Pitch velocity has been the first thing we know about nearly any pitcher since the radar gun came around in 1954. But, what if the way we’ve been thinking about pitch speed is at best inconsistent, and at worst, wrong? Dave Cameron wrote an article on FanGraphs yesterday talking about the way StatCast has changed our measurement of a pitched ball’s speed. Essentially, it may boil down to the use of radar over cameras.
PITCHF/x, the previous tracking system MLB used in its ballparks, worked by training a camera on a defined portion of a pitch’s trajectory, usually at 55 feet from home plate. By tracking the time a ball took traveling a known distance, the camera has the ball’s speed in feet per second, and that’s converted to miles per hour. The problem with this is twofold. First, different MLB players have different release points, and secondly, at the 55-foot mark a ball has already begun to decelerate from its maximum speed.
Pitch speed is at its maximum almost immediately after the ball leaves the pitcher’s hand (let’s call this moment T0) and after that point, the pitch will lose speed as it travels toward home plate. Taller pitchers will naturally have release points closer to home plate, and so T0 will be closer to PITCHF/x’s 55-foot mark. These pitchers then benefit from having their pitches recorded faster because the ball hasn’t traveled as far before reaching the measurement point, and therefore hasn’t lost as much velocity as a shorter pitcher might have.
TrackMan, a component of StatCast’s overall measurement system, generates a radar matrix across the entire field of play. The most famous application of this radar is exit velocity, but the matrix has also helped us better track pitch speed.
TrackMan’s radar doesn’t focus on a single spot in a trajectory, but rather the trajectory itself, identifying the point at which a ball is at its maximum velocity, or again, T0. This compensates for many pitchers who now appear to have “lost” a mile or so per hour.
As I said in Tuesday’s recap, CC seemed to be throwing harder than he did in 2016, tapping 93 mph multiple times through the night, and overall averaging 1.4 mph more on his fastball than last year. It’s unusual for 36-year-olds with physical health concerns to add velocity to their pitches, but more on that in a second. Sabathia’s slider was also harder than in 2016.
In the Jays’ first game of the season, Estrada averaged a neat 90 mph, up 1.9 mph from 2016. Estrada’s curve ball, very much his third best pitch, was also clocked 1.6 mph harder than last season. Marco was a red flag for me as soon as my dad texted me telling me how much harder he was throwing, as Marco’s injury concern is even greater than Sabathia’s, and Estrada has made a living throwing soft already.
Noah Syndergaard was taken out of his start after six innings (and ho hum, seven Ks) against the Braves. Terry Collins pulled his starter due to a blister on Thor’s middle finger. Now, I know how good Syndergaard is, but a blister bad enough to merit removal from a baseball game should cause some change in a pitcher’s performance, right? Right. Syndergaard also threw harder, his fastball up 0.6 mph and his slider, a pitch where most of the stress is placed directly on the middle finger, was also up 0.8 mph over 2016.
Three pitchers, each making one start, still isn’t a great indicator of a trend, so I started playing around with FanGraphs’ velocity tracking, which has switched to the StatCast measurements. Stephen Strasburg was up from his 2016 velocity. Kendall Graveman was up. Lance McCullers was up. Clayton Kershaw of all people gained 1.2mph on his slider and threw his fastball exactly as hard as 2016. Obviously, there’s a sample size issue, but a pattern is starting to emerge.
If the new StatCast measurements are for real, and we’re going to see a league-wide uptick in velocity, or rather, in how velocity is reported, there are three benefits to the Yankees. The first is the likely future opportunity to market more 105+ mph not-strikes. The two more serious ones have to do with pitcher development. A more consistent reading of pitch velocity allows the team to generate a more realistic, controlled comparison between prospects.
More valid pitch speed recordings eliminate one of the variables outside the control of matched prospects, which can help the team get a clearer picture of who is best suited for a rotation spot, a promotion, and so on. As well, with advances in understanding the concept of pitch velocity, having a more accurate and reliable measure of actual velocity (or VA) means your calculation of the delta of VA and perceived velocity (or VP) is going to be more accurate and reliable. This will help give pitching coaches better baselines to work with as they attempt to maximize the delta of VP - VA with each pitcher.
Stats courtesy of FanGraphs