2014 xBABIP Values
This past season, I introduced a xBABIP equation which uses Inside Edge’s hard hit rate and player speed. Well, I have been a little slow on any updates. I have finally gotten around to getting all the 2014 values in one place.
When creating the xBABIP values, I found it correlated more to the next season’s BABIP then any other easily created formulas. Here are the 2014 values with some thoughts on some individual players.
• Highly shifted players (Chris Davis, Mike Moustakas, Brian McCann): I tried with no luck to add pull data to the equation. It didn’t make any difference in the final xBABIP values because just a few hitters are heavily shifted. The shift is definitely part of this trio’s struggles, so their actual BABIP should be less than their expected BABIP.
• Josh Harrison: He came out of no where to hit the ball hard. Additionally, he has a decent amount of speed to get some infield hits. The combination led to high BABIP and AVG in ’14, but the results look legit. His BABIP was only 10 points higher than his xBABIP, so he may not see as much regression as some people think in 2015.
• Mike Zunino: A BABIP around .300 would be huge for his value. His average would be closer to .250 which would be nice to go with his projected 20 homers.
• Lorenzo Cain: The 29-year-old may have had a career season in ’14 with his near .300 AVG supported by a .380 BABIP. His xBABIP was at .305, which is close to his 2012 and 2013 BABIPs which supported an AVG around .260. I could see him as a breakout/trendy pick for 2015 and therefore way over valued.
• J.D. Martinez: The 2014 power numbers may be legit, but his BABIP (.389) driven AVG (.315) looks to regress to near his career values (.333, .272) if he .326 xBABIP is to be believed.
Jeff, one of the authors of the fantasy baseball guide,The Process, writes for RotoGraphs, The Hardball Times, Rotowire, Baseball America, and BaseballHQ. He has been nominated for two SABR Analytics Research Award for Contemporary Analysis and won it in 2013 in tandem with Bill Petti. He has won four FSWA Awards including on for his Mining the News series. He's won Tout Wars three times, LABR twice, and got his first NFBC Main Event win in 2021. Follow him on Twitter @jeffwzimmerman.
Unluckiest (min. 500 PA)
C Davis
M Moustakas
C Carter
B McCann
C Crisp
J Bruce
K Davis
E Encarnacion
M Teixeira
C Santana
Luckiest (min. 500 PA)
A Eaton
C McGehee
C Johnson
J Abreu
C Yelich
Y Puig
J Altuve
JJ Hardy
A Beltre
K Suzuki
Is there any easy way to factor in distribution of hits and account for the babip-reducing effect of shifts?
I wouldn’t call your list of “unluckiest” actually the unluckiest. Of those names you listed, almost all of them are dead pull hitters, and get the shift put on them. Because they are hitting into the shift most of the time they get base hits taken away,lowering their xBABIP. I wouldn’t call those hits that were taken away “bad luck.” And so like the writer said, you’d expect these players to have a noticeably lower BABIP than xBABIP. I wouldn’t equate it to luck and I’d expect your entire list to be on the extremely “unlucky” side of things again this year.
Just because the formula used here (or any formula) does not explain it does not make it luck. (Even if the player cannot repeat the performance does not mean it was luck.)
Variance from a model or formula could be partially due to luck, but it is a lazy assumption that is wrong more than it is right.
Jeff,
The jd Martinez link actually goes to jd Martin’s page.
Sleeper Candidate
I got Johan Santa’ed
Click on the ‘ez’ for now
Getting Johan Santa’d is the official terminology for when that happens. Love it.
Having (sadly) watched a lot of Chris Johnson over the past 2 years, I feel like your xBABIP value for him is way off. The guy is going to run a high BABIP because he hits so many LDs and Texas Leaguers that fall just between the infield and OF. Perhaps you’re weighting hard hit balls too heavily? What about avoiding pop-ups?
What’s the R-squared on this? Couldn’t find any info on it in your original post (or this one), but I may have overlooked it.
I’d also suggest looking at individual BABIP components. A lot of the noise comes from it being an aggregate of different skills.
For example, McGehee’s 2014 BABIP is likely less sustainable because most of the increase was on ground balls hits to the outfield where outliers don’t normally repeat.
how close does josh harrison get to his steamer projection?
Not that this disqualifies the formula but I have a hard time seeing Puig as a low 300 babip type of hitter. To date he has only posted exceptionally high babip, and by the eye test he doesn’t seem to get very many cheap hits.
Does the inside edge data look at batted ball velocity by batted ball type or does it average velocity across all batted ball types? Puig’s ground ball velocities might be harder than the avg players while his avg total velocity might be lower overall because he hits fewer liners or flyballs. His ground balls might go for hits far more often than another player and he may not get the credit for it here.
He’s also very fast.
I get that Wright struggled last year but why on earth would his BABIP have dropped nearly 50 points?
Also, I will say that Granderson’s xBABIP is encouraging (he would have hit .242 with a .345 OBP) although I’m not sure if I buy it. He hasn’t had a full season with a BABIP that high since 2011.
Grandy falls under the ‘highly shifted players’ category
Anyway we can download this? Or sort it?
There is a download button on the visual
While this data is interesting, I feel that the only way to properly assess it is to put it in context. If this stat were included by Fangraphs for all of the seasons for which the necessary data is available, users could compare a player’s recent performance to his past performance to try to determine whether the xBABIP vs BABIP differentials relate to luck or they are specific to the player. For example, if Mark Teixeira consistenly has a positive differential like he did in 2014, then it’s probably because he’s a dead pull hitter and gets shifted on, not because he’s unlucky.
Do you have a formula for pitchers as well? Would you be willing to share that list and/or formulas for both hitters and pitchers? Finally, did you calculate how well your formula correlates Yr1 to Yr2? Thanks much…