Back in June, I took on the issue of groundball pitchers. Well, more specifically, I took on the issue of groundball pitchers and Bill James’ antipathy toward them. This article was prompted by James’ comment, “If you like Groundball pitchers, you’re welcome to them. I don’t want nothin’ to do with them.” In the article, I looked at pitchers last year. I found that if you rank every pitcher in 2015 by groundball percentage and divided them into deciles, generating more groundballs was correlated fairly strongly to fewer home runs and lower ERA and FIP. Here’s the key table:
|
Decile |
ERA |
FIP |
BB% |
K% |
HR% |
|
|
1 |
61.8% |
3.47 |
3.62 |
7.9% |
19.4% |
1.9% |
|
2 |
54.2% |
3.95 |
3.81 |
8.0% |
19.6% |
2.2% |
|
3 |
51.6% |
3.92 |
3.94 |
7.7% |
19.2% |
2.4% |
|
4 |
49.4% |
3.73 |
3.72 |
7.5% |
20.9% |
2.3% |
|
5 |
47.7% |
3.67 |
3.83 |
7.2% |
20.5% |
2.6% |
|
6 |
46.1% |
4.16 |
4.01 |
8.4% |
20.5% |
2.6% |
|
7 |
44.1% |
3.96 |
3.97 |
7.2% |
20.6% |
2.8% |
|
8 |
41.9% |
4.25 |
4.12 |
7.3% |
21.0% |
3.1% |
|
9 |
39.5% |
4.30 |
4.18 |
7.6% |
21.5% |
3.2% |
|
10 |
33.5% |
4.22 |
4.43 |
7.7% |
20.6% |
3.6% |
|
Correlation |
(0.80) |
(0.92) |
0.27 |
(0.75) |
(0.98) |
|
See that? As groundball percentage rises, walks rise a little, strikeouts decline a bit, but more dramatically, home runs decline, because it’s hard to hit a home run on a groundball. That drives a sharply lower ERA and FIP.
I followed up by looking at pitchers from other seasons. I had already done the numbers for 2015, so I went back a decade at a time, to 2005, 1995, 1985, 1975, 1965, and 1955. I found that while the equation more groundballs = better run prevention holds in general, it doesn’t work well in some years, and the relationship is stronger today than it was decades ago. I also found some merit in James’ position, in that both the groundball metric he used (groundball double plays divided by GIDP opportunities) and the pool of pitchers he considered (only those with very long careers) supported his conclusion. However, I stuck by the assertion that in contemporary baseball, if you’re looking at a given pitcher, getting more groundballs is a good thing. It’s not the only thing—among ERA qualifiers, Max Scherzer has the seventh-lowest groundball rate and Justin Verlander the ninth-lowest, and they rank third and sixth, respectively, in PWARP—but it’s a good starting point in pitcher evaluation, all things being equal.
But that pitcher evaluation is in terms of two metrics, ERA and FIP, that are imperfect. ERA is affected by sequencing, FIP assumes no differences in batted ball contact, both are influenced by ballpark factors and catcher framing and opponent quality and defense and…well, if this sounds familiar, it’s because you’ve seen it here before, in last year’s rollout of Deserved Run Average (DRA) Jonathan Judge, Harry Pavlidis, and Dan Turkenkopf. DRA is BP’s pitching metric that synthesizes all individual batting events, adjusts them for their context, and adds baserunning and errant pitches. And it’s been improved, both last year and this year. It’s also been adjusted for groundball tendencies. In DRA, we have a broader measure of pitcher quality and effectiveness than ERA or FIP. So what does it say about groundball pitchers?
Here’s the same chart as above—every pitcher from 2015, divided into 10 equally sized deciles based on groundball percentage—with innings-weighted DRA included along with ERA and FIP.
