The shame of the Jays series was....

[Ohfan67]

I just feel that the smaller the sample, the less you are able to tell whether a low BABIP was due to weak contact or bad luck. I've seen plenty of individual games where a pitcher was quite dominant, got a lot of weak contact but didn't strike out a lot of guys. And, I've seen a lot of games where pitchers were getting creamed and then someone tries to chalk the result up to bad luck because the pitcher's BABIP was high that day.

IMO, those issues persist even with a decent sample size. BABIP is only really useful when combined with a lot of other statistics and even then it's usefulness is pretty limited IMO. But veve was curious so...

[Ohfan67]

I just feel that the smaller the sample, the less you are able to tell whether a low BABIP was due to weak contact or bad luck. I've seen plenty of individual games where a pitcher was quite dominant, got a lot of weak contact but didn't strike out a lot of guys. And, I've seen a lot of games where pitchers were getting creamed and then someone tries to chalk the result up to bad luck because the pitcher's BABIP was high that day.

Frobby: Your post encouraged me to read more about BABIP. It is apparently useless to evaluate on offense. A lineup's BABIP is not a significant predictor of winning percentage, for example. Pitchers' BABIP is, however, strongly negatively correlated with winning percentage. So an interesting aside is that BABIP is only really relevant from the pitcher's perspective. A little weird, but several studies using large data sets have produced the same result.

[LookitsPuck]

I'm not a big fan of BABIP when used to evaluate hitters unless we're comparing historical norms for the hitter relative to a period of current activity (longer than 3 days, though).