On Saturday, March 13, 2021, at the SABR Virtual Analytics Conference, a panel discussion was held on MLB Statcast Pitch Tracking Methods.
Panelists included Clay Nunnally, MLB Baseball Scientist, and Dr. Paul Hawkins of Hawk-Eye Innovations.
- Video: Click here to watch a replay of the MLB Statcast Pitch Tracking Methods panel on YouTube
- Audio: Click here to listen to the MLB Statcast Pitch Tracking Methods panel (MP3; 29:49)
Here are some highlights:
ON THE HISTORY OF PITCH TRACKING
- Nunnally: “Radar guns were the first tracking tool used in Major League Baseball. They measure the frequency shift of going to get the ball. It looks at the component of velocity of the moving object that is the direction of the gun point. This is why you will see scouts sitting behind the plate rather than at third base. Now this is a tool that can be used to analyze pitches and find who who throws fast and who throws hard and use that as a training tool or an analysis tool for the first time. It gives good feedback back to the pitchers so they can understand not only how fast they throw, but what the difference in velocity is between their fastball and their changeup. Some of the first tools for pitch design was getting this type of feedback back immediately.”
ON MLB’S IMPLEMENTATION OF THE HAWK-EYE SYSTEM IN 2020
- Nunnally: “There are 12 synchronized cameras; it can track pitches, hits, players, umpires, coaches, anything that is on the field. One of the things that is impressive about Hawk-Eye is that because there are 12 cameras, we get a redundancy and (more) accuracy because there are so many inputs to the value of where the ball is during play. The spin orientation is directly studied. The thing we began to understand more is that we can see spin decay during the pitch, and we saw spin deficiency occur. Although we knew this happened, we never had the tools to quantify it until the development of Hawk-Eye. Hawk-Eye helped us see the way to determine pitch release with surprising accuracy.”
ON THE PROCESSING OF DATA
- Hawkins: In regard to whether the ball processing is done in real time or minutes later — there is a live data stream that is blowing out of the system the entire time. The biomechanics and ball tracking is done in real time. In addition to this, another process is called the post-process. However, everything keeps up with each other. For example, the 1-second video that is recorded for the spin off the ball is then evaluated and analyzed in the post-process phase which takes about 10 seconds. But it is all done before the next pitch is thrown.”
ON RELEASE POINT AND SPIN
- Nunnally: “How well can Hawk-Eye measure release point? More accurately than we expected. This is attributed to not just looking at the ball because now we look at the biomechanical signature of the pitcher. The point in space and time where the pitcher’s fingers, and the free flight trajectory of the ball intercept, gives us a good indication of where the ball is released. This is an important Hawk-Eye lesson because it does not use just tracking data to figure these things out, but by using biomechanical, contextual data. Spin is the name of the game for this session … Quantifying the spin and being reliable is different than just watching the spin. Hawk-Eye measures spin by looking at the seams, just as humans measure spin. There are very important elements of contrast and a-symmetry on the ball. In general, this matters because we can measure spin and pitch trajectory separately. Because we can measure these separately, we don’t have to convert them using a model. This is model-independent, which makes it so great.”
ON USING DATA TO IMPROVE PLAYERS
- Hawkins: “The data will definitely progress. It’s always amazing to see how much time and money the ball clubs will put into gathering this data for scouting purposes in assessing whether Player A is better than Player B. … The real purpose of a club, and the heart of the sport, is to make your players better. I believe there is an awful lot of work still yet to be done and opportunities to innovatively think about how the data enables one player to get from one level to a higher better level. My hope is that from a fan’s perspective, it is a lot easier to get behind a team if you view a team that has been together as a group for a lengthy period of time and you have (watched) them improve as a team. Rather than a club that just buys the best talent. I believe there is a real value in that.”
For more information on the 2021 SABR Virtual Analytics Conference, visit SABR.org/analytics.
Transcription assistance from Alyssa Rodarte.
Originally published: June 1, 2021. Last Updated: March 30, 2021.