|
Decile |
GB% |
ERA |
FIP |
DRA |
BB% |
K% |
HR% |
|
1 |
61.8% |
3.47 |
3.62 |
3.92 |
7.9% |
19.4% |
1.9% |
|
2 |
54.2% |
3.95 |
3.81 |
4.17 |
8.0% |
19.6% |
2.2% |
|
3 |
51.6% |
3.92 |
3.94 |
4.22 |
7.7% |
19.2% |
2.4% |
|
4 |
49.4% |
3.73 |
3.72 |
4.05 |
7.5% |
20.9% |
2.3% |
|
5 |
47.7% |
3.67 |
3.83 |
4.12 |
7.2% |
20.5% |
2.6% |
|
6 |
46.1% |
4.16 |
4.01 |
4.33 |
8.4% |
20.5% |
2.6% |
|
7 |
44.1% |
3.96 |
3.97 |
4.22 |
7.2% |
20.6% |
2.8% |
|
8 |
41.9% |
4.25 |
4.12 |
4.34 |
7.3% |
21.0% |
3.1% |
|
9 |
39.5% |
4.30 |
4.18 |
4.37 |
7.6% |
21.5% |
3.2% |
|
10 |
33.5% |
4.22 |
4.43 |
4.73 |
7.7% |
20.6% |
3.6% |
|
Correlation |
(0.80) |
(0.92) |
(0.89) |
0.27 |
(0.75) |
(0.98) |
|
DRA concurs with the initial finding—more groundballs, in general, yield a lower DRA, in general. A 10 percentage point increase in groundball percentage results in a 0.24 reduction in DRA, and the correlation is pretty strong.
As with my previous analysis, I checked to see if this has always been the case. Here’s 2005:
|
Decile |
GB% |
ERA |
FIP |
DRA |
BB% |
K% |
HR% |
|
1 |
60.0% |
3.83 |
3.97 |
4.16 |
7.7% |
15.2% |
2.1% |
|
2 |
52.7% |
4.17 |
4.18 |
4.39 |
8.6% |
16.1% |
2.3% |
|
3 |
50.2% |
4.09 |
4.18 |
4.49 |
7.7% |
15.3% |
2.5% |
|
4 |
47.6% |
4.02 |
4.02 |
4.22 |
7.6% |
17.2% |
2.5% |
|
5 |
46.1% |
4.31 |
4.24 |
4.61 |
8.0% |
16.0% |
2.6% |
|
6 |
44.4% |
4.38 |
4.35 |
4.51 |
8.1% |
16.7% |
2.8% |
|
7 |
42.4% |
4.48 |
4.43 |
4.52 |
8.4% |
16.9% |
2.9% |
|
8 |
40.6% |
4.30 |
4.32 |
4.50 |
8.2% |
16.5% |
2.8% |
|
9 |
38.2% |
4.57 |
4.60 |
4.80 |
8.6% |
17.2% |
3.2% |
|
10 |
33.4% |
4.78 |
4.63 |
4.79 |
8.7% |
17.3% |
3.3% |
|
Correlation |
(0.92) |
(0.90) |
(0.84) |
(0.62) |
(0.80) |
(0.97) |
|
That’s still a strong relationship. I should note that pitchers who generate a lot of groundballs get fewer strikeouts than their flyball peers. Even so, groundballs suppressed scoring. More grounders, lower DRA.
Here’s 1995:
|
Decile |
GB% |
ERA |
FIP |
DRA |
BB% |
K% |
HR% |
|
1 |
60.7% |
4.03 |
4.05 |
4.26 |
8.6% |
15.8% |
1.9% |
|
2 |
54.8% |
4.24 |
4.30 |
4.74 |
9.0% |
15.0% |
2.1% |
|
3 |
52.4% |
4.44 |
4.42 |
4.84 |
8.7% |
14.6% |
2.4% |
|
4 |
50.3% |
4.43 |
4.51 |
4.88 |
9.2% |
14.8% |
2.5% |
|
5 |
48.6% |
4.50 |
4.30 |
4.61 |
8.6% |
16.2% |
2.5% |
|
6 |
46.9% |
4.17 |
4.11 |
4.21 |
8.2% |
17.9% |
2.5% |
|
7 |
44.6% |
4.68 |
4.58 |
4.89 |
9.3% |
16.6% |
2.8% |
|
8 |
41.9% |
4.37 |
4.50 |
4.61 |
9.3% |
17.3% |
2.8% |
|
9 |
39.4% |
4.54 |
4.75 |
5.15 |
9.6% |
16.0% |
3.0% |
|
10 |
34.1% |
5.13 |
5.02 |
5.14 |
10.4% |
18.2% |
3.6% |
|
Correlation |
(0.80) |
(0.84) |
(0.61) |
(0.73) |
(0.67) |
(0.98) |
|
Remember when an ERA of 4.00 was really good? Anyway, the correlation’s not as strong, but it’s still decent. A 10 percentage point increase in groundball percentage knocked a quarter run off DRA in 1995.
