

i also know that nobody on the blog needs a reminder that milwaukee's current 14-for-16 run is unsustainable. however, it seems pretty convincing that milwaukee is unlikely to sustain a .342 BABIP for the rest of the month. in fact, only one team has sustained such a high BABIP for a full month this year: the tigers had a team .344 BABIP in june.

it's also heartening to me to look back at a feckless june and a stagnant july and see that the club was really struggling for reasons almost certainly related in part to luck.

n.b., those who will go on at length about BABIP and skill; while there is certainly some relationship between the two, BABIP is heavily affected by luck in small and even medium-sized samples. compare and contrast the cardinals' LD rates in june and august and BABIPs in june and august.

it shouldn't come as any surprise that the st. louis sports media still lacks a good sense of what makes a decent sample size. joe strauss thinks that we should all pay attention to corey patterson's splits against lefties this year (.342 wOBA; .301/.333/.446) in a tiny sample size (89 PA's) and a .377 BABIP. i suppose we should all ignore his career mark (wOBA: .280; .241/.273/.374) against lefties over 1099 PA's and a .289 BABIP. but this is all garden variety sports commenting nonsense from one of the usual suspects.

i was more disappointed to see bernie miklasz, whose work is uneven although sometimes enjoyable, come out this morning blasting people for criticizing his article on albert pujols individual splits against the brewers. i had glanced at the article and mostly ignored it, because he was trying to draw a meaningful conclusion from albert's performance thus far this season against the brewers. obviously, there is nothing predictive about what pujols has done over 11 games. what's more, the importance of what albert had done earlier in the season during what may well have been a hamstring injury slump would have little import for what he might do now.

naturally, karma being what it is, pujols went 4-4 that same night against the brewers.

instead of taking a moment to reconsider the validity of his earlier article on pujols, miklasz took on his critics, complaining that it wasn't his "job to offer foot massages to Pujols." now, aside from the fact that i will never get said image out of my head, it's certainly true that journalists should not handle the home team with kid gloves. and it may well be that he got lots of messages that said, "albert is the gratest ever! u nevar playd the game so shutup!!" to the extent that he was responded to that kind of comment, i guess i can partially understand. he should have been able to tune those responses out.

the rest of bernie's response was meaningless; he claimed to have been vindicated, that pujols' performance showed that "if he isn't hitting, the team isn't the same." which may just win some sort of sportswriter's award for self-evident syllogisms: when a baseball team's players hit, the team does better. tell us more! does the team "look good" when its starting pitchers give up few runs and go deep into games?

it would be unfortunate if we left it at that. the real issue is that sportswriters are perpetually guilty of looking at tiny sample sizes like this and trying to draw some kind of conclusion. bernie shows flashes of being able to understand the importance of these concepts and receptiveness to some advanced stats, but when he responds like this to criticism, i tend to lose some respect.

the underlying issue was not whether he should be doing anything to albert's feet, but whether he should be using valid sample sizes to draw meaningful conclusions. there's a lot of interesting things you could write about albert and what he's doing at the plate; looking a 40+ PAs is not the way to do it.