About Gregory Everett Link to heading

I’m Gregory, a PhD student at the University of Southampton studying how AI can be helpful in football, not just for predicting who’ll win the league, but for solving problems that clubs face every day.

How I Got Here Link to heading

I fell into this area almost by accident. During my computer science degree, I found myself drawn to machine learning modules. As a football fan, I began to wonder if there were ways that the vast amount of data surrounding the sport could be utilised more effectively. It turns out there were, and that became the subject of my PhD.

What Drives My Research Link to heading

What fascinates me most is how football clubs make decisions. They’re dealing with millions of pounds worth of player assets, trying to balance performance with injury risk, and often working with incomplete information. It’s precisely the sort of messy, high-stakes problem that AI should be helping with.

My research focuses on treating football teams as multi-agent systems - essentially, groups of intelligent agents (players) who need to coordinate effectively. I employ techniques such as reinforcement learning and graph neural networks to understand how teams collaborate and how individual player contributions impact overall performance.

One project I’m particularly proud of involves a team selection model that accounts for player injury risks. The model has shown that it could reduce first-team injuries by around 13% whilst cutting the money clubs inefficiently spend on injured players by 11% and maintaining on-pitch team performance. For a Premier League club, that’s potentially millions saved and, more importantly, player careers protected.

I’m also working on making advanced analytics accessible to smaller clubs. Many insights require expensive tracking data that only clubs in the top leagues can afford. I’ve developed methods to extract similar information from basic event data, which somewhat levels the playing field for the clubs without access to tracking data.

Industry & Teaching Link to heading

The industry collaboration side has been crucial. I’ve worked with Sentient Sports throughout my PhD, which keeps me grounded in real-world applications rather than just academic theory. It’s one thing to publish a paper; it’s another to see your work being used by clubs to make better decisions.

I’ve also run undergraduate lab sessions for modules including Intelligent Agents. I’ve found that explaining complex concepts to someone who’s encountering them for the first time is clarifying, as it forces you to truly understand what you’re doing.

My Philosophy Link to heading

What drives me is the belief that AI works best when it augments human expertise rather than replacing it. Football people understand the game in ways that no algorithm ever will. But with the right tools, AI can provide valuable insights and evidence, helping decision-makers make more informed, confident choices.

I’m always interested in discussing this work with others, whether you’re from academia, industry, or just curious about where sports analytics is heading.

Email: gae1g17@soton.ac.uk