UFC 331 pick'em: our readers are backing 13 straight losses

One hundred and seventy nine people have filled in part of the UFC 331 pick'em so far, 529 individual picks across twelve fights, and there is a shape in it that neither the sportsbooks nor our own model share.

Our readers are backing every veteran on this card who is currently losing. All three of them. Between Marlon Vera, Giga Chikadze and Tai Tuivasa there are thirteen consecutive defeats, and in every one of those three fights the reader majority or something close to it is on the man carrying the streak.

Photo: Mass Communication Specialist 1st Class Russell R. Rhodes Jr. / Public domain

Vera is the biggest gap on the card, by a distance

Marlon Vera vs Charles Jourdain is the clearest case. Our readers have it 22 picks to 8 for Vera, which is 73.3 percent. The market, stripped of vig, has Vera at 34.8 percent. That is a 38.5 point disagreement, and it is the widest crowd versus market split on any fight on this card.

The market has its reasons and they are on the record. Vera has lost four in a row: David Martinez by unanimous decision in February, Aiemann Zahabi by split decision last October, Deiveson Figueiredo in August 2024, Sean O'Malley over five rounds at UFC 299. His last win was Pedro Munhoz in August 2023. Jourdain, meanwhile, has won three straight, two of them by submission inside two rounds.

Our model sits with the market, not the readers: 38.1 percent for Vera, which is three points off the board and thirty five off the room. Vera is still the number 10 ranked bantamweight in the UFC, and I suspect that is doing a lot of the work in the pick'em. A ranking is the last thing to go.

Chikadze is the second biggest, and the model does not buy it either

Giga Chikadze vs Joanderson Brito splits our readers almost exactly down the middle, 13 to 12, which is 52 percent for Chikadze. The market has him at 22.5 percent, drifting the wrong way all week from plus 274 to plus 328.

Three straight losses is the record behind that price: Kevin Vallejos stopped him in the second round last December, David Onama and Arnold Allen took decisions before that, and his last win was Alex Caceres in August 2023. Our model gives him 27.3 percent, modestly above the board and nowhere near the coinflip our readers have made of it.

So on the two biggest crowd versus market gaps on the card, the model is on the market's side of the argument. Which makes the third one interesting.

On Tuivasa, the readers and the model are the ones agreeing

Photo: NAPARAZZI / CC BY-SA 2.0

Tai Tuivasa vs Robelis Despaigne is the one where the pattern flips. Our readers have Tuivasa at 32.1 percent, 9 picks to 19, which sounds like a fade until you look at the number he is being faded from. The market has him at 18.8 percent. The model has him at 53.4.

Tuivasa has lost six straight, which is more than the other two combined and the longest active skid on the card. On the pure record he is the least defensible pick of the three, and the readers have him closest to the market. They are more willing to back Vera on four losses and Chikadze on three than Tuivasa on six, which is a perfectly rational way for a room of people to behave and produces exactly the wrong answer if our model is right.

Twelve fights and this is the largest model versus market gap of any of them at 34.6 points. It is the one place on this card where the pick'em crowd and the sportsbooks are aligned and the model is off on its own.

And the man nobody wants to pick

The inverse of all this is Ryan Gandra, who is 3-0 since arriving from the Contender Series and has finished all three inside the first round, including Zach Reese at UFC 329 in July.

The market has Gandra at 80.5 percent against Ozzy Diaz. Our model has him at 79.0. Our readers have him at 59.1 percent, 13 picks to 9. That is a 21 point discount on a fighter who has not been out of the first round since he got here, and it is the largest crowd versus market gap on the card in the other direction.

Put those two ends together and the shape is not "our readers back underdogs." They are pricing familiarity. A ranked veteran on a four fight skid gets more reader support than his price deserves; a three fight winner most of the room has never watched gets less. That is a bias, and it is not a stupid one, because names usually carry information. It just happens to be the specific bias that a model trained on records and ages is built to have no opinion about.

What the night settles

Vera, Chikadze and Tuivasa all fight on the UFC 331 prelims and early prelims. If all three lose, the market was right and our readers were sentimental. If all three win, everybody who built a number this week has some explaining to do, including us.

The most likely outcome is a split, and the interesting version of a split is the one where Tuivasa is the winner. Early prelims start at 5:30 p.m. ET from Crypto.com Arena in Los Angeles, and the picks are still open until the first bell.