Expected goals (xG), explained without the hype
xG is the most quoted number in football analytics — and the most misunderstood. Here is what the model actually measures, where it breaks, and how to use it responsibly.
Football data analytics: long-form articles, YouTube walkthroughs and open code. Learn the numbers behind the game — and build with them.
Long-form pieces on xG, datasets and modelling decisions — process over hot takes, with the code that backs them up.
Browse the blog →Step-by-step videos: data pipelines, notebooks and match analysis you can follow along and reuse in your own work.
Watch on YouTube →Notebooks and small tools, open on GitHub. Clone them, run them, break them and adapt them to your club or project.
See the code →Deep dives into football data: methods, sources and decisions — written to be useful, not just readable.
xG is the most quoted number in football analytics — and the most misunderstood. Here is what the model actually measures, where it breaks, and how to use it responsibly.
Open data is the fastest way to start. This walkthrough goes from raw JSON events to a shot map in a handful of steps — the same pipeline pattern you will reuse with every public dataset.
You do not need a paid provider to start learning football analytics. These five free sources cover events, results, xG, odds and historical tables — with honest notes on the trade-offs of each.
Pipelines, notebooks and match-analysis sessions — recorded as we build.
Subscribe on YouTube and you will catch the first pipeline build as soon as it lands.
Subscribe on YouTubeEverything we build for the videos ends up here: pipelines, feature code and small utilities you can reuse.
The first notebooks from the walkthroughs will be published on GitHub. Follow the profile to get them the day they land.
Follow on GitHubA dataset, a model you want to validate, a video idea, or a project for your club or company — tell us and we will get back to you within 48 hours.