Batted Ball Profiles and what they tell us-- Part 1
by PaulThomas on Dec 26, 2007 2:27 PM PST
37 comments
A fool with a little information is a dangerous thing...
Lately I've been reading up on batted ball profiles, and what they mean for batting averages and other hitting statistics. I thought I'd take a look at the A's roster this year, and see what we can learn from what kinds of balls they hit.
The usual disclaimers: I'm throwing out anyone with under 80 plate appearances. Why 80? Because I'm writing this, and I said so. Also, I'm pitching the guys who've been cut or traded, since (in the inimitable words of Mark McGwire) we're not here to talk about the past. I'm leaving Stewart in, however, as there is a (remote) chance that he might be re-signed.
This leaves us with 13 total guys with enough plate appearances that are still on the MLB roster. To wit:
Barton
Buck
Chavez
Crosby
Cust
Ellis
Hannahan
Johnson
Kotsay
Murphy
Stewart
Swisher
Suzuki
I'll handle 4 today, 5 tomorrow and 4 Friday.
Statistical jargon
Basic breakdown: What kinds of balls are guys hitting? LD is Line Drive percentage. (Very, very good.) GB is ground ball percentage. (Bad, unless you're Ichiro or Maurice Greene.) OFB is outfield fly ball percentage. (Usually about average, but very bad if you're Juan Pierre or Reggie Willits.) IFB is infield fly ball percentage. (Very, very bad. These are almost always pop-ups or foul-outs.)
Results: What happens to the balls when they land? HR/FB is the percentage of fly balls which are home runs. (Most non-homer fly balls end up being outs, so this tells you a lot about whether a guy is benefiting or hurting his team by hitting flies.) BABIP is Batting Average on Balls in Play. (It's a key component of a player's batting average.) OCSlg is Slugging Percentage On Contact. (When a guy actually hits a ball, how hard does he hit it? If anyone's wondering, the difference between "In Play" and "On Contact" is that the latter includes home runs and the former doesn't.)
Forward!
Daric Barton
LD 24.2%
GB 32.3%
OFB 37.1%
IFB 6.4%
HR/FB 14.8%
BABIP .368
OCSlg .754
Barton had one of the strongest big-league debuts of any player in recent memory, as he came up and simply knocked the cover off the ball for three weeks at the end of the season. He showed everything one could want from a hitter-- great power, patience and plate awareness, good contact and the ability to beat out tough balls in the infield (2 infield hits).
Needless to say, everyone who is thinking clearly has to expect a regression from these numbers. It may be that Barton is a 24% line drive hitter, but those are fairly rare and we can't expect it on a permanent basis. He's primarily a fly-ball hitter, so he'll need to hold up either his line drives or his power indices (like HR/FB) to post above-average numbers.
Boy, though-- you can't help but salivate at the thought of a guy who hits a liner one out of every four times he puts the ball in play. By their very nature, liners are hard for defenders to handle-- they come in very fast at awkward trajectories. The really exciting thing about Barton is that you can see two separate ways for him to become a star-- a line-drive machine or, if he builds his power to what he was showing down the stretch, a more traditional first baseman model. If he manages both, he's a Hall of Fame candidate.
Travis Buck
LD 18.7%
GB 43.3%
OFB 33.8%
IFB 3.2%
HR/FB 8.6%
BABIP .354
OCSlg .619
Buck's numbers are sitting comfortably around league average in most respects. He hits a good but not great number of line drives. His raw power (as measured by HR/FB) is definitely below league average, but in the neighborhood of guys like Mike Lowell, Eric Byrnes and Kevin Millar-- all solid major league hitters who typically hit home runs in the high teens or low 20s over a full season.
Buck's strong suit is his BABIP. He hits a high percentage of ground balls, but has the raw speed to beat out a decent number of them (five infield hits this season). A rough rule of thumb for BABIP is line drive percentage plus .120, which would give him an "expected" BABIP of .307. We can expect to see some regression to the mean here next season, but Buck should continue to post BABIPs comfortably above league average.
In the future, A's fans can hope for two things. First, if Buck adds additional power (quite a possibility given his youth) he could substantially up his slugging percentage. Second, if he can add the push-bunt to his hitting arsenal, his excellent home-to-first speed should enable it to be an effective means of obtaining additional base hits.
