Questions on BABIP

[Robideaux]

Through perusing this site and other, my impression is that, in general, pitchers control walks, strikeouts and home runs and the likelihood of a batted ball in play resulting in a hit is pretty uniform and not controlled by the pitcher--thus the primacy of WHIP as a tool for evaluating effectiveness without luck clouding the picture.

 

If this is true, wouldn't it logically follow that the likelihood of a player getting a hit is purely a function of how often he puts the ball in play. Thus, high strikeout hitters would theoretically have lower batting averages than low strikeout hitters. A quick look at the BA leaders from 2006 show that there is definitely a negative correlation between BA and a strikeouts, but there are notable exceptions among the leaders. Jeter, Cabrera, Holliday, Ramirez, Dye, Berkman, and Howard are all among the leaders. 12 of the top 30 leaders struck out more than 100 times and 17 of 30 struck out more than 90 times.

 

It would seem like these high strikeout high BA players buck the uniform BABIP idea. I'm guessing my logic is probably messed up somewhere in there. If anyone could tell me where I strayed I would appreciate it.

 

Second point, even if BABIP remains relatively constant, is it possible that even taking home runs out of the equation, some players give up more doubles and triples proportionally to their total number of balls allowed in play? I don't see much mention of SLG% allowed--it seems like if you were able to combine SLG% with the park factor, it would be a good diagnostic tool.

 

Thanks., I hope this makes sense.

[sbrylski]

Through perusing this site and other, my impression is that, in general, pitchers control walks, strikeouts and home runs and the likelihood of a batted ball in play resulting in a hit is pretty uniform and not controlled by the pitcher--thus the primacy of WHIP as a tool for evaluating effectiveness without luck clouding the picture.

 

From studies I have read, the batter is about 2/3rd responsible for the outcome of a plate appearance (mainly K or BB).

 

And for smaller samples, WHIP isn't the greatest stat because BABIP noise shows up rather prominently. K/BB ratio would work better, or K/9 and BB/9.

 

This is why FIP and xFIP are such great stats. They calculate a pitchers expected ERA based on just Ks, BBs, and HR. It takes out all the luck of BABIP and percentage of runners left on base.

 

It would seem like these high strikeout high BA players buck the uniform BABIP idea. I'm guessing my logic is probably messed up somewhere in there. If anyone could tell me where I strayed I would appreciate it.

 

Its important to note that each type of hit (GB, LD, FB) have different likelyhood's of landing for a hit. A batter does have control over his GB/LD/FB ratios. To calculate a batters expected BABIP, an easy formula is LD% + .120.

 

For example, Joe Mauer has maintained about a 24% LD rate over the past two years and an expected BABIP of close to .360. That, coupled with a lower K rate and a lot of balls in play, leads to his execellent batting average.

 

Ryan Howard struck out 181 times last season, yet held a .313 batting average for two reasons. He had a very respectable LD% of 22%, and led the league in FB/HR ratio by 10%. FB result in outs most often of the three hit types, but they can't be caught when they land in the seats.

 

For pitchers its different. Because a batter has more influence on the outcome of a plate appearance, pitchers tend to have very similar expected BABIP.

 

Does that help, or did I miss the point of the question?

[Robideaux]

Thanks, that pretty much answers my questions.