About xGRAPM

xGRAPM is part analytics product, part experimentation ground. The premise is simple: the data exists to say more interesting things about football than most public analysis actually does. Box scores get repackaged, xG gets cited, and a handful of players get credited with things that were really team effects. I wanted to see whether better methods — borrowed from basketball analytics and adapted for the sport — could produce something more honest.

That might be a wrong bet. It's possible we're already doing the best we can with the data available. But the beauty of public analysis is that anyone can try, and you rarely find out what's possible without someone testing the edges of what's been done. xGRAPM is that test.

The goal isn't to produce numbers for their own sake. It's to give the kind of fan who already thinks carefully about football — who uses xG, debates tactics, makes predictive arguments — access to tools that let them make stronger, better-supported claims. That fan exists. This is built for them.

About Me

I'm Albert — a data engineer at Meta. I grew up in Nigeria, went to college at Carleton in Northfield, Minnesota, and have been working in data ever since. The through-line in most of what I care about is something like: what can we actually know, and how do we know it?

Three things pull at that question for me. Data, because we now have more of it than we know what to do with — the bottleneck has shifted from collection to interpretation, and that shift matters. Games, because they offer something rare: a closed system with a clear objective, where you can actually measure whether an idea works. I mean this broadly — soccer, chess, basketball, Overwatch, FPL all qualify. And AI, because the question of how machines learn to behave intelligently inside those kinds of bounded systems is one of the more interesting open problems around.

xGRAPM sits at the overlap. It's a data problem inside a game, and the methods drawing from machine learning. I built it because it felt like the natural thing to build.

Get in Touch

If you have questions, feedback, or just want to talk football and data, find me on Twitter / X or LinkedIn.