March 03, 2009
Speed In and Out
On Sunday I looked at the effect of speed on getting a good result in different areas of the strike zone. The result was interesting as high speed pitches produced a higher percentage of favorable results, but low speed pitches out performed expectations of the probabilistic model. I was wondering what caused this, and decided to test a hypothesis that it had something to do with the location of these pitches in and out of the strike zone. Fast pitches tend to be straighter; they have less time to curve. If a pitcher needs to throw a strike, a fast pitch is the easiest one to control and put in the strike zone. Of course, these pitches in the strike zone are also easier to hit.
On the other hand, slow pitches tend to be breaking balls, pitches that fool batters. A slow pitch in the strike zone should get hammered, but pitchers get good results by getting batters to chase them out of the strike zone.
The following table shows the hypothesis has some validity. I defined an area as in the strike zone if the positive probability outcome was greater or equal to .7. The in ratio shows the ratio of the actual success rate in the strike zone to the expected success rate in the strike zone. A number over 100 indicates better than expected success. The out ratio is the same for pitches out of the strike zone.
| Speed | Pitches | In Ratio | Out Ratio |
|---|
| 100 | 253 | 106.76 | 110.26 |
| 99 | 818 | 101.64 | 96.20 |
| 98 | 2122 | 101.65 | 106.02 |
| 97 | 5093 | 101.22 | 99.28 |
| 96 | 9539 | 101.68 | 95.02 |
| 95 | 15424 | 101.16 | 92.60 |
| 94 | 21580 | 101.03 | 91.71 |
| 93 | 27943 | 100.25 | 91.56 |
| 92 | 32751 | 99.85 | 92.11 |
| 91 | 34993 | 99.85 | 91.74 |
| 90 | 34648 | 99.62 | 89.29 |
| 89 | 31283 | 99.60 | 91.09 |
| 88 | 27000 | 98.64 | 93.71 |
| 87 | 23559 | 100.25 | 97.84 |
| 86 | 21150 | 98.92 | 103.82 |
| 85 | 19903 | 99.85 | 109.52 |
| 84 | 19123 | 101.42 | 112.06 |
| 83 | 18613 | 100.14 | 113.77 |
| 82 | 17530 | 99.94 | 117.67 |
| 81 | 15812 | 99.36 | 112.67 |
| 80 | 13483 | 100.02 | 113.52 |
| 79 | 11136 | 99.53 | 116.01 |
| 78 | 9528 | 100.64 | 107.46 |
| 77 | 8239 | 100.67 | 110.07 |
| 76 | 6903 | 100.57 | 104.89 |
| 75 | 5719 | 100.46 | 105.88 |
| 74 | 4672 | 99.57 | 101.36 |
| 73 | 3687 | 97.38 | 96.81 |
| 72 | 2934 | 101.78 | 100.63 |
| 71 | 2379 | 97.77 | 101.70 |
| 70 | 1808 | 101.65 | 104.59 |
| 69 | 1401 | 99.19 | 96.59 |
| 68 | 1234 | 99.43 | 104.57 |
| 67 | 927 | 98.94 | 99.43 |
| 66 | 682 | 97.88 | 93.14 |
| 65 | 434 | 93.34 | 98.54 |
| 64 | 276 | 101.54 | 93.06 |
As you can see, the faster pitches are more effective in the strike zone, the off-speed pitches are more effective out of the strike zone. For every speed from 74 MPH to 86 MPH, a higher ratio exists outside the zone. The PITCH f/x data is pretty much confirming what I would expect.
Please donate to the Baseball Musings Pledge Drive.
Posted by David Pinto at
10:28 PM
|
Pitchers
|
TrackBack (0)
As I understand it, Pitch F/X also has the release point and trajectory information. (If it doesn't have trajectory information, it could be inferred from the release-point to capture-point differences.)
Is there a correlation between movement at the same speed? That is, if I sub-divide pitches by "difference between release and capture points (over a minimum e.g. the 6' pitcher's 3-quarter release point to the capture point is the minimum; subtract that and show that)", do the pitches with a bigger difference produce more strikes? do the smaller differences?
