An example of how to create a xG model using R and Wyscout event data
A football Expected Goals (xG) prediction model leveraging advanced machine learning techniques.
Fun with Rocket League tracking data
An Expected Goals (xG) model built with StatsBomb open data. The project covers data scraping, cleaning, feature engineering, model training, and evaluation to estimate goal probabilities from shot events.
Python toolkit that turns football match-event data (WhoScored/Opta) into tactical visualizations, xG/xT models, and automated PDF match reports
A machine learning project that builds an Expected Goals (xG) model using StatsBomb event data. Includes feature engineering (distance, angle), logistic regression modeling, and visualizations such as xG heatmaps and player/team comparisons.
Football Analytics Intelligence Platform (FAIP)
xG model on real StatsBomb data (4,309 shots, 3 competitions): tuned logistic regression vs gradient boosting, 5-fold CV, bootstrap significance testing, benchmarked against StatsBomb's official xG.
AI-powered Over/Under 2.5 football predictions. Runs automatically every day via GitHub Actions — predicts before the match, grades after. Full transparency, zero paywalls.
Add a description, image, and links to the xg-model topic page so that developers can more easily learn about it.
To associate your repository with the xg-model topic, visit your repo's landing page and select "manage topics."