Trying to Explain Vicente Padilla’s Platoon Split

I often find inspiration for my article in the words of my fellow writers, as their insightful observations give rise to further questions for me. Anyway, today is no different, as I am particularly intrigued by Dave Cameron’s observations about the components of a platoon split. The platoon split is clearly seen in fielding-independent measures like strikeout, walk, and ground-ball rate, but Dave C. found (via David Appelman) that there is only a slight platoon split for BABIP and HR/FB, a couple of points for the former and tenths of a percent for the latter.

Dave then questioned whether this general rule is the case for all pitchers, particularly if you have a pitcher who throws for a non-standard arm slot. He brought up a handful of guys. Since I had written about Vicente Padilla before and had his data laying around, I thought he would be an interesting test case.

Padilla has enormous xFIP split (5.11 versus LHBs and 3.83 versus RHBs) which comes from his K, BB and GB% platoon splits, but beyond that he also has a slight HR/FB split and a huge BABIP split (.324/.273). Going forward should we expect him to have just a slight difference in his BABIP against LHBs and HRBs like most pitchers, or is there something different about Padilla?

First off, Padilla does not have a great pitch against LHBs. He has a rarely thrown, ineffective changeup/splitter and even his okay curve shows a pretty big platoon split (while most curves do not). So against LHBs, he is left throwing his fastballs — two- and four-seam — 70% of the time to LHBs, and, of those, over 75% are his four-seam fastball. That four-seam fastball shows a big platoon split in BABIP, .280 against RHBs and .330 versus LHBs. Where does this come from? The BABIP of a pitch depends on a number of things, but the pitch’s horizontal location plays an important role. So I looked at that relationship for Padilla’s and the average RHP’s four-seam fastballs.

For most pitchers the BABIP of a four-seam fastball decreases the further inside the pitch is. Against RHBs Padilla has a similar trend, but generally his fastball have a much lower BABIP than average. Against LHBs on the outside his BABIP is similar, but on inside pitches, instead of dropping off, it picks back up. So it looks like, for some reason, LHBs can make solid, high-BABIP contact on Padilla’s inside fastballs. Whether this has to do with Padilla’s delivery or the movement on his fastball or whether the result is just noise is unclear. In all, I think Padilla’s problem against LHBs stems from not having a solid off-speed pitch and a four-seam fastball that — maybe because he cannot go inside against lefties — displays an out-sized platoon split compared to average.