Score insights
Historical score patterns
Every square is equally available, but final scores do not land evenly. Explore the completed-game baselines behind our modeled square chances.
67,000+ finished games
Our first historical model uses ten completed seasons from established public datasets. It powers the pregame baseline while live estimates also account for the current score and time remaining.
What these percentages mean
Each percentage is the historical chance that a completed game landed in that board cell. It is not a prediction of which team will win and it is not a sportsbook line.
Final-score history is a strong starting point because it captures sport-specific scoring patterns. Live modeling is a separate problem: game state, clock, score differential, and scoring pace all matter. We identify live estimates as experimental until each sport passes out-of-sample calibration testing.
See the patterns in play.
Create a free board and watch the current winning cell move as the score changes.