## Visualizing strikeouts, walks, and ground balls using standard deviations

One of the strikeout kings - Tim Heitman-USA TODAY Sports

We know strikeouts are good, walks are bad, and ground balls are better than other types of batted balls, but how exactly does the data look when we divide up pitchers by standard deviations?

Some relationships between a statistic and overall performance are predictable. For example, we know that higher strikeout pitchers perform better than low strikeout pitchers on average. Yu Darvish and Max Scherzer are good examples of this. The relationship between ground balls and overall performance is a little less clear. Strikeouts are good, walks are bad, and ground balls are usually better than line drives or fly balls, but the question we're looking at here is how ground ball rate relates to overall performance without controlling for the other two factors. Batted ball data is available since 2002, so this research will draw on the last twelve seasons.

I have calculated standard deviations for strikeout rates, walk rates, and ground ball rates in order to determine the relationships between these rates while grouped by standard deviations from the mean (or median depending on the nature of the distribution in the given year). The results, presented here, show standard deviations with K% (Table 1), BB% (Table 2), and GB% (Table 3). I chose several performance metrics to paint the widest picture. ERA, FIP, xFIP, SIERA, ERA-, FIP-, xFIP-, RA9-fWAR, and fWAR are included in this analysis with median values that correspond to the given grouping of K%, BB%, or GB%. In the tables, "-2+" means at least two standard deviations below the mean, "-1+" means between 1 and 2 standard deviations below the mean, "0-1" means between the mean and 1 standard deviation below the mean, and so forth.

K% SD ERA FIP xFIP SIERA ERA- FIP- xFIP- RA9-WAR WAR GB%
-2+ 6.04 5.85 5.47 5.63 136.5 144 130 -0.7 -0.4 55.80%
-1+ 5.46 5.22 4.92 5.09 125 120 117 -0.1 0.3 46.20%
0-1 4.89 4.81 4.65 4.73 111 110 109 0.7 0.9 43.30%
0+1 4.32 4.31 4.21 4.24 100 99 99 1.7 1.7 42.80%
1+ 3.79 3.80 3.70 3.74 88 88 90 3.0 3.1 43.10%
2+ 3.19 3.16 3.32 3.26 74.5 73.5 78 4.8 4.4 40.70%

The first table shows the relationship between strikeout rate and overall performance. As one would expect, better performance is well correlated with a higher strikeout rate, which is obviously true in the metrics that use strikeout rate as an input. It is very rare to find a successful pitcher with a very low strikeout rate, which is evident in metrics like ERA and RA9-WAR. Interestingly, ground ball rate decreases as strikeout rate increases. This is also a known phenomenon; pitches higher in the zone tend to generate more strikeouts while also generating fewer ground balls.

BB% SD ERA FIP xFIP SIERA ERA- FIP- xFIP- RA9-WAR WAR GB%
-2+ 3.82 3.70 3.91 4.07 86 84 88 3.2 3.5 42.70%
-1+ 4.03 3.99 3.99 4.07 92 92 93 2.6 2.7 43.30%
0-1 4.36 4.37 4.21 4.29 100 100 100 1.7 1.8 43.90%
0+1 4.66 4.69 4.51 4.55 106 107 106 1.0 1.1 43.10%
1+ 5.05 4.95 4.81 4.83 118 112 113 0.3 0.6 42.90%
2+ 5.50 5.55 5.10 5.23 126 127 122 0.0 0.1 40.90%

The second table shows the relationship between walk rate and overall performance. Similar trends are observed, though less pronounced than with strikeout rates. The range of performance is less extreme, and contrary to strikeout rates, good pitchers can be found in each group.

GB% SD ERA FIP xFIP SIERA ERA- FIP- xFIP- RA9-WAR WAR K% BB%
-2+ 4.54 5.11 5.03 4.87 112 120 122 0.5 0.1 17.4% 9.8%
-1+ 4.63 4.62 4.63 4.52 108 110 110 0.9 1.0 17.5% 8.9%
0-1 4.55 4.56 4.53 4.53 106 107 106 1.1 1.3 16.0% 8.7%
0+1 4.37 4.29 4.25 4.39 103 101 100 1.4 1.6 15.7% 8.5%
1+ 4.10 4.15 4.06 4.20 98 99 97 1.6 1.55 15.0% 8.8%
2+ 3.91 3.97 3.95 3.94 90 94 93 2.3 2.25 13.8% 7.9%

The final table shows the relationship between ground ball rate and overall performance. As the ground ball rate increases, performance improves across the board. Similar to walk rate, good pitchers can be found in each group; an extremely low ground ball rate is not the death sentence that an extremely low strikeout rate is. At the extremes, it appears there could be a relationship between ground ball rate and walk rate, but the sample sizes at the extremes are fairly low.

While the general trends are present, there is still considerable variation among many of the groups. In future analyses, I will be evaluating combinations of groups, for instance high K%/high GB% (The Winter Soldiers), to see how different ground ball rates affect performance within groups of walk rates and strikeout rates.

. . .

All statistics courtesy of FanGraphs.

Kevin Ruprecht is a contributor for Beyond the Box Score. He also writes at Royal Stats for Everyone. You can follow him on Twitter at @KevinRuprecht.

More from Beyond the Box Score

## Trending Discussions

forgot?

As part of the new SB Nation launch, prior users will need to choose a permanent username, along with a new password.

I already have a Vox Media account!

### Verify Vox Media account

As part of the new SB Nation launch, prior MT authors will need to choose a new username and password.

We'll email you a reset link.

Try another email?

### Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

### 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.