Photo via Pexels
Every Premier League club now employs a data analytics team. But how does it actually work, what does it change, and which clubs are using it most effectively?
When Brighton's data team identified a relatively unknown striker in the Belgian Pro League as an undervalued asset — bought for £18 million and sold three years later for £105 million — the football world sat up and paid attention. That was in 2021. In 2026, every Premier League club has a data analytics team, every transfer decision is informed by statistical modelling, and AI-assisted video analysis tools are as standard as a training ground gym. But what does this transformation actually look like, and are all clubs using data equally effectively?
The most consequential application of data analytics in modern football is recruitment. Clubs use statistical models to identify players who outperform their immediate context — a striker scoring at a poor club whose xG-outperformance suggests he will produce at a higher level, or a defensive midfielder whose pressing metrics rank in the top 5% of European players despite operating in a minor league. These models allow clubs to find value that traditional scouting networks, limited by human bandwidth and geographic reach, would miss.
Brighton's model — the most celebrated — combines event data (every touch, pass, and shot from every game across dozens of leagues) with physical tracking data (GPS, sprint distances, acceleration profiles) to build a picture of a player that extends far beyond goals and assists. Their success rate in identifying players who perform above their transfer fee is the best in the Premier League over the past five seasons.
Beyond recruitment, clubs use AI-assisted video analysis to prepare for opponents. Tools that can identify a team's pressing triggers, their most common set-piece routines, or the defensive positions a particular player is weakest at — all of which previously required dozens of hours of human video analysis — can now be generated automatically within minutes of a game ending. Some clubs use these tools to prepare opposition reports for the press and media; the full depth of the analysis is kept internal.
Brighton remain the benchmark for recruitment analytics. Liverpool's data science team, the first in English football to hire PhD-level statisticians as full-time employees, is the most sophisticated for in-game modelling. Manchester City have invested heavily in physical performance analytics, using GPS data to predict and prevent soft tissue injuries at a rate that is the envy of the division. The gap between the data-rich clubs and those still relying primarily on traditional scouting is narrowing but has not closed: the technology is available to all, but the expertise to use it well is still scarce.
The most honest data practitioners in football are quick to emphasise that analytics enhances rather than replaces human judgment. A model can identify a player who is statistically undervalued; it cannot assess whether his personality fits the dressing room, whether he will perform when the crowd turns hostile, or whether a manager's specific tactical demands suit his instinctive game. The best clubs use data to narrow the shortlist; the final decision still belongs to people.
Arsenal sit top of the Premier League after ten gameweeks, but Manchester City are breathing down their necks. We assess every contender's chances.
Erling Haaland leads the Premier League golden boot race with 11 goals, but Bukayo Saka and Mohamed Salah are mounting a credible challenge. Here's the full top-10.
Ten gameweeks in, we name the Premier League's eleven best-performing players of the 2026-27 season — with one significant controversy.