Less variance, but still a strong correlation, in 1985:
|
Decile |
GB% |
ERA |
FIP |
DRA |
BB% |
K% |
HR% |
|
1 |
68.9% |
3.82 |
3.78 |
3.95 |
9.4% |
12.4% |
1.6% |
|
2 |
61.2% |
3.88 |
3.76 |
4.08 |
8.2% |
13.3% |
2.0% |
|
3 |
57.3% |
3.97 |
3.86 |
4.22 |
8.8% |
13.1% |
2.0% |
|
4 |
54.3% |
3.90 |
3.84 |
4.31 |
7.5% |
12.6% |
2.2% |
|
5 |
52.4% |
3.75 |
3.90 |
4.09 |
9.0% |
14.2% |
2.2% |
|
6 |
50.3% |
3.97 |
3.86 |
4.28 |
8.4% |
14.5% |
2.3% |
|
7 |
48.6% |
3.83 |
3.77 |
4.17 |
8.7% |
14.9% |
2.1% |
|
8 |
45.9% |
3.68 |
3.85 |
4.31 |
8.5% |
14.6% |
2.3% |
|
9 |
43.4% |
4.18 |
4.19 |
4.34 |
8.6% |
14.5% |
2.9% |
|
10 |
36.7% |
3.95 |
4.12 |
4.49 |
9.2% |
16.0% |
2.9% |
|
Correlation |
(0.25) |
(0.71) |
(0.87) |
0.00 |
(0.89) |
(0.93) |
|
The run environment in 1985 was more than half a run per team lower than in 1995, so a 10 percentage point increase in groundball rate knocked 0.15 off DRA rather than 0.25 a decade earlier. But the correlation’s still negative—more groundballs equals more run prevention—and it’s quite strong; stronger than for ERA and FIP, in fact.
Things were different in 1975:
|
Decile |
GB% |
ERA |
FIP |
DRA |
BB% |
K% |
HR% |
|
1 |
65.2% |
3.68 |
3.67 |
4.18 |
8.5% |
11.4% |
1.6% |
|
2 |
59.1% |
3.67 |
3.72 |
4.46 |
8.9% |
11.0% |
1.5% |
|
3 |
56.5% |
3.85 |
3.64 |
4.46 |
9.0% |
11.8% |
1.5% |
|
4 |
54.3% |
3.89 |
3.80 |
4.36 |
9.3% |
13.0% |
1.9% |
|
5 |
52.6% |
3.77 |
3.71 |
4.27 |
9.2% |
12.9% |
1.7% |
|
6 |
50.8% |
3.58 |
3.78 |
4.31 |
9.3% |
12.6% |
1.8% |
|
7 |
48.6% |
3.61 |
3.58 |
3.92 |
8.9% |
15.1% |
2.0% |
|
8 |
46.8% |
3.68 |
3.63 |
4.20 |
8.7% |
14.2% |
2.0% |
|
9 |
44.9% |
3.76 |
3.70 |
4.28 |
8.8% |
14.3% |
2.1% |
|
10 |
39.6% |
3.62 |
3.88 |
4.50 |
9.5% |
13.4% |
2.1% |
|
Correlation |
0.26 |
(0.31) |
(0.01) |
(0.47) |
(0.77) |
(0.87) |
|
As with ERA and, to a lesser degree, FIP, the relationship between groundballs and run prevention as measured by DRA broke down some in 1975. But while the correlation coefficient for DRA is negligible, it’s worth noting that the most extreme flyball pitchers had the highest DRA, and the most extreme groundball pitchers had the third-lowest DRA in 1975. So it’s not totally random.
The year 1965 looks more like recent years:
|
Decile |
GB% |
ERA |
FIP |
DRA |
BB% |
K% |
HR% |
|
1 |
67.1% |
3.20 |
3.36 |
3.54 |
8.5% |
14.3% |
1.7% |
|
2 |
60.6% |
3.67 |
3.47 |
3.75 |
8.5% |
15.0% |
1.9% |
|
3 |
58.5% |
3.29 |
3.44 |
3.72 |
7.3% |
14.4% |
2.1% |
|
4 |
55.7% |
3.41 |
3.22 |
3.50 |
8.2% |
16.1% |
1.8% |
|
5 |
52.8% |
3.40 |
3.50 |
3.53 |
7.7% |
15.9% |
2.3% |
|
6 |
50.5% |
3.54 |
3.66 |
3.87 |
8.3% |
14.6% |
2.3% |
|
7 |
48.7% |
3.82 |
3.51 |
3.83 |
8.3% |
16.2% |
2.2% |
|
8 |
46.8% |
3.35 |
3.53 |
3.73 |
8.4% |
17.4% |
2.4% |
|
9 |
44.2% |
3.63 |
3.55 |
3.87 |
8.4% |
17.5% |
2.5% |
|
10 |
40.2% |
3.70 |
3.75 |
3.93 |
8.7% |
16.9% |
2.8% |
|
Correlation |
(0.58) |
(0.70) |
(0.69) |
(0.24) |
(0.80) |
(0.92) |
|
A 10 percentage point increase in groundball rate projects to a 0.13 reduction in DRA, with a nice correlation.
