A.I. Won’t Solve the NBA’s Newest Crisis
Predicting injury risks is just half the battle
Welcome back to Club Sportico, where we break down the intersection of sports and money—with an extra bit of humor and opinion. Today, we’re talking about predicting the future.1
Once again, the healthiest team in the NBA Playoffs took home the Larry O’Brien Trophy this year. Before OKC claimed its crown, a single injury—a torn Achilles tendon—took down Damian Lillard, Jayson Tatum, and Tyrese Haliburton in the postseason, after the league saw zero such injuries all of last year. The previous high across an entire season was four; now the record is seven.
Naturally, A.I. is on the case.
In addition to a panel of experts tasked with studying the problem, NBA commissioner Adam Silver said this week that he’s optimistic about the possibility of artificial intelligence models that can ingest every frame of video from every game a player has played to detect patterns and determine what circumstances lead to Achilles injuries.
It’s not the first time Silver has turned to A.I. to help solve the league’s challenges. The commissioner has pointed to improved computer vision helping to speed up officiating reviews and cited monitoring models that could identify suspicious betting behavior.
Player health is a particular area of interest when it comes to technological developments. Two years ago, it was sprained ankles plaguing the league. I wrote about a similar initiative to the league’s current Achilles research that took place then, ranging from video review of injuries to floor studies and even an investment in a new ankle brace startup. I learned that since 2019, the NBA has used machine learning to study connections between injuries and player movement. As much as anything, reversion to the mean may have quieted that particular storm.
I’m not dismissing A.I. It probably can help in this case. Similar tools are already being used to replace simple metrics like distance traveled in soccer or pitch counts in baseball with more holistic strain measurements. As early as 2022, experts were wondering when we’d have systems that would identify injury risk levels before something snapped, Minority Report-style.
But implementing that science on the hardwood will prove tricky. This is a new form of moneyball we’re talking about. First, data was used to determine who should be on the team. Then it drove decision-makers to push for a faster-paced style of play that appears to be breaking athletes down at a faster rate. Now, stars could be asked to trust computers to make medical determinations on whether they should be on the court at all.
I’m not sure who would be most upset if a model told LeBron James he ought to take a seat in crunch time of a big game: the fans who paid to see him or James himself. I don’t need superintelligence to predict how messy that situation would get.
In the meantime, quicker fixes could come from equipment improvement, training changes, and more oversight at the youth level. Experts believe wear and tear built up from a young age is contributing to the problem, especially with players still in their prime now seeing their Achilles tendons give out. But then again, we run into a similar problem. Will a teenager really abide by an algorithm?
Club Sportico is a community organized by Sportico, a digital media company launched in 2020 to cover the business side of sports. You can read breaking news, smart analysis, and in-depth features from Eben, Jacob and their colleagues at Sportico.com, and listen to the Sporticast podcast wherever you get your audio. Contact us at club@sportico.com.
Today’s artwork was created by Jacob because our normal creative mind, Matt Palacio, is (a) at a bachelor party and (b) still recovering from his own Achilles tendon tear, so giving him this assignment felt a bit too soon.