This is a rant I posted on BABIP in February. Today (April 8th) Baseball Prospectus published a new article with research claiming BABIP is extremely flawed. I have been saying this for years and it feels good to finally be vindicated. Scroll down to see the article.
I am a big fan of sabermetrics and use them in my baseball handicapping but one stat I can't stand is BABIP. BABIP stands for 'batting average on balls in play' and measures a hitters average on pitches that he makes contact with and puts in play.(BABIP is used for pitchers too but for this rant I will focus on hitters) The theory is that BABIP will generally be the same for all players and will regress to the league average. So say one year the league average for BABIP is .300 and a player finishes at .380 the thinking is that he got lucky and next year he will regress back to the mean.
To me this is complete junk. Better hitters will make more solid contact than worse hitters. This will mean better hitters will have a higher BABIP because when they make contact it's more likely the balls they hit won't be fielded for outs.
To me, the only way to use BABIP is with video review. Video review would be able to tell you whether a higher than average BABIP is due to skill or luck. If the video review shows the player hitting a lot of bloop singles off of weak contact that are dropping for hits because fielders cant reach them then yes his BABIP will likely regress to the mean. However if most of his hits are coming off of solid contact and line drives then his BABIP likely isn't a product of luck and won't regress.
One of the dumbest arguments I have ever seen a writer make was when Keith Law of ESPN was arguing that performance enhancing drugs might not have been the reason for Melky Cabrera's inflated numbers and quoted his higher than average BABIP as proof.* Are you kidding me?
*higher than average compared to the rest of the league as well as higher than average compared to Melky's career BABIP
MLB front offices use video review to compliment their statistical analysis but the average fan doesn't have the same ability and access. Accepting sabermetrics in general doesn't mean we have to blindly accept that every advanced stat out there is accurate. Next time you hear someone referencing BABIP without video review be skeptical.
UPDATE: APRIL 8th: The eggheads at Baseball Prospectus published an article confirming what I have believed for years. That BABIP is junk and makes no sense. It feels good to be vindicated!!