The Gap #4: the book and the crowd disagree on McGregor, Conor vs Holloway, Max

Prediction markets and crypto books are pricing the same fights very differently this week, and the McGregor gap is the one demanding your attention first.

The fights

The lead story is Conor McGregor against Max Holloway on 12 July. Polymarket has McGregor's win probability at 34.7%, which translates to a fair price of 2.89. Stake is offering 3.20 — a 4.8 percentage-point gap between what the crowd thinks and what the book is charging. Put simply, the book is paying you as if McGregor wins roughly 31 in every 100 fights, while the prediction market says it's closer to 35. That spread is wide enough to be meaningful, though McGregor's layoff and the calibre of opponent mean the probability itself carries real uncertainty — the gap is genuine, but so is the risk on either side of that estimate.

Two fights on the 27 June card show the next-largest gaps, both sitting at 3.4 percentage points. Julius Walker against Abdul-Rakhman Yakhyaev has Walker priced at 5.20 on Stake, against a crowd fair price of 4.59 — a meaningful difference when you're working at those odds, because a half-point move on a fiver represents real expected-value swing. Eric Nolan against Farman Hasanov also clears the threshold, with Stake at 2.49 against a fair price of 2.40, though at those shorter odds the absolute cash difference per unit staked is more modest.

Two further fights sit at 2.3 percentage points — Sharaputdin Magomedov versus Michel Pereira, and Nazim Sadykhov versus Matheus Camilo. Worth noting on both: the book price is actually slightly below the crowd fair price in each case, meaning Polymarket is the less generous source here. Neither clears a level we'd treat as actionable, and we'd rather call that plainly than dress it up.

The football

Argentina versus Austria at the World Cup kicks off later today and carries a 3.2-point gap: Polymarket at 66.7% against Duelbits' implied 63.4%, with the book sitting at 1.53 against a crowd fair price of 1.50. At those short odds the gap is real in percentage terms but slim in decimal terms — three cents on the price. It clears our threshold, just, though Argentina at 1.53 on a knockout-stage match offers little room for variance to work in your favour over a small sample.

Mexico against Czechia on 25 June shows a 2.7-point gap with Mexico at 1.97 on Duelbits against a fair price of 1.96 — the probabilities diverge, but the decimal prices have almost converged entirely, which tells you most of the gap is noise at this level of precision. Japan against Sweden on the same date shows 2.4 points, Norway against Senegal 2.4 points, and Belgium against New Zealand 2.0 points exactly. All four of those football markets carry gaps too small to act on with confidence.

The skips

Belgium to beat New Zealand sits at exactly 2.0 percentage points — the minimum we track — and the book price of 1.21 sits below the crowd fair price of 1.26, meaning Stake is actually the stingier source. That is not an edge. Japan versus Sweden and Norway versus Senegal both read similarly: the numbers diverge just enough to print a gap, not enough to justify a bet. Brunno Ferreira against Ikram Aliskerov also lands at 2.0 points with a crowd fair price of 3.37 versus a book price of 3.40 — a whisker, nothing more.

Odds shift from the moment this digest is captured to the moment you place a bet, so the price at the book when you click is the only price that matters.