I started the year talking about how machine learning is helping doctors more accurately predict when people will die. I’m going to end the year revisiting our common machine learning theme, but this time talking about how it is impacting pro sports. If you’ve read the book Moneyball by Michael Lewis or saw the movie starring Brad Pitt, you know the Oakland A’s and their manager Billy Beane were the first team to use predictive analytics to win games in the early 2000s. Today, virtually every major league baseball team is using analytics and we now see it making its way to other professional sports league like the NFL. Some teams, like the Baltimore Ravens, are embracing more than others as was recently witnessed in their win over the Dallas Cowboys.
Last weekend I happened to be doing some work on my laptop, or possibly getting my fantasy football lineup submitted for the weekend, and an article popped up titled, “How machine learning is unlocking the secrets of human movement – and reshaping pro sports,” by Bill Briggs. We’re not talking about your traditional analytics on deciding when to throw a curveball versus a changeup or when to go for a two-point conversion versus kicking an extra point, they are literally measuring how athletes run, jump, and move, to predict success in their chosen profession.
The company featured in this article/marketing piece for Microsoft is Peak Performance Project (P3) in Santa Barbara, CA. The founder, Dr. Marcus Elliott, served as the first director of sport science in MLB for the Seattle Mariners, and in the same capacity for the New England Patriots as the NFL’s first team to employ someone to introduce new training techniques to reduce injuries.
Most of this article focused on the NBA, but P3 has clients from the NFL, MLB, international soccer, track and field and more. They have been working with players from the last six NBA drafts and have a database of more than 600 current and former players. Dr. Elliot explains, “We’re not interested in how high they jump or how fast they accelerate. We’re interested in the mechanics of how they jump, how they accelerate or decelerate. It’s helping us unluck the secrets of human movement.” Each year experts try to determine which players will make the successful jump to the NBA, which will become superstars, role players, or fail to make a roster. P3 is taking it a step further by using thousands of biometric data points captured to predict who will retire early, which players will suffer career ending injuries, and which lesser ranked prospects will exceed expectations and reach stardom.
As a Milwaukee Bucks fan I’ve seen both ends of the spectrum as I watched my favorite team draft and immediately trade Dirk Nowitzki to the Dallas Mavericks where he went on to a successful 21-year career, and more recently draft a young little known player from Greece, Giannis Antetokounmpo, in the 15th round of the 2013 NBA draft who is now one of the faces of the NBA. I’d be curious what machine learning would say about either player today, especially Giannis. I watch the violent nature with which he jumps or takes one long euro-step and I’m amazed he isn’t more injury prone. Here is a link to a YouTube video from a recent game https://www.youtube.com/watch?v=lUKNQUVWoL4.
Many in the NBA make a visit to P3 a part of their annual summer routine for re-testing and to determine if their movement patterns have gained asymmetries that could cause injury, or to make sure they are properly maintaining their physical systems to give them a competitive edge. P3 successfully identified Dallas Maverick Luka Doncic as having the ability to stop quickly in a pre-draft workout. Elliot said based on their finding Doncic and 2017-2018 NBA MVP James Harden, “were in the same player branch.” That was significant because he said this of James Harden, “he has a better stopping or braking system than anybody we’ve ever assessed in the NBA.”
I’m not sure the average reader of this blog will get this reference, but P3 is proving despite what Allen Iverson said back in 2002 that it is about “Practice” and current 76er players seem to agree. See link to AI’s original practice rant https://www.espn.com/sportsnation/story/_/id/15479405/what-forgotten-allen-iverson-practice-rant.
While practice certainly isn’t what most people look forward to, at Extract we like to give our instance of machine learning plenty of it so we can produce better results for our clients in their game time situations.
About the Author: Troy Burke
With 30 years of experience providing clients with stellar service and strategic solutions for growth and development, Troy is committed to ensuring his customers receive the highest quality solution, training and support with every implementation. He frequently speaks on the topic of redaction and is actively involved with National Association of Court Management, Property Records Industry Association and several other government organizations.