How AI Is Changing Sports Commentary and Recruiting
By Kallie Polgrean
It had been a decade in the making. Hockey player Sean Durzi was in his prime at age nineteen, a talented defenseman hoping to be drafted in the National Hockey League. On the night of the draft, Durzi would wait eagerly for his name to be called and a team to claim him. But his name was never called.
It was a shock and a major ego bruise. He had been ranked one of the top forty prospects, yet he had not been chosen. The ranking process had been completed by an artificial intelligence system, yet NHL coaches declined to listen to the technology’s conclusion that Durzi was a talented player. It would be another year until Durzi was finally drafted to play for the Maple Leafs in June of 2018.
“It’s a dream come true,” Durzi wrote on his Twitter account only a few months ago, after his drafting.
According to Christopher Boucher, professional hockey teams made a mistake when they denied a spot on their roster for Sean Durzi. Boucher is the manager of hockey analytics at a company called Sportlogiq.
Sportlogiq is responsible for the AI system that ranked Durzi among the top forty prospects. The sports-centered business employs intelligent technology to make the scouting process more seamless, and allow coaches to view all of the statistics on a player in comparison to other players.
The system uses data of an athlete’s grades, highlight footage, teacher and coach recommendations, as well as other relevant information that is usually included in a regular college application. As the technology surveys more and more athletes, it improves in ranking accuracy and speed.
“Now he’s much higher ranked and will cost them an earlier draft pick,” Boucher remarked prior to Durzi’s official signing with the Maple Leafs.
If NHL managers and coaches had heeded the advice of Sportlogiq’s AI back in 2017, perhaps they would have experienced the blossoming of a very successful hockey player. Maybe the rookie Durzi would have led a team to the Stanley Cup or set records in defense.
Artificial intelligence has entered nearly every aspect of our lives, so it is no surprise that it now has a role in how we watch sports games and how prospects for professional teams are recruited.
In recent years, tech companies such as IBM, Sportlogiq, and television networks specialized in broadcasting athletics have become interested in incorporating AI with recruiting processes and sports commentary.
Professional leagues worldwide have begun using Sportlogiq’s system of ranking athletes based on their statistics and prowess. AI has the capacity to make or break somebody’s professional sports career, depending on how it analyzes their individual data along with every other hopeful prospect.
“The goal is for colleges and pro teams to have access to data from kids they might never see otherwise,” says Cam Potter, who works with Brooklyn Dynamics in analyzing data for many professional baseball teams.
Brooklyn Dynamics — which also uses AI to rank MLB players — and Sportlogiq are essentially working in the same sphere of AI relating to sports, but the two companies have influences in different sports. Regardless of what league it assists in the recruiting process, artificial intelligence has undoubtedly altered the landscape of professional sports.
In perhaps an even more astounding way, commentary for sports games is also being ‘infringed’ upon by technology. Researchers in artificial intelligence are attempting to replicate a human’s ability to narrate a sports game onto a computer system.
“We have the computer do things like watch soccer games until it figures out how it can generate commentary the same way people do,” says John Smith, fellow at IBM.
Other research papers have been published trying to do the same thing. The task is not a simple one. Programmers must give the computer the ability to form coherent and exciting sentences that captivate the audience. Vivid adjectives and verbs are necessary when describing sports lest the game becomes dull.
IBM’s Watson, the computer system famous for beating humans in a game of jeopardy, was programmed to provide highlights of the Wimbledon tennis tournament.
“New for 2018, Enhanced AI-powered automated video highlights for Wimbledon fans. As a learning system, Watson has been taught to better recognise player emotion increasing the quality of the output, while also increasing speed in turnaround time by 15 minutes,” reported IBM in an online statement.
The company also offered watchers a live chat feature with an artificially intelligent bot on Facebook messenger allowing fans to access scores, new developments in matches, and player information.
It is more cost efficient to use AI to create highlight reels and form commentary because machines don’t require monthly or weekly wages for completing a task. With AI, a television network must pay an upfront fee for the technology, but afterwards there is no paycheck necessary.
Companies like IBM and Sportlogiq will try and try, but artificial intelligence will likely never fully replace human commentators and draft pickers. These occupations require sentience and emotional understanding of players, as well as an admiration for the game. Artificial intelligence may excel in some areas, but it can’t surpass humans in everything.
“When a cheap laptop beats the smartest mathematicians at some tasks but even a supercomputer with 16,000 CPUs can’t beat a child at others, you can tell that humans and computers are not just more or less powerful than each other — they’re categorically different,” says PayPal founder Peter Thiel.
Sources:
