How Vex AI works

Each fighter is described by a statistical profile sourced (in production) from licensed and official feeds: significant strikes landed/absorbed per minute, striking accuracy/defense, takedowns per 15 and accuracy/defense, submission attempts, control time, knockdowns, plus engineered indices for cardio, durability and finishing ability. Context inputs include reach, stance, age, layoff duration, recent form and opponent quality.

Raw stats are normalized to 0–100 within weight class, then blended into nine interpretable category sub-scores: Striking, Wrestling, Grappling, Submission, Cardio, Durability, Physical, Form and Competition.

The composite rating is a transparent weighted sum of the sub-scores. Win probability uses a logistic (Bradley–Terry-style) function of the rating gap:

R   = Σ (weight_c × subscore_c)
P(A) = 1 / (1 + e^( -k × (R_A − R_B) ))

We simulate the bout hundreds-to-thousands of times. Each round carries a finish hazard scaled by the leader's finishing ability against the opponent's durability; unfinished fights go to a decision. Aggregating runs yields method-of-victory distribution (KO/TKO, submission, decision), round momentum, and empirical confidence intervals.

• Uncertainty inflation: probabilities are pulled toward 50% when data is incomplete or the styles are high-variance.
• Hard caps: displayed probabilities are clamped (≈8–92%) — MMA is too chaotic for false certainty.
• Ranges, not points: every prediction ships with a confidence interval and a volatility tag.

Absent stats are shrunk toward the weight-class mean (a Bayesian prior), the data-completeness meter drops accordingly, and imputed fields are flagged. We never silently fabricate a number.

Every pre-fight probability is logged immutably. After results we compute Brier score and log-loss, plot reliability curves, apply Platt/isotonic recalibration on a rolling window, and periodically refit weights. Each change ships as a new Vex AI version with a public accuracy record.

For any prediction we show each category's contribution (weight × skill gap mapped to probability points), the top drivers in plain language, and links back to the underlying stats and sources.