MODEL ACCURACY

Every win probability we publish is calibrated against real ATP & WTA matches. When we say 60%, it should win about 60% of the time — here's the proof, good or bad.

Calibration — predicted vs. reality

We sort every graded match into 10 buckets by the win probability we gave the favorite, then check what fraction actually won. A perfectly honest model sits on the diagonal: the dots should hug the dashed line.

Our model (dot size = matches) Perfect calibration
PredictedWon (actual)MissMatches

What these numbers mean

Brier score is the average squared error between the probability we gave and what happened (lower is better; 0.25 = a coin flip, 0 = perfect). Straight-up accuracy is how often the side we favored actually won. Calibration is the most important: across the whole range, our stated probabilities match real-world frequencies.

These are measured on the same surface-aware power-rating + serve/return blend we use to price every match on the site — not a curated subset.

This is a calibration record, not a betting record. It measures whether our win probabilities are honest — not profit, units, ROI or a pick history. On the main tours the betting market is extremely sharp, so a well-calibrated model is the foundation, not a promise of beating the closing line. For entertainment and informational purposes only. 21+. Please bet responsibly.