Elo rating system
Match similarly rated teams and update ratings with a predictable Elo calculation.
Probability of winning
For Team A with rating Rₐ playing Team B with rating Rᵦ, we estimate Team A's probability of winning as:
The result is always a number between 0 and 1.
Rating change
After a match, the rating change is sensitivity K multiplied by the difference between the actual outcome S (1 = win, 0 = loss), and E.
The result is positive for a win and negative for a loss. A tie is treated as S = 0.5.
Sensitivity and provisional ratings
- Choose low sensitivity for K = 25, standard for K = 50, or high for K = 75.
- During your configured provisional matchup count, K doubles so a new rating settles faster.
- A team's total change in one round is capped at three times its applicable K value.
- A forfeit is an official loss and carries the same rating weight as another decisive result.
Pools and display
Ratings are based on the frozen rating entering the round. In N-team rotation, a pool is the group of teams whose round results are evaluated together; it may span multiple physical courts when a manager configures that manually. Projected changes wait until every required score in that pool is filled.
Full matches use one round-wide result group instead of treating each physical court as a separate pool. Rating changes appear only after all required round scores are filled.