January 08, 2006
Who Gets the Grounders
I've wondered for a while if pitchers or hitters had more of an effect on balls in play being in the air or on the ground. What if pitchers were the only cause of the ball being lifted in the air or not? Then, we wouldn't expect the probabilities to change no matter what type of batter he was facing. If the batter was responsible, then changing the type of pitcher wouldn't matter.
To study this, I selected a group of pitchers that gathered 400 IP from 2002-2004. There were 101 pitchers in the group, and I divided them into quartiles on the probability of a ground ball. Quartile 1 is the group with the lowest probability of a ground ball, quartile 4 the highest. I also selected batters with 1000 plate appearances in that time frame. There were 243 batters in the study, also divided into quartiles on the same statistic. Here's a table representing the probabilities of the four quartiles.
Probability of a Ground Ball
| Quartile | Pitchers | Batters |
| 1 | .38 | .35 |
| 2 | .42 | .41 |
| 3 | .45 | .45 |
| 4 | .52 | .50 |
Next, I pitted each pitcher quartile vs. each batter quartile:
Pitcher Quartiles vs. Batter Quartiles, Probability of a Ground Ball
| Pitchers | Batters |
| Quartiles | 1 | 2 | 3 | 4 |
| 1 | .31 | .37 | .39 | .44 |
| 2 | .33 | .40 | .43 | .50 |
| 3 | .37 | .42 | .46 | .52 |
| 4 | .44 | .48 | .53 | .60 |
As you can see, the type of pitcher and the type of batter both matter. So if you really need a ground ball, get your best ground ball pitcher in against a ground ball hitter.
Posted by David Pinto at
08:56 PM
|
Offense
|
TrackBack (0)
Does the second table use data on individual at-bats from pitchers in a quartile vs. batters in a quartile?
What are the other hitting statistics here. I remember reading several years ago that there was a platoon effect involved here - pitchers who got lots of ground balls had an advantage against hitters who hit a lot of them, kind of like lefty pitchers against lefty hitters. Is that effect here?
Larry
Pete, that's correct.
I haven't looked at the other hitting statistics yet. I just wanted to see if one side had more effect on ground balls than the other.
Nice study, David :) The 4-by-4 chart drives the point home quite easily.
This table is sweet...it actually looks pretty balanced between the hitter and pitcher in terms of how much the outcome is affected.
David,
A corollary study with similar methodology was published in 1990. The inaugural edition of The STATS Baseball Scoreboard divided both hitters and pitchers into 3 groups according to their groundball/flyball ratio for each of the years 1987-89, and published the outcomes for each of the group matchups for each of the three years. The outcomes were expressed using traditional batting statistics. The study's conclusion found a kind of platoon effect - in terms of OBP and SLG, ground ball hitters were more productive against flyball pitchers, and flyball hitters were more productive against groundball pitchers (though their HR did decrease).
The detailed results they published show that for the groundball batter vs groundball pitcher group, homeruns and doubles declined while singles increased. This is consistent with the idea that more groundballs were hit, and fewer fly balls and perhaps fewer line drives. Also, while both the groundball groups were notably below average in strikeout rate, and the strikeout rate dropped further in absolute terms when groundball batter and pitcher group faced off, it did not not drop as much as might be predicted from how far below average each was - strikeouts were relatively more common.
It's intuitively plausible (to me anyway) that this would be the result. Groundball pitchers, through location or deception or both, overall tend to make hitters miss the ball "high" - when they make contact, they'll more often make contact above the center of the ball than below. Flyball pitchers have a greater tendency to get hitters to miss the ball low. Similarly, groundball hitters will tend more often to miss the ball high. I'd expect pitchers' and hitters' tendencies in the same direction to reinforce each other, and tendencies in the opposite direction to cancel each other out.
Robert Adair's book The Physics of Baseball provides a framework for understanding why this might happen. In his book Adair diagrams and contrasts the "level swing" of Rod Carew and the "uppercut swing" of Reggie Jackson; when Reggie makes a timing mistake and swings a little too early, the bat trajectory will cause him to make contact on top of the ball (if at all) and pull a ground ball, a late swing would result in an opposite field flyball or popup. With the uppercut, a well timed swing would result in a well-struck fly ball. Of course, there are other sorts of errors a batter can make besides swinging early or late (eg horizontal or vertical errors in placement of the sweet spot of the bat in the path of the center of the ball even if he has the timing right).
I'd be very interested to see what the results would be if you chose to extend the study to the full set of hit types. I'd expect when fly ball batters faced fly ball pitchers that the batters would be even more underneath the ball - their popups turn into swings and misses (bumping up strikeouts), their easy fly balls into popups, their line drives might turn into fly balls, their hard ground balls into line drives, and so on. Vice versa for groundball hitters vs groundball pitchers.
You guys ever see the movie "Pi"?
Not to take away from your conclusions, but it seems redundant to deduce a formula to tell you that a groundball is dependent both on the pitcher and batter (I mean, duh!). But Joe Arthur makes an interesting point in his 3rd paragraph. I would find a percentage of swinging k's (or even contact made) either high or low based on groundball/flyball pitchers and uppercut/downward hitters. Seems like that'd be useful info for a manager in a playoff game: i.e. Vinny Castilla and his uppercut swing are stepping to the plate... who should you bring in to face him?
Anybody have access to something like that?
Joe Arthur, thanks for the very thoughtful comment. I will be looking at this more.
Nat, yes, I did see "Pi." It came up as a Netflix recommendation. A very strange movie indeed.
As for reduncancy, yes, it may be obvious, that both have an effect, but it's nice to know for sure.
Here's another thought - if you've got a pitcher who's making the hitters hit the ball a little lower, but the guy is a fly-ball hitter, maybe he's hitting line drives, while the ground ball hitters hit line drives off the flyball pitchers.
Fly-ball hitters pop-up against fly ball pitchers and ground0ball hitters hit worm-burners against ground ball pitchers.
Just a theory. What does the data say, David?
Larry
Yeah, "Pi" was really strange.
I only bring it up because that guy discovered a formula for life. Then he was hunted for it and/or went crazy but that's beside the point ... This is what I love about baseball. Everyone studies it like alchemy in an effort to discern patterns when baseball is so much more like life: unpredictable. And that's what makes it such a great sport. I'm not saying statistical patterns don't exist. They do in life too. But baseball analysis is like a methodical study to find a formula for Chaos. And much like alchemy was, it's entirely fascinating.
Just my two philosophical cents, and commentary on the Hardball Times, Bill James, et al.