A Blog by Jonathan Low

 

Jul 1, 2023

How Baseball Scouts Use AI To Draft Players

AI is increasingly playing a role in evaluating performance, including likelihood of future injuries based on biomechanics. JL 

Lindsay Scott reports in the Wall Street Journal:

Major League Baseball is introducing analysis of player potential via artificial intelligence. The league has partnered with a biomechanics company that documents a prospect’s movement patterns using two iPhone cameras. It translates images captured into metrics that quantify elements of player movement. It believes the data it generates can detect player’s flaws, forecast their potential and flag potential for injury. “We have metrics on kinematic sequence, stride length, ball contact timing. We also have very early injury warning detection. If you have too much arm flare, there may be potential overload on the elbow, which can lead to Tommy John surgery.”

The most important thing a baseball team can do ahead of the amateur draft is predict future success for developing young ballplayers. Scouting has changed drastically since an explosion of data and technology entered the game a decade ago, and now Major League Baseball is introducing an even more high-tech tool: analysis of player potential via artificial intelligence.

The league has partnered with Uplift Labs, a biomechanics company that says it can document a prospect’s specific movement patterns using just two iPhone cameras. The setup was available for use in evaluating prospects who agree to participate at the MLB draft combine last week in Arizona. 

Uplift says it uses artificial intelligence to translate the images captured by the phone cameras into metrics that can quantify elements of player movement. It believes the data it generates can detect player’s flaws, forecast their potential and, possibly, flag their potential for injury.

 

“We have metrics on things like kinematic sequence, stride length, ball contact timing,” said Sukemasa Kabayama, who founded Uplift in 2017. “At the same time, we also have this new kind of very early injury warning detection. Let’s say if you have too much of an arm flare, you know there may be potential overload on the elbow, which can unfortunately lead to Tommy John surgery.” 

The integration with Uplift demonstrates the league’s growing interest in centralizing some elements of research and development. As many clubs invest in similar technology and information, MLB has been considering ways to provide that data on a universal basis. 

“Biomechanics and the analysis associated with it is something that we know is a pretty significant piece of where the game is headed,” said Bill Francis, MLB’s senior director of baseball operations. “Traditionally, it has been very hard to do at scale because of the expensive hardware.” 

The use of Uplift was optional for players at the combine. Agents say there is a divide in who may benefit from submitting biomechanical data pre-draft and who could be hurt by it.

The league and players’ union have implemented signing bonus protections for any player who submits to a pre-draft physical, but the top players in any draft class typically opt out of the process. Under the new policy, if a player submits to a pre-draft physical, he is ensured that he will receive at least 75% of the slot value as a signing bonus, or become a free agent. 

If a player is already projected to go near the top of the draft, MRI data or movement data may be used by a team to find potential red flags in the player’s athletic profile, which could be used to negotiate a smaller bonus for the player as he enters professional baseball. 

The league has partnered with Uplift Labs, a biomechanics company that says it can document a prospect’s specific movement patterns using just two iPhone cameras. PHOTO: UPLIFT LABS

But some player agents believe that movement data could help players who are likely to go in later rounds of the draft, showing deeper skill sets that may not be immediately apparent in the player’s high school or college results. Baseball’s information age has presented a consistent dilemma to those who aspire to play the game at a professional level: The same information that could be used to turn you into a better player could also be used to devalue you.  

“I think whenever something is introduced, it’s natural to wonder why, and to proceed with caution,” said player agent Tom Koehler, a former major-league pitcher. “The real question becomes, ‘Why is it being used?’ Is it being used as a barometer to see how these guys develop throughout their career and how to prevent injuries moving forward? Or is it being used to hurt draft position based on how somebody’s body works?” 

Baseball scouts suddenly have a lot more information in their search for the mythical “five-tool” player who has speed, fielding prowess, can hit for average and power and possesses arm strength. Gone are the days when teams relied on a scout’s career’s worth of anecdata to determine how the player might perform at the big leagues. 

In the early 2010s, teams began researching the pitching or hitting characteristics that lead to on-field success. Pitchers have always thought about the spin on their pitches, for instance, but once technology was able to quantify “spin rate,” teams were able to study it and learn just how important it is to a pitcher’s success. Once they learned the true value of spin rate, they were able to target players who may have had a poor strikeout rate in college, but a good spin rate on their fastball, which can be harnessed in development to make him a more effective pitcher. 

The use of biomechanics is the second true wave of baseball’s innovation age. In just a few years, teams have gone from building “biomechanics labs” that use complicated systems that require a player to wear minimal clothing and attach sensors to various parts of their bodies to build a digital portrait of how the body is moving. This is typically known as “markered” technology. 

But the ideal system for tracking biomechanics would be in markerless, camera-based technology like Uplift, which can be trained on a player in a game situation rather than a lab. The way a player moves on an indoor mound, throwing into a net is hardly a perfect data set when it comes to determining how the player will move when he is bursting with adrenaline while facing a live batter in a major-league game.

A sample of an Uplift pitching report. PHOTO: UPLIFT LABS

“In our limited testing, Uplift shows huge promise for clinical markerless motion capture in sport biomechanics,” said Kyle Boddy, the founder of Driveline Baseball. “Their team is one with a lot of experience in not only motion capture but wearable technologies, showing that they have a broad set of skills in the biomechanics field.”

 

Uplift’s approach is to attempt to democratize access to markerless biomechanics technology, and MLB is betting that the convenience of the product will give it an easy foray into offering biomechanical analysis at the draft combine. Competitive advantages tend to last for only a matter of months in professional baseball, but the information race has now moved far beyond on-field results, and into a player’s core athletic traits. 

All teams will have access to the same Uplift data at the combine, but there is still an arm’s race within baseball to determine how to get the most out of a player’s movement profile. One team may have determined that they are skilled at developing players with a certain subsection of skills—making players with those skills more valuable to them in the long run. Another team may understand that there is value in a movement tendency, but that as an organization they have yet to figure out how to make the most of this ability. 

“I liken it to the radar gun, a tracking technology that is quite ubiquitous in baseball now,” Kabayama said. “It’s a tool that assists people in the player performance or evaluation process.” 

The use of biomechanical technology presents a big ask for potential future major-league players: How much potentially damaging or beneficial information about yourself would you give potential future employers to determine whether they want to hire you or not?

0 comments:

Post a Comment