Stage 1: Baseline Ratings
The ten attribute ratings are added and divided by ten.
Explore
Learn the simulation engine rules: attribute matchup advantages, finishing probabilities for KO and submissions, and variance formulas.
Ratings are 0–100 editorial estimates normalized within each division, allowing comparisons across the full roster. The simulator excludes weight class and sex from its calculations. So, for example, Zhang Weili can defeat Francis Ngannou when her normalized profile produces the higher result. The simulations do not predict real-world outcomes of bouts.
The ten attribute ratings are added and divided by ten.
Each attack rating is compared with its paired defense, then multiplied by the listed coefficient. The seven adjustments are added and capped at ±8 points.
A seeded random multiplier from 0.970 to 1.030 represents the variation between a fighter's expected and actual performance.
Average Knockout Power with the higher of Boxing or Kicking, then compare the result with the opponent's Chin.
Average Submissions and Wrestling, then compare the result with the opponent's Wrestling.
Each chance starts at 3%. Add 0.12 percentage points for each offense point above 75 and 0.18 points for each offense point above the defense. Each chance is limited to 0.75%–10%, and the four chances share a 25% cap. A triggered knockout or submission overrides the final score.
If no finish occurs, each fighter's final matchup-adjusted score is multiplied by its seeded variance and the higher score wins. An exact tie is resolved by a deterministic seeded coin flip.