⚽ World Cup 2026 — AI Prediction Engine LIVE

A self-learning machine-learning model that predicts every match of the 2026 World Cup — and retrains itself after every final whistle. Random forest + Poisson xG, loop-learned ELO, live ESPN data.
🕐 Vietnam (UTC+7)
--:--:--
Source
🏆 Tournament Win Probability — All 48 Teams
← Select a team to see their profile
🔄 Loop Learning — ELO Updates from Actual Results

After every played match, the model updates each team's ELO rating (K=60 WC weight). These updated ratings feed directly into all upcoming match predictions.

🧠 Self-Learning Engine — What the Model Has Learned

Loading…

Type a team name to see their corner history

How this project works

A self-learning World Cup prediction system — model, backend, and live data pipeline.

The model

The learning loop

The live backend

Stack

Python · scikit-learn · SciPy · pandas · vanilla JS (zero front-end dependencies) · GitHub Actions · Render