Our metrics
Win Probability
The model's estimated chance that a team wins a game, from the share of simulated games it wins. A 70% favorite should still lose about 3 games in 10.
How we use it
Each game is simulated drive by drive 2,000 times. The home team's win probability is the share of simulations it wins outright (ties count as not a win).
Win probability is a statement about many games, not one. It is judged by calibration: across all games we gave 70%, the favorite should win about 70% of the time. Our track record shows how close we come.
Where to see it
Related terms
- Calibration
- Whether forecast probabilities match how often things happen: of all the teams given a 70% chance, about 70% should win.
- Brier Score
- A measure of probability forecasts: the average squared gap between the forecast and what happened (1 if it happened, 0 if not). Lower is better; always saying 50% scores 0.25.
- Monte Carlo Simulation
- Estimating the chance of an outcome by playing it out many times with random draws and counting how often it happens.
- Home Field Advantage
- The edge a team gets from playing at home. Across the NFL it has historically been worth a few points per game and has shrunk in recent seasons.