Most fights on a card, model probability and market-implied probability land within a few points of each other. The interesting fights are the ones where they don't.
On UFC 330, three bouts show a meaningful gap between statistical models and the current market — and each gap has a different cause.
Three kinds of disagreement
The first gap is a name-value fade: the market prices a popular veteran above what declining output and absorbed strikes justify. Models are ruthless about age curves; casual money is not.
The second is an information gap — a camp change and a move up in weight that the market has priced heavily but models treat as neutral until fight-night data exists.
The third is the classic grappling blind spot: takedown-heavy fighters are systematically underpriced against strikers with untested defensive wrestling, because knockouts are memorable and control time is not.