Eric Chavez
LD 17.1%
GB 36.4%
OFB 35.1%
IFB 11.4%
HR/FB 12.0%
BABIP .268
OCSlg .573
Dear lord, that's a fuggly infield fly percentage. To a pretty large extent, those of you who complained about Chavez's lack of clutch-ness last year were quite right. He had a disturbing tendency to pop up in situations where a popup was the worst possible outcome.
That said, Chavez has never had anything like that kind of a popup percentage before this season. In past years it was in the neighborhood of 6%. If-- big if-- he returns healthy from the various surgeries he had this offseason, I suspect his popup percentage will drop toward what it has been for most of his career.
Chavez has good raw power, but he made very poor contact with the ball last season. This is reflected in his poor BABIP and OCSlg, which are below average for everyone in the former case and way below average for power hitters in the latter. A's fans need to hope that a newly pain-free Chavez is able to get around on his swings and square up more balls in 2008. If not, we'd better get used to biting our nails when he comes up with men on 2nd and 3rd and one out.
Bobby Crosby
LD 19.8%
GB 48.1%
OFB 26.4%
IFB 5.7%
HR/FB 8.8%
BABIP .254
OCSlg .415
Crosby: now starring as the Soviet Union! (A riddle, wrapped inside an enigma, etc etc.)
I'm confounded, I really am. Crosby's batted-ball numbers look like Derek Jeter's. And yet he has a BABIP of .254. His OCSlg is absolutely putrid-- despite his considerable power, when he makes contact with the ball, he looks like Juan Pierre.
If I was a hitting coach, I would tell Crosby to forget about trying to hit fly balls. Yes, he has power, but it's useless as long as he's making such poor contact with the average ball. The guy has good speed; if he slapped the ball around a little he might up his BABIP considerably. His best season (2005) was also the season in which he hit the most grounders and the fewest flies.
Really, though, I'm stumped. How can someone have a BABIP which is so close (only 6.5 percent or so off) to his line drive rate? If anything, given his solid speed, his BABIP should be HIGHER than the standard LD rate plus .120. Is he hitting a whole bunch of really soft liners right at people? Doesn't he ever get hits on non-homer fly balls? Maybe he's just unlucky, and Beane has stuck with him for as long as he has because he believes that sooner or later, his BABIP has to come up to match his batted ball profile.
That's all for today. I hope this was somewhat interesting; if anyone has any suggestions, I'd be more than happy to incorporate them (different stats you'd like to see, different ways of organizing them, whatever). Does my analysis make sense? Any ways you feel the team should make use of this information? Inquiring minds want to know.
More tomorrow.
0 recs |
37 comments
Email
Print
Comments
Crosby's problem...
is that he hits too many ground balls. Look at your numbers! Half of the balls he hits are ground balls. And almost all of them are easily fieldable because those are the ones he rolls over on. The pitches that we complain about, ie: low and outside, with Crosby standing three feet outside the batters box.
by Zonis on Dec 26, 2007 2:26 PM PST reply actions 0 recs
But perhaps
this is a consequence of hitting grounders with fly-ball swings. If you hit the top of the ball with an uppercut swing, you're likely to beat the ball into the dirt for an easy out. If you're swinging through a lower, flatter plane, that same ball might shoot the hole and get you a base hit.
I mean, it's almost impossible to hit a low and away pitch for a fly ball; the best you can hope for is to rap it the other way. Maybe it gets through, maybe it doesn't, but at least you have a chance of getting a hit.
by PaulThomas on Dec 26, 2007 3:07 PM PST up reply actions 0 recs
I blame the manager, whatever signal that Bobby
thinks is, roll over and weakly ground out to short, the A's need to change it.
by theblackpearl on Dec 26, 2007 3:13 PM PST up reply actions 0 recs
I don't remember Geren
being quoted on the subject but Macha constantly harped on Crosby's HR swing, and how they tried to get him to cut down on his swing but he wouldn't listen. I always felt the best thing a manager could do with Crosby is to sit him on the bench until he changes his approach but since he is seemingly injured all the time it never happened although it should have.
by china bob on Dec 26, 2007 4:12 PM PST up reply actions 0 recs
Wow
Did you really just say that it's almost impossible to hit a fly ball on a low and away pitch? Umm, ok (laughing) - carry on.
