Los 40 USA
Sign in to commentAPP
spainSPAINchileCHILEcolombiaCOLOMBIAusaUSAmexicoMEXICOlatin usaLATIN USAamericaAMERICA

TECH IN SPORTS

How AI could replace referees in soccer

Tired of the person with the whistle being biased against your team? Well, maybe you would prefer it if a machine was left to do all the decision making.

Update:
Tired of the person with the whistle being biased against your team? Well, maybe you would prefer it if a machine was left to do all the decision making.

Fans across the world believe that those humans officiating the games have something against their team. The expletives rain down from the stands on the referees who are perceived to ‘have it in for us.’ Well, the future may see them replaced by robots, so says an industry expert.

Could robots replace referees?

Artificial intelligence has already made its presence felt in the realm of officiating, notably through the utilization of VAR (Video Assistant Referee) and goal-line technology, but its influence is expected to significantly expand in the future. With the increasing availability of high-quality data, it could become conceivable for AI-powered machines to officiate matches, eliminating the requirement for an on-field human referee.

“Computer vision will become increasingly effective in the coming years, and the number of cameras on the pitch will only continue to rise,” said Aldo Comi, the chief executive of the leading global football analytics provider Soccerment, during an interview with the PA news agency. He continued, “The amount of data that is tagged, and the quality of the models trained with that data will increase exponentially, and as a result, we will have AI models capable of making refereeing decisions based on what they observe on the pitch. So, we might reach a point where we no longer need a referee.

“Before that, we may still have a referee, but perhaps the linesmen will be the first to vanish from the game. The referee may be connected to a virtual assistant, guiding them to make better decisions. Ultimately, in the span of 20 or 30 years, the referee will probably be purely an AI. I’m not suggesting this is necessarily positive; I’m merely saying it’s likely to occur.”

The rise of AI and machine learning in football won’t be confined to officiating; it is already a part of the game. The use of data analytics has assisted clubs like Brighton and Brentford in disrupting the established Premier League hierarchy by identifying high-quality signings that they have later sold for significant profits.

AI to provide “better understanding” with big data

However, the advancement in data integration may lead to managers using virtual coaches to aid them in team selection and tactical decisions.

“AI can become a source of new perspectives on the game in the coming years,” added Comi. “If you provide AI with enough high-quality data, you can have a virtual assistant with a better understanding of what’s happening on the pitch. By analyzing data with AI, you can train models to better predict future events, such as understanding the probabilities of what will occur in the next five or ten minutes.

“Through predictive analytics, AI can provide prescriptive analytics, suggesting, ‘Things are expected to progress this way, and to improve your chances, I have 10 ideas.’ These suggestions from AI will be filtered by the assistant coach, who will then convey the information to the manager, and it will be up to the manager to decide whether to accept them. AI will exist, but not as a replacement for professionals; it will serve as high-quality support. Clubs that embrace this technology will likely outperform those that do not.”

Comi, whose company serves several Serie A and Serie B clubs, acknowledges that people will need time to trust AI, but there is already evidence of its effectiveness.

“It will take time to build trust, but much like data analytics, there are enough success stories to show that you can outperform others. We’ve seen it with Brentford and Brighton. The advantage that AI can provide is many times more significant than traditional data analytics.