Do Pitchers Induce More Ground Balls in Double Play Situations?

ARLINGTON TX - OCTOBER 22: Mariano Rivera's arm stayed on after this pitch. (Photo by Stephen Dunn/Getty Images)

Earlier this week I was gallivanting around Fangraphs' criminally underused Q&A section when a user caught my eye with this question for the saber world: "Does a batter's GB% raise, drop or remain even with a runner at first?"

Seems like a reasonable assumption, right? In bases-empty situations, a pitcher may not concern himself so much with inducing a ground ball as opposed to simply just working an out by any means necessary. But, with a runner on first, of course, a ground ball out becomes especially attractive as it creates the opportunity for the ever-valuable double play.

I ran the data and found that since 2002, GB% (as in GB/batted balls) with Bases Empty was 44.9% across the league. With a runner on first and the double play now a possibility, that number actually drops to 43.7%. Now, I'm going to ignore the drop and assume its just noise, but I will admit to being mildly amused by this. (This is using retrosheet's definitions of batted ball types, keep in mind, and BIS data may differ.)

Curious, I then ran the data again and included only situations with less than 2 outs, when the double play would be beneficial. But even then there was no change:

BATTED BALL BY BASE-STATE

BASE STATE GB% FB% LD% PU%*
Bases Empty 44.7 37.1 18.2 7.9
"1--" 43.7 37.3 19 7.6
"-2-" 46.8 35.2 18 7.5
"12-" 44.3 37.4 18.3 7.5
"--3" 44.7 37.5 17.7 7.6
"1-3" 44.9 37.7 17.4 7.5
"-23" 46.1 36.4 17.4 7.4
"123" 44.2 38.4 17.4 8.2

*FB% includes "pop-ups"

And then I ran the data once more using Plate Appearances as the denominator, just to make sure some non-batted ball events weren't interfering with the results, but still nothing:

BASE_STATE GB% FB% LD% PU%
Bases Empty 31.8 26.4 12.9 5.6
"1--" 30.3 25.9 13.2 5.3
"-2-" 30.3 22.8 11.7 4.9
"12-" 29.7 25 12.2 5
"--3" 30.2 25.3 12 5.1
"1-3" 32.4 27.3 12.5 5.4
"-23" 28.3 22.4 10.7 4.6
"123" 32.7 28.4 12.8 6.1

Of course, inducing ground balls isn't every pitcher's forte. So I kept that same time frame, 2002-2011, and restricted the search to only those pitchers who posted a ground ball rate of at least 50% over at least 500 innings. I queried their GB% with both bases empty and in "double play situations", which I defined as base-states 1--, 12-, 1-3, and 123, with fewer than 2 outs.

On average that group posted a GB-rate of 54.4% with the bases empty, and in double play situations 54.04%. So no change there either, but that may not be saying much-- if your best weapon is a sinker or 2-seamer that induces a high rate of grounders, why limit it to just Double Play Situations?

So by this point, you're probably thinking the same thing I was thinking: enough of this nonsense, just show me which pitchers actually do induce more grounders in double play situations. Can do:

# First Last IP GB% GB% BE GB% DPS Diff
1 Billy Wagner 566.7 43.9 40.8 49.3 8.5
2 Scott Linebrink 637.7 37.8 36.1 43.4 7.3
3 Brad Lidge 639.3 41.7 39.1 45 5.9
4 Mariano Rivera 740.3 54.7 53 58.2 5.2
5 Adam Eaton 927 40.3 38.3 43.3 5
6 Carlos Silva 1246.7 48.9 47.5 51.9 4.4
7 Jason Jennings 1089 46.6 46.2 50.6 4.4
8 Kirk Saarloos 509 56.7 55.1 59.4 4.3
9 Jason Johnson 872 51 49.5 53.6 4.1
10 Chris Young 782.3 28.2 27.4 31.4 4

Now, I've run some pretty bizarre queries in my time. I'm not kidding you-- I've sifted out things as ridiculous as night-game wOBA with RISP and 2-outs on get-away days in September for the Montreal Expos. Believe me, there is an unending universe of small samples that lie deep, deep within the retrosheet files.

But never in my life have I run a top ten list that returns the one-and-only Mariano Rivera next to a one Adam Eaton. I mean, he doesn't even deserve a hyperlink!

So, we are likely seeing quite a bit of noise here. But, if I was to be convinced that anyone had the capacity to induce ground balls on demand in situations where the batter was positively trying to avoid such an outcome, I would put my money on The Sandman. Not to take anything away from Billy Wagner, whose 8.5% difference makes an interesting case as well. Considering that the 10th greatest surge in GB-rate during DPS was only 4%, Wagner's 8.5% is certainly compelling.

Both Brad Lidge and Scott Linebrink had brief (ok, very brief) stretches where they seemed to have performed close to the Wagner/Rivera level, but I'm nowhere near sold on that. Carlos Silva frequently posted ERA's lower than his FIP during his time with Minnesota, and inducing an inordinate amount of double plays could have been a contributor to that in theory. And it's conceivable Chris Young, an extreme fly ball pitcher, could have altered his pitch selection slightly in double-play situations, becoming a not-so-extreme fly ball pitcher in those conditions. But that's a long walk.

I have little doubt there is a deliberate effort on the pitchers part to induce grounders when they can benefit them the most. But what we may be witnessing in these tables is the opposing efforts of the hitter responding to the situation as well, counteracting and neutralizing that deliberate effort.

In extreme cases, the elite pitcher may win out. In other cases, a dude might just be getting lucky.


Increase your stamina and follow @JDGentile on twitter today!

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

Join Beyond the Box Score

You must be a member of Beyond the Box Score to participate.

We have our own Community Guidelines at Beyond the Box Score. You should read them.

Join Beyond the Box Score

You must be a member of Beyond the Box Score to participate.

We have our own Community Guidelines at Beyond the Box Score. You should read them.

Spinner

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9351_tracker