UFC 333: Merab Dvalishvili vs. Petr Yan (Bantamweight, Main Card)

Logged 19 Oct 2026, 19:01 UTC · Brier 0.1388

• Yan won the first meeting by unanimous decision in December 2025 and enters as champion, which is the single strongest signal; rematch losers repeat their loss more often than not.
• Merab is a former champion with a comparable record (21-5 vs 20-5), so this is a high-level competitive rematch rather than a fluke-driven mismatch, keeping his win probability well above a longshot.
• Both fighters are coming off a ~10-month layoff with no 2026 fights before this one, so ring rust and uncertainty apply roughly symmetrically and slightly widen the distribution rather than favor either side.
• A small share of outcomes (no contests ~1.4% in the division, plus rare draws or postponements) resolve 50-50 rather than YES, trimming the effective YES probability slightly below Merab's raw win chance.
• No injury or availability information could be established, so no adjustment is made for that; this is a genuine information gap close to fight night.

What would change it: Credible news of an injury, bad weight cut, or decline for either fighter in fight week, or evidence that Yan's UFC 323 win was fluky (e.g., injury to Merab mid-fight), would move the estimate materially.

-3.3 probability points from the market at generation · med confidence · Polymarket

Base rate: In UFC immediate title rematches, the loser of the first fight wins the rematch roughly 35-40% of the time historically; the retrieved facts confirm the prior result (Yan UD over Merab at UFC 323) and a 1.4% no-contest rate in the division, but no rematch-repeat rate was retrieved, so the rate is stated from knowledge of the class.

Chosen by Crossbook's published selection rule (auto-v1), not by hand.

Formed blind at 34.0%, without sight of the price; published at 37.3%, the market’s 40.5% moved 50% of the way toward it by fixed rule. None of the cited facts could be shown to be one the market had not already priced. That check is recorded, not enforced.