Keeping score: Has Premier League ceded too much power to maths geeks?

As data and AI remake the league Asia adores, fans can only hope the analytics revolution does not rob the game of its romance.

- Bhavan Jaipragas

The world's most-followed domestic football league returns this weekend, supercharged by data analytics like never before.

The English Premier League (EPL), Singapore and much of Asia's favourite sports league, may not be the Major League Baseball (MLB) or the National Basketball Association (NBA), where big data has long reigned and maths whizzes, who could just as easily be running algorithms at Goldman Sachs, shape everything from recruitment to in-game tactics. But it is by no means a bystander in this numbers revolution.

As former Liverpool director of research Ian Graham recounts in his 2024 book How to Win the Premier League: The Inside Story Of Football's Data Revolution, since 2016 all 20 clubs have had access to granular, sophisticated data: raw positional information on every player and the ball, recorded every 40 milliseconds for all 380 league games. The same algorithms that power facial recognition and instant translation are used to parse this trove.

Liverpool, EPL champions last season and now one of the sport's data-science powerhouses, go further, with the ability to track the position of each player's feet, ankles, hips, shoulders, elbows, hands, eyes and even ears in real time for the full 90 minutes. This "pose data" can then be fed into artificial intelligence models and analysts can "hallucinate" future player movements.

As in other sports that embraced the quant revolution, the biggest impact has been on talent scouting. Clubs now use data to model how a player would fit into a squad before they are signed. This summer, alongside the familiar bidding wars for star players, the quieter contest has been over the growing influence of data teams.