by NewPosterMojo on Dec 27, 2007 5:13 AM PST up reply actions 0 recs
Don't ground balls have a higher BABIP
than fly balls? The problem is that they typically result in either singles or outs. Hence the crappy OCSlg. If it helps, Crosby's PrOPS last year (.732) was nothing like Jeter's (.771), but more like Juan Uribe or Jason Bartlett.
by WaddellCanseco on Dec 26, 2007 3:09 PM PST reply actions 0 recs
Not familiar with PrOPS
Do you know the formula? (Not necessarily the exact mathematical details, just the inputs and what it's supposed to measure.)
by PaulThomas on Dec 26, 2007 3:21 PM PST up reply actions 0 recs
PrOPS stands for "Predicted OPS."
The Hardball Times says:
It was developed by J.C. Bradbury amd introduced in this article. PrOPS isn't really a new stat; it's a formula for predicting what a player's OPS is likely to be in the future based on his batted balls, strikeouts, home runs and walks.
It's too new to have been proven yet, but I think a discussion of batted ball types might benefit from mentioning it.
by WaddellCanseco on Dec 27, 2007 9:08 AM PST up reply actions 0 recs
OK that was a badly worded post.
When THT says it's not a "new stat", they mean it's not a stat, not that it's been around a long time.
When I said "it's too new" I meant it hasn't been around long enough to see how well it correlates with actual OPS.
by WaddellCanseco on Dec 27, 2007 9:14 AM PST up reply actions 0 recs
Good job
I really enjoyed the piece. I'm anxious to read about Ellis, and see what his stats say about him. Surely Cust must hit the most fly balls, except when he is striking out of course. Bill James wrote that Cust swings and misses more than any player in the league. Not that we didn't know that. I'm guessing DJ hits more ground balls than anyone else.
by china bob on Dec 26, 2007 3:10 PM PST reply actions 0 recs
Where can one find these statistics?
Do we have league aveages?
by jarforcefatherofforce on Dec 26, 2007 4:28 PM PST reply actions 0 recs
All data from fangraphs.com
I don't have league averages handy, but I can give you league medians (i.e. what Player 81 out of 162 posted--and yes, I know that "median" really should be the averages of Player 81 and Player 82, but seriously) for most of these.
Line Drive: 18.8%
Ground Ball: 42.0%
Fly Ball (note: this is both infield and outfield flies combined): 38.9%
Infield Fly Balls per Fly Ball: 8.7%
Running a little math here, that would give a median score of:
Outfield Fly Balls: 35.0%
Infield Fly Balls: 3.9%
Home Runs per Fly Ball: 10.0%
BABIP: .312
I don't know what the median is for OCSlg, because I calculated that by hand. I seem to recall working out that league average was about .520 not too long ago, however, if that's any help.
by PaulThomas on Dec 26, 2007 4:39 PM PST up reply actions 0 recs
Strictly speaking
Either #81 or #82 or any number in between counts as a median.
What I want to know is how you can make a ranking at all, if you're looking at four different numbers (well, functionally three).
Or do you have a separate median for each figure?
by iglew on Dec 26, 2007 9:20 PM PST up reply actions 0 recs
The latter
You'll notice the figures don't add up to precisely 100%.
by PaulThomas on Dec 26, 2007 9:37 PM PST up reply actions 0 recs
This reminds me of...
a post by Graham at LL that used pitchers peripherals to determine a luck neutral performance metric.
http://www.lookoutlanding.com/story/...
It's about the M's of course, but the spreadsheet he llinks has pitchers from every team.
by MrIncognito on Dec 26, 2007 4:47 PM PST reply actions 0 recs
Interesting read, PT
My hunch on Crosby is that the problem is how hard he doesn't hit the ball (when he hits it at all). Sharp grounders often find holes while soft grounders don't - hence the low BABIP. Partly because he is so easily made off-balance (by pitchers and by his own swing), and partly because he's usually hitting behind in the count, Crosby hits mostly routine grounders and lazy fly balls.
