Current Model Performance
View detailed metrics and performance statistics for our prediction models.
KNN Model
Test Accuracy
—
Cross-Val Score
—
Optimal K
—
XGBoost Model
Test Accuracy
—
ROC-AUC Score
—
Trees
1000
Model Analysis Visualizations
Model Comparison
KNN Confusion Matrix
KNN Probability Distribution
Winner Model Accuracy Comparison
Winner Confidence Distribution
KNN Confusion Matrix (Winner)
XGBoost Confusion Matrix (Winner)
About These Models
Model Comparison
Both KNN and XGBoost models are trained on historical Premier League data to predict Over/Under goals and match winners.
View Interactive Predictions →Performance Metrics
The confusion matrices show how accurately each model classifies matches. The probability distributions indicate model confidence in its predictions.
Historical Accuracy
Performance metrics are updated weekly as new matches are played and predictions are validated against actual results.