Midseason xBABIP Update: The Overperformers

by
July 12, 2016

Yesterday, I calculated every qualified hitters’ xBABIP and compared it to each of their BABIP marks to determine which hitters have most underperformed. Today will bring a look at the opposite side of the coin — those hitters whose BABIP marks most exceed their xBABIP marks. These aren’t necessarily sell high candidates, but you figure their batting averages should decline, perhaps significantly, the rest of the way.

Midseason xBABIP Overperformers
Name LD% True FB% True IFFB% Oppo% Hard% Spd BABIP xBABIP Diff
J.T. Realmuto 17.9% 26.0% 5.1% 25.5% 29.4% 4.2 0.370 0.300 0.070
Jonathan Villar 22.9% 15.6% 3.7% 27.9% 32.3% 5.7 0.410 0.343 0.067
Xander Bogaerts 17.4% 25.8% 5.7% 27.0% 31.3% 4.5 0.369 0.303 0.066
Brett Lawrie 20.4% 35.0% 6.2% 23.5% 28.8% 3.1 0.340 0.284 0.056
Jonathan Schoop 19.4% 32.2% 4.6% 24.0% 29.1% 3.2 0.348 0.293 0.055
Derek Dietrich 21.0% 33.3% 2.2% 24.2% 30.7% 3.0 0.363 0.309 0.054
Ian Desmond 22.9% 25.6% 1.1% 29.3% 32.1% 6.4 0.402 0.349 0.053
David Freese 19.2% 20.3% 0.0% 25.4% 35.6% 1.7 0.383 0.331 0.052
Manny Machado 20.4% 37.6% 6.8% 25.8% 36.9% 2.3 0.346 0.294 0.052

J.T. Realmuto is our new Jason Kendall, offering rare speed from the catcher position. But that .317 batting average is a complete fluke. Realmuto is hitting fewer line drives and far more pop-ups, while keeping the rest of his batted ball profile nearly identical to last year, and yet his BABIP has spiked from .285 to .370! The BABIP, and resulting average, is going to crash and burn, which is going to reduce his opportunities to swipe bases. He would be a perfect sell high, but I’m not confident that fantasy owners realize how valuable he actually is with those steals.

Jonathan Villar has always been a high BABIP guy, but a .410 mark?! C’MON! Even crazier is that he has done that despite posting the third highest IFFB% in baseball! Of course, that doesn’t mean a whole lot because his FB% is third lowest in baseball. His True IFFB% shows us that pop-ups haven’t actually been a problem. Hope he keeps hitting ground balls then! xBABIP does believe he has true high BABIP skills, but obviously, it’s not going to remain above .400 for much longer. It’s part of the reason he has already attempted 41 steals.

Boy have I whiffed on Xander Bogaerts! He’s actually posting a near identical BABIP to last year, yet like Realmuto, his LD% has declined, while his IFFB% has surged. He has even hit more fly balls (bad for BABIP), while getting back to his high Pull% ways before his change last year. It baffles me how he continues to hit for such a high average. xBABIP thinks he shouldn’t be much better than the league average. Combined with his unexpected speed, I think he makes for a perfect sell high guy, but hey, I have been completely wrong on him so far, so perhaps I’m not the guy to be offering Bogaerts advice.

Lots of fly balls and pop-ups have led to a…career high BABIP? Brett Lawrie is thanking his lucky stars. If his strikeout rate doesn’t improve, his second half batting average could get ugly.

Despite a horrific 13/70 BB/K ratio, Jonathan Schoop is in the midst of a breakout offensive performance. Of course, it’s all been driven by a career high BABIP, as he has maintained his power spike of last year. But he’s not doing anything to suggest he might sustain such a BABIP. Still, the power should continue to be there and even the .263 average Steamer project over the rest of the season is acceptable.

It’s been one heck of a rebound season for Ian Desmond and his wOBA actually sits at a career high. Part of that is due to a surge in HR/FB rate, and the other is thanks to that insane BABIP. He has never posted a mark above .336 in a season and although a career high LD% and career low IFFB% has helped, it’s hard to believe it’s sustainable. I don’t think he’s an automatic sell high, but couldn’t hurt to see what you could get. It would be hard to replace his combination of power and speed though.

Manny Machado still hasn’t cured his pop-up problem, and this year he’s hitting more fly balls than ever before. So what does that add up to? A career high BABIP of course! While he’s not going to hit .318 the rest of the way, he should still be good for a positive contribution in the category. The one real concern is that he has only attempted three steals and has been caught on each try. The lack of speed take a small bite out of his value.





 

Mike Podhorzer is the 2015 Fantasy Sports Writers Association Baseball Writer of the Year. He produces player projections using his own forecasting system and is the author of the eBook Projecting X 2.0: How to Forecast Baseball Player Performance, which teaches you how to project players yourself. His projections helped him win the inaugural 2013 Tout Wars mixed draft league. Follow Mike on Twitter @MikePodhorzer and contact him via email.