As for Barton, I think he will become one of your two projected "let's hope"s and not the other - I think he will become a "line drive machine" with ho-hum power (more Boggs than Brett). Which is fine by me because I think he'll become a really, really good one - actually, if he can just avoid the pesky regression monster, he already is!
by Nico on Dec 26, 2007 5:02 PM PST reply actions 0 recs
Pesky Regression Monster?
Hm. Let's hope that "Regression is a sometimes phenomenon".
Re: Crosby, I checked out his splits when in various counts-- they actually looked pretty normal to me. He takes a first pitch strike about 60% of the time (basically average), hits a lot worse when the count's against him and hits a lot better when the count's in his favor. Basically, his splits look like your average .240 hitter.
This was only an eyeballing, but I did not come away with the impression that his problems were related to poor performance relative to the ball-strike count.
by PaulThomas on Dec 26, 2007 5:32 PM PST up reply actions 0 recs
What I wonder is whether
Crosby is in a poor ball-strike count more often than league average. If everybody were to hit "a lot worse when the count's against him and a lot better when the count's in his favor," some guys would still hit .300 while others hit .200. Just curious whether a higher pct (compared to league avg) of Crosby's ABs come from 1-2 counts than 2-1 counts, etc.
Mostly, though, I still think the key is how hard he hits the ball. Very few sharp one-hoppers; very many routine 7-hoppers. I think that's why he hits a lot of ground balls yet can't break the .250 barrier - but what causes this, if true, is separate fodder for speculation.
by Nico on Dec 26, 2007 7:42 PM PST up reply actions 0 recs
Good stuff ... quick suggestion ...
it'd be helpful to have rankings and/or various benchmarks (eg top/bottom 10/25%) ...
on Crosby, I agree, it's weird. Not only does he post a solid ld%, he also hits a lot of ground balls, which, if memory serves, are more likely to result in hits (though much less likely to result in xbh) than fly balls.
Do you have Barton's minor league numbers? That'd probably help to identify how much of his numbers are sample size and how much are talent.
by devo on Dec 26, 2007 6:38 PM PST reply actions 0 recs
I haven't seen any minor league batted-ball
breakdowns, so unfortunately that aspect of things can't really be compared.
We can figure out his career BABIP and OCSlg though. For his minors career (not including his postseason rampage this year, which might up each of these a couple of points) his BABIP is .338; his career OCSlg is .545.
For the 2007 regular season, his BABIP was .324, while his OCSlg was .506. Throw in his otherworldly postseason and his MLB stint, and his overall 2007 looks pretty much the same as the rest of his career.
by PaulThomas on Dec 26, 2007 8:01 PM PST up reply actions 0 recs
You've probably seen this already, but
minorleaguesplits.com has some minor league batted ball data, and complete data for Barton's '05, '06 and '07 seasons.
I apologize if I'm simply restating the obvious here. And great post, by the way.
by CletusSJY on Dec 26, 2007 10:25 PM PST up reply actions 0 recs
No, by no means
Thanks for the heads-up! I'll definitely look at it for Hannahan, Murphy and Suzuki.
by PaulThomas on Dec 26, 2007 10:47 PM PST up reply actions 0 recs
The problem with Crosby...
is his incredibly high SUCKops. percentage of 100%, making him the suckiest loser in all of sucktown.
by SwisherSweet on Dec 26, 2007 8:10 PM PST reply actions 0 recs
Wow, you know
when you actually log onto suck.com you realize how many metrics there are for the simple act of sucking.
by Nico on Dec 26, 2007 9:44 PM PST up reply actions 0 recs
The SUCKops formula is flawed
I prefer to use the modified ZsuckOps.
by iglew on Dec 26, 2007 10:51 PM PST up reply actions 0 recs
I used to favor ZsuckOps too,
until I read, on suck.com, that ZsuckOps has a 9.8 suckosity rating as a metric - as rated by ZsuckOps itself. That's why I only use SuckU and Value Over Sucky Replacement (VOSR).
by Nico on Dec 27, 2007 7:54 AM PST up reply actions 0 recs
Excellent Post
One of the best analyses I've ever read on AN.
by richwol on Dec 26, 2007 11:26 PM PST reply actions 0 recs
Nice analysis - thanks.
Since this is first in a series, one tiny point about data presentation (a professional imperative and a personal interest, particularly after reading Tufte).
