UFC Fight Night 289 odds: every big line move went toward our model
UFC Fight Night 289 odds movement: Vieira, Brener, Jackson and Bellato all moved toward our market-blind model, plus Till vs Romero and a Cage Warriors flip.
Our prediction model never sees a betting line. It is built that way on purpose: it knows records, rates, ages, layoffs and the level of competition, and it has no idea what anyone is paying for either corner. That makes fight week a clean experiment. The model publishes its number, the market spends nine days moving, and on Saturday morning you can check which way the money walked.
For UFC Fight Night 289, the answer is uncomfortable for anyone who likes to think the market is always ahead. Every big move on the card went toward the model.
The four lines that moved
Photo: Profigrappling / CC BY-SA 4.0
Our consensus line is the average across sportsbooks only, 18 of them on this card as of Saturday morning. The first number we logged for most of these bouts was on September 17. Four fights moved by seven points of implied probability or more once the bookmaker margin is stripped out. Here they are, with the model's number beside them.
| Bout | First line | Now | Market moved | Model |
|---|---|---|---|---|
| Rodolfo Vieira vs Robert Bryczek | -108 | -169 | 50.0% to 60.2% | 68.0% |
| Elves Brener vs Josiah Harrell | +169 | +107 | 35.8% to 46.3% | 52.4% |
| Montel Jackson vs Ricky Simon | -155 | -224 | 58.5% to 66.3% | 66.7% |
| Rodolfo Bellato vs Christian Edwards | -250 | -180 | 68.8% to 61.6% | 51.3% |
Four for four. Three favourites got shorter and one got longer, and in every case the line closed the distance to a number that was published without looking at it.
Rodolfo Vieira opened as a pick'em against Robert Bryczek. The model had him at 68 percent. He is 11-5 as a pro and 6-5 in the UFC, a 19-6-1 professional submission grappler whose whole case is the takedown, and Bryczek defends 71 percent of those in his three UFC fights. A coin flip was always a strange opening price for that matchup, and the market has spent the week agreeing.
Elves Brener is the most interesting of the four. He opened at +169 against Josiah Harrell, a man who is 11-1 and whose one loss is his entire UFC career. The model has had Brener as a slight favourite at 52.4 percent. The market has moved ten and a half points in his direction and still has him as the underdog at +107. If it keeps walking at this pace it arrives at the model's number somewhere around the walkout.
Montel Jackson is the tidy one. The model said 66.7 percent, the market now says 66.3. Jackson is 15-4 and absorbs 1.40 significant strikes a minute in the UFC, which is the kind of defensive number a model trusts and a first line sometimes underrates. Ricky Simon is 22-7-1 and knows the room, but the gap is gone.
Rodolfo Bellato is the move in the other direction. He opened at -250 against Christian Edwards, and the model called it close to even at 51.3 percent. Bellato is 13-3-1 and has landed 5.37 significant strikes a minute in the UFC, but he also absorbs 5.86, and a favourite who takes more than he gives is one the market seems to have lost its nerve on. It has shed seven points in a week.
The drift that went the other way
It would be a better story if every line on the card had followed the model, and it did not. Of the other eight bouts, seven drifted away from the model's number. None of them moved by more than five and a half points.
That is the part I find most interesting. The big moves, the ones where somebody put real conviction into a price, were all in the model's direction. The small ones scattered. I am not going to pretend one card proves anything about who knows more; it is twelve fights, and a model that has never seen a price can be wrong in ways the market is not. But when the only lines that moved hard all moved one way, it is worth saying out loud which way that was.
Where they still disagree
Four fights on this card still carry a gap of more than 12 points after all that movement.
The main event is one of them. Raul Rosas Jr. opened at -180 against Raoni Barcelos and the money has come back toward the veteran, to -156, or about 58 percent. The model has Rosas at 70.6. Rosas is 12-1 and turns 22 next month; Barcelos is 22-5 and 39. The market is paying more respect to Barcelos's 87 percent UFC takedown defense than the model is, and against a fighter who averages 6.10 takedowns per fifteen minutes that is a fair thing to argue about.
The co-main runs the other way. Norma Dumont opened as a slight underdog against Ailin Perez at +111 and has drifted to +134, about 41 percent. The model has Dumont at 23.9 percent, which means it likes Perez far more than even the drifting market does. Perez is 13-2 and averages 4.90 takedowns per fifteen minutes in the UFC.
Brady Hiestand remains the widest gap on the card: +296 against Rinya Nakamura, about 24 percent, against a model number of 40.3. That line barely moved all week, less than a point. Either nobody with an opinion has touched it or everybody who has agrees with the price.
The fourth is Mehemmedeli Osmanli against Ilimbek Akylbek Uulu, two fighters with no UFC results on their records yet. The market has Osmanli at -291, about 71 percent. The model says 58.7, and with no octagon tape on either side, a model built on careers is naturally more cautious than a price.
The European Saturday, market only
Photo: Эркек / CC0
The three cards that go off before the UFC prelims have no model numbers, so this is the market talking to itself. It has still said some loud things.
The biggest move on any of the four cards is on the Cage Warriors 210 prelims. Lewis Byrne, 5-0 as a pro, opened at -160 against Oliver Tero and is now a +128 underdog. That is 18 points of implied probability gone from a fighter with no professional losses, and it is the only favourite across all four cards to flip all the way to underdog. At Oktagon 94, Edmon Avagyan went the opposite way under the Stand and Bang rules, from +153 to -128 against Dominic Schober, a 15-point swing on a man with one professional MMA fight.
At BKFC 94, Darren Till has gone from -189, the first line we logged back on August 27, to -281 against Yoel Romero. Romero is 49. Till is 33. A month of that age gap sinking in has been worth eight points. Karl Thompson flipped from even money to -144 against Andy Thornton on the same card.
And one main event is going the wrong way for its favourite: Christian Eckerlin has drifted from +113 to +139 against David Kozma in Frankfurt, in front of a German crowd, with a German fighter in the favoured corner a week ago and the Czech one in it now.
None of these come with a model check. We will know tonight whether the late money on those cards was smart or just loud, and by Monday we will know whether a model that never looked at a price was the thing the UFC market spent a week catching up to.
Every bout's current line and model number is on the UFC Fight Night 289 event page, and the pick'em is open until the first bell. This is information about what the market believes, not betting advice.