Finally, 1955:
|
Decile |
GB% |
ERA |
FIP |
DRA |
BB% |
K% |
HR% |
|
1 |
61.3% |
3.97 |
3.86 |
4.29 |
9.4% |
10.7% |
2.3% |
|
2 |
56.7% |
4.19 |
4.04 |
4.47 |
10.1% |
11.2% |
2.5% |
|
3 |
53.9% |
3.97 |
3.90 |
4.47 |
9.7% |
11.1% |
2.3% |
|
4 |
51.5% |
4.33 |
4.08 |
4.32 |
10.8% |
11.5% |
2.3% |
|
5 |
48.1% |
3.99 |
4.19 |
4.30 |
9.2% |
11.1% |
2.9% |
|
6 |
46.1% |
3.94 |
4.02 |
4.35 |
10.9% |
12.8% |
2.5% |
|
7 |
44.6% |
4.22 |
3.96 |
4.13 |
11.1% |
12.4% |
2.3% |
|
8 |
42.5% |
3.73 |
3.86 |
4.00 |
10.5% |
13.6% |
2.3% |
|
9 |
39.5% |
3.86 |
3.97 |
3.83 |
9.6% |
13.6% |
2.7% |
|
10 |
34.5% |
3.79 |
4.12 |
3.86 |
10.1% |
13.4% |
2.9% |
|
Correlation |
0.50 |
(0.29) |
0.83 |
(0.25) |
(0.90) |
(0.54) |
|
As with ERA, the relationship between groundball percentage and DRA is flipped in 1955. A 10 percentage point increase in groundball percentage is correlated to a 0.12 increase in ERA and a 0.24 increase in DRA. Groundball pitchers were less successful at preventing runs than their flyball brethren.
So what does this tell us? Does DRA confirm or negate the previous findings suggesting that groundball pitchers are more effective at run prevention? There appear to be three conclusions:
1. In modern baseball, using the most sophisticated measure of run prevention, higher groundball rates are well correlated to fewer runs allowed. It’s not an ironclad relationship—Scherzer and Verlander don’t get a lot of grounders—but inducing groundballs is a positive attribute.
2. To James’ point, if we look over the arc of baseball history, the conclusion above decays with time. Getting groundballs is really good today. It wasn’t necessarily good 40-60 years ago. So while we can say “groundball pitchers are generally better” today, that statement’s time-limited.
3. This illustrates the limitation of single-variable analysis in baseball. Inducing groundballs is positively correlated with run prevention. So is getting strikeouts. But grounders and whiffs are negatively correlated—groundball pitchers get fewer strikeouts than flyball pitchers. So getting a lot of grounders, just like getting a lot of strikeouts, isn’t enough to guarantee success. Among ERA qualifiers this year, Edinson Volquez has the seventh-highest groundball rate, and Ian Kennedy’s no. 18 in strikeout rate. Their ERA/FIP/DRAs are 4.99/4.38/4.85 and 4.03/4.92/3.95, respectively; the American League average is 4.21.
I’ll concede that James has a point about durable pitchers over the past 60 years. But I’ll stick with what I wrote in June, “I’m not backing away from the view that in contemporary baseball, groundball pitchers, in aggregate, are more valuable the flyball pitchers, in aggregate,” and DRA’s got my back.
Thank you for reading
This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.
Subscribe now
I also notice that in all the post-1955 years, there's a local minimum in DRA that slides between the 4th and 7th decile, and corresponds with a dip in walk rate and a peak in K rate, while leaving HR rate unaffected. Looks like a potential sweet spot where enough of the missing groundballs are being turned into outs (or at least not into baserunners) to keep the runs down. Might be interesting to double-check OBA-against to make sure it follows there.
Good catch about the DRA minimum in 1965, 1975, and 1995 and the near-minimums in the other years. It makes sense that more Ks, in particular, would enhance the run-suppressing qualities of grounders.