Your orderings of the possible outcomes are (i) different in their descriptive and analytic listings, swapping GB and LD between Basic Breakdown and Individual Stats, and (ii) not ordered, for example by desirability (high to low, LD, OFB, GB, IFB), hence missing the opportunity to include that additional information.
by green star oakland on Dec 27, 2007 12:52 AM PST reply actions 0 recs
I edited the description
As far as ordering them, though, there's no obvious determiner of whether ground balls or outfield fly balls are better-- it's probably the case that OFBs are better for a majority of hitters, but there are way too many exceptions for this to comprise any kind of general rule.
by PaulThomas on Dec 27, 2007 8:16 AM PST up reply actions 0 recs
I belive an important statistic would be ...
to quantify what % of batted balls on the outer half of the plate are pulled. Both Crosby and Chavez have a terrible tendency to roll over on pitches, resulting in weak grounders to the infield. That separates them from Buck/Barton, and it manifests itself in the BABIP, although there is not a single category that isolates it. It would probably show a high correlation coefficient with grounders, but it would also be true with fly balls I'd guess. The only problem with this stat with Crosby might be that he stands so far away from the plate that the ball doesn't need to be on the outer half to be 'outside' relative to his swing?
by iceplant on Dec 27, 2007 8:53 AM PST reply actions 0 recs
i was thinking along the same lines...
what is the breakdown of balls put in play to left, center, and the right side of the field?
that would be an interesting statistical breakdown for babip and the like.
by ST on Dec 27, 2007 9:34 AM PST up reply actions 0 recs
I thought this would be in ESPN.com's scouting
section, but it was strangely vacant for Crosby, so... eh?
You can check out hit charts for players at MLB.com, though. Crosby is not a dead pull hitter; he probably hits 40% or more of his balls to the right side of second base. It's evidently the velocity, not the direction, of his batted balls that's the issue.
Perhaps a more specific analysis using Gameday data could tell us whether Crosby is putting the "wrong swing" on an excessively large number of balls (in other words, pulling outside pitches, inside-outing inside pitches, etc.). If his grounders to the right side are all on inside pitches and his grounders to the left side are all on outside pitches, he's probably not going to do too well even if he gives the superficial appearance of spreading his batted balls around.
Unfortunately, it would literally take days to perform this for even a single player with the easily accessible technology. I like fiddling with this stuff, but not THAT much.
by PaulThomas on Dec 27, 2007 10:12 AM PST up reply actions 0 recs
but you may be overlooking the evidence
by staring too closely at it. Maybe the problem is that he guides pitches to the opposite field instead of driving them, thereby reducing his natural power. Then he is pulling outside pitches, and they aren't finding holes? Both of these scenarios (and probably one compensating for the other, so they are indications of the same condition) do suggest that he is a dead pull hitter, and he's trying to adjust. (I'd bet a large % of balls hit in the air to the left side of the field do fall safely for Crosby.)
So if you're a pitcher, all you need to do is throw pitches on the outer half of the plate. He has attempted to adjust by stepping away from the plate, maybe so he doesn't pull everything, but that has just made the outer half of the zone unreachable. Maybe he should try moving up in the box, so he hits the slider/curve before it has fallen out of reach?
by iceplant on Dec 27, 2007 11:49 AM PST up reply actions 0 recs
I know he's no longer on the team...
...but could you add Scutaro to the list? I'd like to see how he compares with some of the others.
by UncleLeo on Dec 27, 2007 10:31 AM PST reply actions 0 recs
Scutaro's numbers
LD 20%
GB 39.3%
OFB 36.3%
IFB 4.4%
HR/FB 5.7%
BABIP .278
OCSlg .409
Brief take: hits a lot of liners, but doesn't hit the ball very hard. Gets very limited utility out of non-LD contact.
by PaulThomas on Dec 27, 2007 10:45 AM PST up reply actions 0 recs
Paul
I think you can get batted-ball breakdowns at firstinning.com, too. Great website. It doesn't go back as far as thebaseballcube.com, but it has an easier interface and more esoteric stats.
Great work. :)
by notsellingjeans on Dec 27, 2007 1:47 PM PST reply actions 0 recs
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Related FanPosts on Athletics Nation
Related FanShots on Athletics Nation
FanPosts
Community blog posts and discussion.
Certain photos copyright © 2009 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