22 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
johnforthegiants
7 years ago

Is chamberlain’s formula valid for team xbabip? I ask because i just used it to calculate xbabip for the teams with the 8 best records in the nl and their average xbabip was .315, which seems way too high.

0
Mike Podhorzermember
7 years ago
Reply to  johnforthegiants

sure, it should work at the team level. the league BABIP is around .300, so it would make sense the best teams would average a higher mark. calculate it for the league and see if it’s close to actual league-wide BABIP.

0
johnforthegiants
7 years ago
Reply to  Mike Podhorzer

Don’t have the initiative at the moment, I’ll take your word for it for now. The Giants came out the best at xBABIP .326, seems too good to be true (actual BABIP is .302), but no one was under .308 (that was the Mets). All of them but the Marlins are underperforming, the Nationals, Dodgers, and Mets by 30-37 points. It didn’t seem right.

0
Mike Podhorzermember
7 years ago
Reply to  johnforthegiants

Out of curiosity, I decided to calculate the league average and you’re right! League xBABIP is .315, which is well off from the actual. Although the R-squared is only in the mid-0.40 range, you would expect individual hitters to be off, but the league as a whole to be pretty close. I’ll talk to Alex.

0
Alex Chamberlainmember
7 years ago
Reply to  Mike Podhorzer

Hey guys — Mike brought the issue to my attention. I think the model itself is fine, theoretically speaking. It’s the variation in hard-hit rate (Hard%) that’s the issue — it has steadily climbed over the years, to the point where it’s almost a full 10 percentage points higher than it was in 2002. Anywhere else you look (LD%, Oppo%, etc.), league trends have stayed almost perfectly constant.

I’ll make some tweaks and get back to Mike with the info. Rather than posting something new, I’ll likely update the original post (because Mike cites it so frequently) with some notes on the updates. I imagine the likely result of the tweaks will be a slightly lower coefficient for Hard% and a slightly higher constant term (and natural variation elsewhere). Won’t end up being wholesale changes — just enough to make this thing appropriately accommodate the modern hitting environment.

Hope that helps, all! Let me or Mike know if you have an questions.

0
johnforthegiants
7 years ago
Reply to  Alex Chamberlain

Please write an article about this when you do it. This kind of influences everything. Last week I got into an argument with a Red Sox fan about their hitters benefiting from outrageous BABIP luck (.336 at the time, and their peripherals didn’t look like anything special) and he came with a calculation that their xBABIP was .328, it turned out he’d made a mistake and in was .321 but I still felt stupid, but then when I did calculations I did today, everyone looked too high.

Two other points about xBABIP:
(1) As I wrote in response to yesterday’s article about underachievers, it seems to me that using OPPO% can be misleading in the sense that some hitters can hit so much to the opposite field that they can be effectively shifted against that way (Matt Duffy this year is an example of this). It might be better to use the smallest of Pull/Cent/Oppo&.
(2) Spd for individual years seems to be highly idiosyncratic. For example, this year for some reason Buster Posey has a Spd rating of 5.1 while Hunter Pence’s is 2.3. Obviously Pence is going to get way more infield hits than Posey. It would probably be more reasonable to use career Spd (I guess with some adjustment when people get older). Also, I don’t know if Spd considers if a batter hits left-handed or right-handed, and if it doesn’t, some adjustment should be made.

0
Alex Chamberlainmember
7 years ago
Reply to  johnforthegiants

Sure, I can write another post. It’ll likely be brief. I’ll try to get it up tomorrow.

One approach I took to using batted ball direction was calculating the variance between Pull%, Cent%, and Oppo%, where zero would be a perfect 1/3-1/3-1/3 split. It provided little additional explanatory power to the model (if any — I can’t recall in precise terms) and was a doozy to calculate every time. Meanwhile, given the limited nature of batted ball direction, using any direction works fairly well for the model. Also, while in theory it is correct that a hitter who hits oppo a ton could see irregular shifts, we just don’t really see it in practice.

Speed is idiosyncratic, yes. I’ll be the first to admit there are certainly more refined approaches to estimating player speed. But in the grand scheme of things, the difference between Posey’s 5.1 speed score this year and his 2.6 career rate is about 11 points of BABIP. It’s enough to make a difference, but it’s really not that big of a difference. The following numbers are for illustrative purposes only, but if Posey has a .315 BABIP and either a .310 xBABIP (with a 2.6 Spd) or .321 xBABIP (5.1 Spd), and we’re splitting hairs about his performance, then we’re losing sight of the fundamental purpose of the tool.

0
johnforthegiants
7 years ago
Reply to  Alex Chamberlain

Believe me it’s happening with duffy this year and it’s frustrating. The shift to the opposite field in the outfield which some teams do is hurting his average. Just look at his spray charts and you’ll see why.

0