Lets Stop Using Pitcher vs. Batter Splits

This Article Originally Appeared on Friar Forecast

Randy Johnson retired yesterday. He was a good pitcher. Arguably the best of his generation.

I could cite a bunch of stats that demonstrate how great Johnson was. I just don’t see the point. We all know he was good, we can leave it at that.

That said, the urge to use stats to highlight Johnson’s greatness can be great.

In a twitter post, Tim Sullivan wrote : “How tough was Randy Johnson on left-handed hitters? Consider: Tony Gwynn went 1-for-12 against him with four strikeouts.”

Although a neat little factoid, that Gwynn went 1-for-12 with four strikeouts against Randy Johnson actually tells us very little about Johnson’s success against lefties.

Gwynn went 0-for-8 against Mark Wohlers , and 1-for-13 against Kent Bottenfield .

Over his career, Wohlers had significantly more success against righties than lefties. Bottenfield had approximately equivalent success against righties and lefties, but nobody would ever use him as an example of a pitcher incredibly tough on lefty batters.

Against lefties he surrendered a 0.776 OPS—a bit worse than league average.

That Johnson struck Gwynn out four times in 12 at-bats is also not significant.

Because Johnson was very good at forcing all batters to swing and miss, we would expect Gwynn to strike out more against him than against the average pitcher.

That said, sometimes, just due to small sample size, Gwynn struck out a lot against certain pitchers. In 15 plate appearances against Mike Madden , Gwynn struck out five times. Madden only struck out 5.5 batters per nine innings in his short career.

Randy Johnson, of course, was incredibly tough against left-handed batters. Lefties only managed a 0.571 OPS against Johnson during his career.

His SO/BB ratio against lefties was 4.12. Amazing stuff. Using Gwynn’s struggles against Johnson does little to show such domination.

My point here is not to call out Tim Sullivan. His factoid was almost certainly intended for amusement, and not serious analysis.

It's just that all too often, people do use samples as small as Gwynn’s 12 plate appearances against Johnson to back up larger claims.

I have found such information to be disseminated most often on radio and television broadcasts. I implore Dick Enberg, the Padres' new play-by-play guy, to fight the urge to provide viewers with pitcher-batter splits. The information is not meaningful.

Special thanks to Baseball Reference’s play index for facilitating my research for this short piece.