Technology marches on. Now, thanks to MLB's Advance Media division, we have the chance to get a lot more detailed analysis of the pitches thrown to, and hit by players in the majors last season. For example, that three-run homer Jeff Salazar swatted out with two outs in the ninth against San Franciso? It left Brad Hennesey's hand at 89.2 mph, and arrived over the plate at 83.5 mph. It was also 7.9 inches away from where the ball would have arrived without spin: however, this dropped it nicely into Salazar's wheelhouse, from where it was promptly dispatched it into the right-center bleachers. If you look in the dictionary beside "hanging breaking ball", you'll likely see its picture.
This kind of detail would have been unthinkable only a couple of seasons ago, but is now increasingly possible, thanks to a triangulation process between a number of cameras in certain ballpark: one above home, one above first and another in center field. This allows the path of the ball to be tracked, and the information is available for all to see in MLB's GameDay application. Early on last year, they also used to show the strike-zone as well, but this information was quietly erased, presumably because the umpires did not appreciate their work being "marked" real-time, and the results shown to anyone with an Internet connection. This is, to some extent fair, since the strike zone was set for each batter by the operator based on the image from the center-field camera. Some guy hired off monster.com could therefore make a pro umpire look very, very bad indeed.
However, the raw data is also available online [here explains how to locate it, should anyone want to put it through further analysis], and there is some really good work being done with it. For example, Josh Kalk has taken the data, crunched the numbers and used it to compile a set of player cards, both for hitters and pitchers - we'll be using his data as the basis for this series of articles. You can use the speed and break (horizontal and vertical) of a ball to work out the kind of pitch thrown - John Walsh wrote about this over at The Hardball Times: his graph for Brandon Webb is shown below:
Now, there are a number of limitations regarding this data. Firstly, it doesn't seem like we have pitch f/x data for all games. Brandon Webb threw 3,436 pitches in 2007, but Kalk's data shows only 1,583, less than half, I'm not sure why this is missing: apparently the technology was supposed to be in place for 28 of the 30 parks this year, but perhaps it was only implemented.as the season went on. The information also is difficult to break down by factors like runners on base, which are certainly an aspect that will affect pitch-selection [with a speedy runner on first, a pitcher will tend to throw more fastballs]. The same goes for game situation: if the tying or go-ahead run is at the plate or on-base, that would seem likely to affect the choice
That aside, it is still possible to draw some overall conclusions abour pitch selection. We'll hold over beginning the analysis of specific D-back pitchers and hitters until next time, but we again turn to the invaluable work of John Walsh, for an overview of what pitches get thrown in specific counts.
Cnt | FB% | SL% | CB% | CU% -----+------+------+------+----- 0-0 | 0.63 | 0.15 | 0.12 | 0.09 0-1 | 0.52 | 0.20 | 0.15 | 0.12 0-2 | 0.51 | 0.21 | 0.18 | 0.09 1-0 | 0.63 | 0.15 | 0.08 | 0.13 1-1 | 0.53 | 0.19 | 0.13 | 0.14 1-2 | 0.48 | 0.22 | 0.19 | 0.11 2-0 | 0.75 | 0.11 | 0.04 | 0.10 2-1 | 0.64 | 0.16 | 0.08 | 0.13 2-2 | 0.51 | 0.21 | 0.16 | 0.12 3-0 | 0.84 | 0.05 | 0.03 | 0.08 3-1 | 0.80 | 0.10 | 0.03 | 0.07 3-2 | 0.66 | 0.17 | 0.08 | 0.09 -----+------+------+------+----- All | 0.59 | 0.17 | 0.12 | 0.11
To explain the above, the first column is the count of balls and strikes, and the remaining show the fraction of fastballs, sliders, curves and change-ups thrown for that count. For example, 63% of first pitches were fast balls, 15% sliders, 12% curves and 9% change-ups. Some general thoughts based on the above stats and also Walsh's other analysis:
- There's not that much difference between 0-0 and 1-0. A few more changeups, a few less curves. A first-pitch strike, however, and you're significantly likely to see a breaking pitch.
- If the count reaches 2-0, however, the hitter can look dead-red, with a three-in-four chance of a fastball.
- Any at-bat with a three-ball count has a good chance of a fastball. Even at 3-2, odds are still about two-in-three, better than average.
- The slider and, even more so, the curve are both put-away pitches, most often seen with two strikes.
- However, the fastball is still king. There's only one count [one ball, two strikes] where it does not represent the majority of pitches.
- There is a clear difference if the pitcher and batter are opposing arms [e.g. southpaw pitcher vs. rightie batter] Fastballs and sliders are close to the same, but you are much more likely to see a change-up [16%-5%] and less likely to get a slider [14%-21%].
- Some results don't differ all that much: the percentage of pitches called as balls varies only from 36% to 40% across the board. Fastballs are more likely than average to be a called strike, but much less likely (only 6% of the time) to be a swinging strike.
- Fast balls are hit more often, and when they are, the damage is usually more. SLG% off fastballs is 50 points more than off curves.
We'll get into the more specific analysis of the D-backs next time. But one possible future enhancement with staggering potential, was described by Mike Jacobs, head of the company behind Pitch f/x, Sportsvision. "We could send live pitch data to an Xbox 360 or a simple gaming application on a phone where the user can try to hit live pitches." This would take armchair participation to a whole new level: you can grab your Wii, stand in front of your TV, try to hit that Brandon Webb sinker...and ground harmlessly out to the infield, just like Nomar [0-for-17 lifetime vs. our ace]. Sounds like fun to me.