Elite pitchers usually have two things in common: great pitches and the ability to command those great pitches. With better technology and understanding, the baseball community has gotten much better at determining what makes pitches great. However, our ability to measure command has not evolved in the same way.
Well, in order to understand this, we first need to differentiate command from control. Control is the general ability of a pitcher to throw strikes. It can be evaluated by simple metrics like zone percentage and walk rate. Command, on the other hand, measures the ability of a pitcher to locate his pitches exactly where he wants them to go whether that be in the zone or not.
At STATS, we go through every pitch of every game to determine actual intent. Our team usually starts with four key variables – count, pitch-type, the pitcher’s trends and the catcher’s mitt location and body position during his setup. STATS then factors in over 20 other key variables, including things like the hitter, game situation and the runners on base.
From there, the intent of a pitch is narrowed to one of 13 zones, which correspond with a concept. For example, a pitch that is supposed to be located on the outside edge of the plate is known as “black or better” because the intent is to paint the corner or be just off the plate.
Let’s look at some examples of how STATS Pitch Intent data is revolutionizing the way that pitchers are being evaluated:
Here are two sliders from Yankees right-hander Masahiro Tanaka that end up in the same location, low and inside and out of the strike zone.
But notice the different situations before the pitches.
In the first video, Tanaka has a 1-1 count on Yolmer Sanchez, a relatively low-power hitter at the bottom of the White Sox lineup. The pitch is a slider, so it moves in toward a left-handed hitter. Generally, this would not be used as a chase-pitch off the outside edge because it would start out of the zone. Given that information, the game situation, and the position of the catcher, Tanaka misses on a pitch that is supposed to be thrown for a strike on the outside part of the plate and Sanchez singles to right field.
This breakdown also illustrates why the catcher’s mitt is not a good indicator of intent on its own. This pitch is intended to be a strike, yet the catcher sets up with his mitt below the batter’s knees for some reason (anticipating a block, just out of habit, etc.). Take a look one more time.
Now, let’s look at the second video’s situation.
This time, Tanaka has a 2-2 count on Chicago’s Yonder Alonso with a runner on second and two outs. Tanaka has a pitch to work with, so he can afford to throw the slider for a ball in hopes of getting a swing-and-miss strikeout. The catcher sets up inside for a common back foot slider location, and Tanaka hits the spot, unlike in the previous video, nearly getting Alonso to chase.
This is the beauty of looking at intent rather than results. If you just looked at a heat map of Tanaka’s slider against left-handed batters, you would see that he throws them in two common places – for strikes low and outside and balls low and inside.
Both sliders shown in the videos, represented by those two black dots in the heat map, ended up in that low and inside zone and seemingly fit in with Tanaka’s pitching trends. However, with Pitch Intent data, we can distinguish between those two black dots, one as a missed location and the other as a hit location.
Once that data is collected, we can compare it with the results of the pitches to generate a plethora of more accurate metrics. In terms of simply evaluating command, we have developed a Command+ rating that compares the command of every pitcher, sortable by pitch-type, with the league average (set at 100). If you look at the table below, you can see that Tanaka’s Command+ rating for all his pitches is actually the best in the league at 132 or 32% better than the MLB average.
|Pitcher||# of Pitches||Command+|
Now that STATS has been collecting this Pitch Intent data for a few years, teams can track how a pitcher’s Command+ rating has improved (or gotten worse) for each pitch in his arsenal.
Take Lucas Giolito. The breakout White Sox pitcher went from being one of the worst pitchers in the league last year to winning AL Pitcher of the Month for May. Using Command+, we can illustrate that a lot of Giolito’s improvement can be attributed to his improved command.
|Pitch Type||2018 Median Command+||2019 Median Command+|
Giolito’s command has gotten better on multiple pitches, but he has commanded his changeup particularly well this season, improving 12% relative to the league average. Combine that with the fact that he now throws his changeup 23.4% of the time compared to 15.8% last season, and you start to get a clearer picture of why he has made such a big leap.
Beyond evaluating pitchers, STATS Pitch Intent data can also be used to evaluate other aspects of the game, like a catcher’s ability to frame pitches or how pitchers are attacking a hitter during a slump.
STATS is always working to develop new tools to change how we see sports through data, and for evaluating pitcher command, Pitch Intent is our new darling.