UFC 334 odds movement: seven weeks of money toward Josh Hokit, and where the model disagrees

On September 23, three days after the sportsbooks posted UFC 334, Ciryl Gane was -290 against Josh Hokit. With the bookmakers' margin removed, that is a 70.8 percent favorite. This morning he is -220 across 18 sportsbooks, or 65.8 percent, the lowest he has been since the fight was priced.

That's five points in seven weeks, all of it in one direction. Fifteen of the 18 books in the consensus have moved toward Hokit since their own opening number. Three haven't moved. None has moved toward Gane.

The rest of the card tells a different story, and in five fights our model, which never sees a betting line, disagrees with the market by more than the main event's whole drift. Here's every line on Saturday's card, bout by bout (main card 9 p.m. ET, 02:00 UTC Sunday). As always, none of this is betting advice. The prices are information.

The main event, week by week

For six weeks the line leaked slowly, a cent or two at a time. Then it went from -246 on Friday to -228 on Saturday, the biggest one-day move since the line settled, and it has kept going since. Across the 18 sportsbooks, opening prices on Gane ran from -300 to -225. Their latest quotes sit between -245 and -200.

Our model gives Gane 61.3 percent, so even after seven weeks the market is still 4.5 points more confident in the champion than the model is. The model's biggest inputs both lean Hokit: he is seven years younger, 29 to 36, and he wrestles. I wrote up what a Hokit win would look like separately.

The co-main that won't move

Kayla Harrison against Amanda Nunes has done nothing. Harrison opened at -199 with five books on September 20 and is -193 with 18 today, 63.1 percent with the margin out. Once five or more books were quoting it, the consensus never left a band between -184 and -202. Twelve books have shaded toward Nunes, one toward Harrison, and the median move is 0.6 points.

The model gives Harrison 67.5 percent, more than the market does. It's the only one of the two title fights where the model sits on the favorite's side. The full case is in the odds piece on that fight.

Every bout, open and now

"Open" below is the first consensus with at least five sportsbooks, so a single early book can't set it. Percentages are for the first-named fighter, margin removed.

Adrian Yanez vs Juan Díaz has no consensus line yet.

The money that agreed with the model

Three fights moved hard, and in all three the market moved toward the model rather than away from it.

Photo: Sexto Round / CC BY 3.0

Caio Borralho opened as a near coin flip against Yousri Belgaroui, -115 with a single book on October 27 and -138 once five had it. He's -157 now. Ten of 18 sportsbooks have shortened him and none has lengthened him. The model had him at 59.5 percent before most of that happened, and the market is now within about a point of it.

Stephen Thompson against Charles Radtke is the same shape: -115 on the first book on November 2, -123 with five, -146 today. Ten books have moved toward Thompson, none away. Here the market has run a little past the model, 56.7 percent to 53.9.

Donte Johnson is the biggest mover on the card. His fight with Cesar Almeida only exists because Baisangur Susurkaev came out of it on October 27, and the line opened at -152 once five books had it, on November 3. It's -193 now, with 12 of 17 books shortening Johnson. The market has gone from 57.1 percent to 63.1. The model is still well ahead of it, at 75.6.

Where the model and market disagree

These are the four gaps worth reading twice.

Bilal Hasan vs Luis Gurule. The market has Bilal Hasan at -984, an 86.7 percent favorite, and that line has barely moved in two weeks. Our model gives him 65.1 percent. That's a 21.6-point gap, the widest on the card.

Drew Dober vs Chris Duncan. Drew Dober has drifted from +115 to +136, so the market agrees he's the underdog and has grown more sure of it. The model is far more sure: it has Dober at 24.8 percent against the market's 40.6, which makes Chris Duncan the model's biggest favorite on the prelims after Johnson.

Uros Medic vs Kevin Holland. The market has Uros Medic at -156, 58.4 percent. The model calls it a coin flip, 49.9. The consensus touched -176 on November 5 and has come back since, and 12 of 18 books now sit longer on Medic than where they opened. The market is drifting toward Kevin Holland, slowly, in the direction the model already points.

Macy Chiasson vs Bia Mesquita. Bia Mesquita has shortened from -220 to -243, and 13 of 17 books have moved her way. The model goes further still: Macy Chiasson 25.9 percent against the market's 32.1.

The only flip

One fight has changed favorites. Evan Elder opened at -119 against Samuel Sanches and was still -121 on Sunday. Overnight the line crossed: at 08:50 UTC this morning it read Elder -108, Sanches -113. Fifteen of 17 books have moved away from Elder since their own opening number, and none toward him. The model had this one at 48.2 percent for Elder all along.

It's a small fight at the bottom of the early prelims, and a one-point move either way. But it's the only bout on the card where the market has ended up on the other side of where it started, and it landed exactly where the model was sitting.

The pattern

Put the card together and the market has done two different things. In the main event it has spent seven weeks taking money off the favorite, and it's still more confident than the model. Everywhere it moved hard on the undercard, it moved toward the model. Borralho, Thompson, Johnson, Mesquita and Sanches all went the model's way.

The four widest gaps left are Hasan, Dober, Medic and Johnson. If the market keeps doing what it has done for two weeks, those are the lines to watch move by Saturday.