Remember me

Curtis Granderson Defense-Independent Batting Statistics

July 26, 2015

Curtis Granderson produced some exceptional seasons with the Tigers and Yankees, but a broken forearm and finger limited him to 61 games in 2013, his last with the Yankees. He was already 33 years old in 2014, the first year of his new four-year, $60 million contract with the Mets. With his power numbers down that ensuing season, one might wonder whether Granderson had entered the decline phase of his career. However, Granderson’s batted ball profile told a different story.

Baseball Info Solutions charts the location of and times every batted ball, and with that information, we can evaluate hitters based on their quality of contact rather than the actual results. For example, if a hitter hits a hard liner to shallow left field that just happens to be caught, its result is an out that seems no different than a lazy pop out. However, we can compare that line drive to balls with similar trajectories historically to discover how often those types of balls become outs and then award credit based on those ratios. In that example, the hitter might earn 0.8 hits because this type of batted ball tends to become hits (even though it did not in that specific instance).

The aggregation of those expected values, which we call Defense-Independent Batting Statistics (DIBS), have proven to be more effective than actual results at predicting future results. Meanwhile, check out Granderson’s actual batting line and DIBS batting line from 2014:

Curtis Granderson, 2014
  PA H 2B 3B HR BABIP AVG Batting Runs
Actual 654 128 27 2 20 .265 .227 3.5
DIBS 654 143 32 4 25 .305 .253 20.3

 

Even in his best seasons, Granderson did not have exceptional batting averages. Still, DIBS expected him to have 15 more hits than he actually had last season. Meanwhile, his power numbers were especially below expectations. DIBS suggested Granderson should have hit five more doubles, two more triples, and five more home runs than he did based on his batted ball trajectories.

Curtis Granderson, 2015
  PA H 2B 3B HR BABIP AVG Batting Runs
Actual 404 88 16 1 14 .301 .251 6.8
DIBS 404 91 19 2 15 .316 .258 11.9

 

So far this season, Granderson has produced numbers much more in-line with both his 2014 DIBS numbers and his 2015 DIBS numbers. His actual batting average of .251 is just two points off his DIBS average from last season, and his 14 home runs through 404 plate appearances has him on pace to hit 23 this year. On the surface, Granderson seemed to decline sharply in 2014 and then bounce back so far this season. However, his DIBS numbers suggest that his 2014 production underachieved what would be expected with the quality of his batted balls, and so far in 2015, he has maintained a similar level of play.

 
 

COMMENTS (7 Comments, most recent shown first)

albriz
I saw Pete Rose doing BP in SF in the late 70's. The first pitch he hit was in the hole between shortstop and 3rd. The second was in the hole between 2nd and 1st. The 3rd was up the middle. The fourth was down the left field line (by then I realized what he was doing). The next was down the right field line. The next was in the left field gap, and the last was in the right field gap.
There is solid contact, and then there are those few who can place their hits like Wee Willie Keeler (I have no idea if he could really "hit them where they ain't").
All this just to say that I'm guessing Granderson tries to make solid contact. Rose, Tony Gwynn, Oliva, and many others were able to make solid contact AND place the ball. It would be interesting whether or not the stats bear out this theory.​
9:13 PM Aug 2nd
 
evanecurb
I remember, years and years ago, hearing Bob Watson of the Astros describe the difference between the big leagues and the minor leagues. He said the percentage of hard hit balls that become outs is much, much higher in the major leagues than in the minors. Watson, you may remember, was a line drive hitter who was very slow, so he had to hit the ball out of the infield in order to reach first.
9:23 AM Jul 28th
 
shthar
Once we remove the reliance on actual results we'll be able to get some REAL research done here. The kind people pay for.
7:23 PM Jul 27th
 
Zeke**
@steve: I assume the HR numbers are basically capturing park effects...Granderson must've hit at least 5+ line drives / big flies that would have been HRs in a high percentage of parks but were long outs when and where he happened to hit them.
5:10 AM Jul 27th
 
steve161
Very interesting, but I don't understand the home run numbers. Either the ball goes over the fence or it doesn't. What accounts for the difference: robberies by an outfielder? actual fence vs idealized fence? ball hitting seagull?
6:30 PM Jul 26th
 
greggborgeson
John, this is great info.

I've always felt that studies of streakiness that I've seen were fatally flawed because they have been dependent on actual results, with the random factors you described. If, as you've shown, actual results can be so misleading in a sample size of 654 at bats, think how much of an impact it has when measuring streakiness over a dozen or fewer at bats. It seems to me that it renders those studies utterly meaningless.

Have you done any studies of streakiness related to quality of contact? I know you tend to dismiss streaks -- but I'd like to see that validated. My guess is you will find at least a significant degree of predictability.


4:26 PM Jul 26th
 
hotstatrat
Interesting . . . how much more closely do all player's stats reflect the DIBS estimated stats of the previous year's stats than the actual previous year's stats - or by comparison a version of their previous year's stats adjusted for an errnant BABiP? (. . . noting the advantage that it may be hard to actually know what a player's true BABiP is, while DIBS is independant of balls dropping where they aint.)
2:51 PM Jul 26th
 
 
©2024 Be Jolly, Inc. All Rights Reserved.|Powered by Sports Info Solutions|Terms & Conditions|Privacy Policy