Monday, January 19, 2009

Hitters, Luck, BABIP, and the Nationals

There's some really interesting research being done looking on BABIP--batting average on balls in play (not to be confused with Bay Area Blacks in Philanthropy).

The basic skinny is that there's a lot of randomness involved in what happens when a ball is put in play. Fielding, park effects, weather, dumb luck... all these play a part in how often hit balls are converted into outs. Research shows that pitchers really have no repeatable ability to affect BABIP.

Hitters are different. While there's a lot outside their control that affects BABIP, they can affect it with "true" skill. This analysis by Chris Dutton and Peter Bendix at The Hardball Times finds that hitter’s eye, line drive percentage, speed score and pitches per plate appearance all correspond with a higher BABIP, while there's a negative correlation between BABIP and pitches per extra-base hit, fly ball/ground ball ratio, spray and contact rate. All this I think fits what you'd imagine, but Dutton and Bendix did the fancy math to confirm.

Once Dutton and Bendix were able to quantify the extent to which each of these stats influence BABIP, they were able to create a "luck-neutral expected BABIP" for every player in baseball with a certain number of plate appearances. Here are the Nationals players from most to least lucky in their 2008 BABIPs:

NAME BABIP (actual) x-BABIP x-BABIP dif
Felipe Lopez .324 .287 .037
Elijah Dukes .313 .291 .021
Ryan Zimmerman .306 .285 .020
Cristian Guzman .333 .315 .018
Jesus Flores .309 .293 .016
Ronnie Belliard .320 .317 .003
Lastings Milledge .287 .297 -.010
Jose Castillo .290 .313 -.022
Josh Willingham .273 .297 -.024
Willie Harris .268 .306 -.038
Austin Kearns .242 .284 -.042
Corey Patterson .210 .262 -.051

Because most people aren't used to looking at stats like BABIP, I took the Dutton and Bendix xBABIPs and crunched the numbers on what the Nationals' OBPs and BAs would have been in 2008. Again, from most to least lucky, here's what our guys would have done if luck hadn't been a factor in BABIP:

NAME Actual OBP xOBP Actual BA xBA
Felipe Lopez 0.343 0.315 0.283 0.252
Elijah Dukes 0.381 0.369 0.264 0.250
Ryan Zimmerman 0.333 0.317 0.268 0.218
Cristian Guzman 0.345 0.330 0.316 0.300
Jesus Flores 0.296 0.285 0.256 0.244
Ronnie Belliard 0.372 0.370 0.287 0.285
Lastings Milledge 0.330 0.337 0.268 0.276
Jose Castillo 0.292 0.309 0.246 0.264
Josh Willingham 0.364 0.379 0.254 0.272
Willie Harris 0.344 0.371 0.251 0.281
Austin Kearns 0.311 0.340 0.217 0.251
Corey Patterson 0.238 0.279 0.205 0.248

Just a few observations here:
  • As I've noted before, Kearns had a lot go against him in 2008, and he almost surely wasn't as bad as he looks at first glance. A .340 OBP would have been more respectable, but really the biggest problem for Kearns was his total lack of power. And a .340 OBP isn't all that good, especially for a corner OF who's strength is supposed to be his on-base skills. I don't think there's a ton of evidence for a big bouce-back for him, though probably there will be some.
  • As good as Willie Harris was in 2008, it's pretty stunning to consider that he really "deserved" a .371 OBP. I can't quite shake the feeling that maybe I'm missing the boat not calling for starting Willie in left in 2009, and this is another bit of evidence that he might be better than Willingham.
  • There's nothing here to offer encouragement on the Jesus Flores front. Next year will be a very big year for him. I'm still hopeful, but I wouldn't be at all surprised if we're looking ahead to Derek Norris or Adrian Nieto by this time next year.
  • Cristian Guzman was very, very lucky, and I really think we ought to have traded him to one of the contenders in need of a shortstop.

4 comments:

Anonymous said...

Biggest thing that jumps out at me is Zimmerman's xOBP and xBA are crap. It validates the thinking that he took a step (or two) back last year. Perhaps his injuries were a factor, but it really brings up the question of whether he really is the future. There's certainly no reason to sign him to a big, long term deal right now. People calling for that aren't looking at the numbers.

Steven said...

Max--Yeah, it's a concern. I'm not where you are, but sad to say for Nationals fans your position is not an unreasonable one.

I think between the injury and the fact that he's always been a slow starter--it just doesn't outweigh how very, very good he was in '06. And he's still very young.

But you're right that there's really no silver lining in the 08 numbers.

e poc said...

I don't think you're missing any boats with Willie. There's really nothing special there. His career wOBA is .303. Unless his 100 at-bats against lefties last year represents a real change in skill level rather than just a fluke, his offense will take a dip this year, especially if he plays every day. The defense is nice, but really only nice enough to make him a starter on a team that doesn't have other options. Here are our current outfield options and what CHONE sees them doing in 150 games at a corner:

Dukes: +11 offense, +2 defense, +13 total
Willingham: +8 offense, -8 defense, +0 overall
Milledge: +0 offense, +0 defense, +0 overall
Kearns: -3 offense, +7 defense, +4 overall
Harris: -10 offense, +14 defense, +4 overall

CHONE isn't the end-all, be-all of player evaluation, of course, but those projections seem pretty fair. As you can see, Harris isn't noticeably better than Kearns, Willingham, or Milledge. Milledge and Kearns are probably the best of the lot (aside from Dukes, of course), as they're the only ones who have much upside on both the offensive and defensive ends.

The frustrating part is that while neither Kearns, Willingham, or Harris has a part in the future of the team, they could all have decent trade value, but we don't have enough spots to show them off. If it were up to me, the outfield would be Dukes, Milledge, Kearns, left to right, and Willingham would play first to spell NJ every once in awhile (and then permanently when NJ gets injured). I'd use Willie as a super-sub and defensive replacement.

e poc said...

Steven: what's your take on the difference between the Nats and Zimmerman's arbitration figures? A remarkably big difference - what do you think it means?