Orlando Magic general manager John Hammond, left, and president of basketball operations Jeff Weltman are pioneering the use of artificial intelligence in scouting. (AP Photo/John Raoux)

In today’s NBA, competition is no longer constricted within the lines of the hardwood.

From state of the art practice facilities to nutritionists, sleep specialists to cryotherapy chambers, organizations are constantly immersed in an arms race to find the next off the floor innovation that could equate to more wins.

Now, the Orlando Magic are attempting to up the ante once again.

Last month, the Magic announced a new partnership with sports data and intelligence company STATS. They will be the first NBA organization to enlist the AutoSTATS artificial intelligence software for college scouting and data purposes to supplement draft preparation.

STATS released the following in a press release in February:

“AutoSTATS delivers comprehensive player-tracking data directly from video through patented AI and computer vision technology. The new technology gives the Magic exclusive access to the college tracking data currently unavailable at this scale due to the scarce use of in-venue tracking systems.”

While the new partnership is expected to begin immediately, it remains unclear to what extent the software will be used by the Magic and what intent STATS has for further implementation around the league.

STATS is the parent company of SportVu, a live-game player tracking software that uses multiple cameras in the rafters of each NBA arena to provide teams with a plethora of advanced statistics. While SportVu is utilized by the entire NBA, its reach on the collegiate landscape has been more difficult to tackle given the cost and physical logistics associated with it. However, using nothing more than a regular broadcast feed rather than in-venue cameras, AutoSTATS will presumably have the ability to make collegiate data more accessible to NBA and college customers alike.

“Collecting data through broadcast is the future of tracking and we couldn’t be more thrilled to launch our technology through this partnership,” the release continued. “This information can then be applied to their proprietary models to improve player evaluation and development.”

How other coaches and executives, such as Clippers’ head coach Doc Rivers (center) and president of basketball operations Lawrence Frank (right) utilize AI in scouting remains to be seen. (AP Photo/Charles Rex Arbogast)

As with any new technology, there are plenty of questions that remain with the new software that only time will be able to answer.

“Personally, I’m pretty skeptical about how it could work,” one NBA front office and analytics staffer told Forbes. “Because of the different camera angles you get in college arenas, it makes it really difficult to teach a computer how to consistently recognize things. But if it works, that’s super valuable.”

Analytics personnel around the NBA believe AutoSTATS could be useful if past games are back-logged, which they believe will be the case. Eventually, subscribers will be able to utilize a library of data similar to what SportVu provides the NBA, but for current and former college players.

But that’s not where AutoSTATS’ reach ends.

What’s more groundbreaking is the possibility of using the artificial intelligence for player evaluation. According to a release from Business Wire, AutoSTATS will utilize a product called OpenPose which “unlocks new layers of body-pose information, providing a deeper quality of player tracking data like body position, shot form, torque, and other aspects of the game.”

Traditionally, NBA executives have depended on technology as a means to acquire data, which has been used to supplement draft and personnel decisions. The key word here is supplement, as technology has not yet been able to mimic the human element of scouting. Such elements include the oft-mentioned ‘eye test’ as well as chemistry decisions. Now, AutoSTATS threatens to make at least one of those a thing of the past.

Though technology and its place in player evaluation can be a polarizing issue, the potential of something like OpenPose can be a game changer. If proven to be accurate, a long-standing subjective criterion such as shooting mechanics may finally be flipped objective. That’s intriguing for a front office. But like many other industries threatened with automation, that could be bad news for the small community of scouts and others in basketball operations. As front offices have continued to grow around the league, owners will undoubtedly be tasked with choosing whether to keep large staffs or go the way of artificial intelligence.

Even those whose jobs AI would target can’t deny the potential luxury it may afford.

“Would it be able to tell me the angle of their spine, or how much their feet rotated? If so, perfect,” said NBA shooting coach Dave Love, who last worked with the Magic and previously the Cleveland Cavaliers.

Love, who has been credited for making major improvements in the shooting mechanics of players such as Tristan Thompson and Aaron Gordon, would utilize such data but cautions that there’s still an irreplaceable human element that’s equally, if not more, important.

“It supplements. It gives you a better idea WHAT might need to change, but it comes down to the coach to determine how to change.”

More traditional, stubborn executives will scoff at the use of such technology and will be hard to sell on it, like it was getting the same folks to budge from Beta to VHS, VHS to DVD and DVD to digital when it came to watching film. But even the most open-minded and tech savvy front offices will proceed with caution, and rightfully so. To Love’s point, the teaching and human element cannot be underscored enough. It’s the how, not the what, that San Antonio Spurs’ assistant coach Chip Engelland has mastered that’s allowed him to be credited with transforming the shooting mechanics of players such as former Spur Kawhi Leonard.

While the Magic have kept their plans for the software close to the vest, the rest of the league will be watching intently for signs, if any, of how they utilize it. There’s a balance between supplemental and overload of information, and that is a determination solely left to humans to make – at least for now.

STATS did not return a request by Forbes to comment on this story




Please enter your comment!
Please enter your name here