AI crowd · UFC Fight Night: Prochazka vs. Stirling
All 10 AI forecasters pick Stirling at UFC Fight Night: Prochazka vs. Stirling
Every AI forecaster took Navajo Stirling over Jiri Prochazka, at an average of 60%. The crowd split down the middle on Dan Hooker vs Brian Ortega and went against the odds in three fights.
The setup
10 AI forecasters, played by three AI models. Each one reads the same fact sheet on every fighter (record, size, career stats, UFC fights and this week's news) and gives a winner and a chance for every fight. None of them sees the betting odds, our model or each other's picks, and none can search the web. The crowd's number is the average of their chances.
The models
- Claude Sonnet 5.5Anthropic · 4 forecasters
- Claude Opus 5.5Anthropic · 3 forecasters
- Claude Fable 5.2Anthropic · 3 forecasters
Every one of the AI crowd's 10 forecasters picked Navajo Stirling to beat Jiri Prochazka on Saturday, but none of them by a wide margin. Their average chance for Stirling was 60%, almost exactly the betting market's 61%.
The crowd is 10 forecasters played by Claude Sonnet 5.5, Claude Opus 5.5 and Claude Fable 5.2, reading one fact sheet blind to the odds.
On the last card, UFC 334: Gane vs. Hokit, the crowd got 9 of 12 winners right. The betting favourites also went 9 of 12, and our model was right on 6 of the 8 fights it rated. The crowd's worst miss was Bilal Hasan, whom it gave 75% before Luis Gurule beat him on a three-round decision.
Most sure, least sure
The crowd's strongest call was Ikram Aliskerov over Jared Cannonier: 10 of 10 forecasters, at 77%. That is a little above the odds, which give Aliskerov 72%. Jake Matthews (65%), Diyar Nurgozhay (64%) and Dom Mar Fan (62%) were also unanimous picks.
Dan Hooker vs Brian Ortega came out dead even, five forecasters each way and an average of 50%. The split ran along model lines. All three Claude Fable 5.2 forecasters took Hooker, at 59%, while the Opus and Sonnet forecasters leaned Ortega.
One Claude Fable 5.2 forecaster pointed to size: "Hooker has four inches of height and six of reach on Ortega, who is debuting at lightweight after a 455-day layoff." A Claude Sonnet 5.5 forecaster looked at Hooker's recent results instead: "Hooker has lost three straight by stoppage. Ortega has a long layoff and an unproven lightweight move but should have the grappling edge." The betting market is far less torn, with Hooker at 63%.
Where the crowd and the odds part ways
In three fights, the crowd picked against the betting market's favourite. The biggest gap is at heavyweight. Eight of 10 forecasters took Tallison Teixeira over Shamil Gaziev, though only at 52%, while the odds make Gaziev a 64% favourite.
Our model sides with the market, at 62% for Gaziev. A Claude Opus 5.5 forecaster made the case for Teixeira: "Younger and much longer at 83-inch reach; Gaziev was knocked down three times in his last three fights and fades late."
The other two disagreements are close to even on both sides. The crowd gave Aljamain Sterling 53% against Kevin Vallejos, whom the odds favour at 54%. It gave Shoaib Yousaf 55% against John Mitchell, the market's 56% favourite.
Opus and Fable go their own way
Claude Opus 5.5 was the dissenter on Sterling vs Vallejos. Two of its three forecasters took Vallejos, for an average of 52%.
Opus also leaned Santiago Luna over Aleksandre Topuria, two of three at 52%. Fable backed Topuria three of three, Sonnet split two and two, and our model has Topuria at 67%.
Beyond Hooker, Fable broke from the pack twice more. Two of its three forecasters took Amir Albazi over Alessandro Costa, at 52%, and John Mitchell over Shoaib Yousaf, at 51%. Every Sonnet and Opus forecaster went the other way in both fights.
| Fight | AI crowd | Our model | Betting market |
|---|---|---|---|
| Prochazka vs Stirling | Stirling 60% | 🔒 | Stirling 61% |
| Sterling vs Vallejos | Sterling 53% | 🔒 | Vallejos 54% |
| Hooker vs Ortega | Even | 🔒 | Hooker 63% |
| Horiguchi vs Almabayev | Horiguchi 57% | 🔒 | Horiguchi 64% |
| Cannonier vs Aliskerov | Aliskerov 77% | 🔒 | Aliskerov 72% |
| Topuria vs Luna | Topuria 54% | Topuria 67% | Topuria 72% |
| Gaziev vs Teixeira | Teixeira 52% | Gaziev 62% | Gaziev 64% |
| Albazi vs Costa | Costa 55% | Costa 54% | Even |
| Matthews vs Abdullayev | Matthews 65% | Matthews 62% | Matthews 68% |
| Motta vs Fan | Fan 62% | Fan 69% | Fan 61% |
| Nurgozhay vs Lorenz | Nurgozhay 64% | Nurgozhay 76% | Nurgozhay 69% |
| Yousaf vs Mitchell | Yousaf 55% | — | Mitchell 56% |
Our model and the betting market as of Nov 21, 3:00 AM ET.
Every forecaster, fight by fight
Each dot is one AI forecaster, placed by the chance it gave its pick. Left of the middle line it picked the red-corner fighter, right of it the blue corner. Dots pile up in 5-point steps, Claude models above the line and GPT models below it.
Model by model
Model by model
Each model's own call, from its forecasters given no angle, next to the whole crowd. The number is the chance it gave its pick; red means it picked the red-corner fighter, blue the blue corner, and the deeper the colour, the surer it was.
| Fight | AI crowd | Opus 5.5 | Sonnet 5.5 | Fable 5.2 |
|---|---|---|---|---|
| Prochazka v Stirling | 60 for Stirling | 63 for Stirling | 59 for Stirling | 58 for Stirling |
| Sterling v Vallejos | 54 for Sterling | 52 for Vallejos | 57 for Sterling | 54 for Sterling |
| Hooker v Ortega | 50 for neither | 53 for Ortega | 53 for Ortega | 59 for Hooker |
| Horiguchi v Almabayev | 57 for Horiguchi | 55 for Horiguchi | 57 for Horiguchi | 61 for Horiguchi |
| Cannonier v Aliskerov | 77 for Aliskerov | 76 for Aliskerov | 76 for Aliskerov | 79 for Aliskerov |
| Topuria v Luna | 54 for Topuria | 52 for Luna | 53 for Topuria | 61 for Topuria |
| Gaziev v Teixeira | 52 for Teixeira | 55 for Teixeira | 52 for Gaziev | 53 for Teixeira |
| Albazi v Costa | 55 for Costa | 55 for Costa | 60 for Costa | 52 for Albazi |
| Matthews v Abdullayev | 65 for Matthews | 63 for Matthews | 70 for Matthews | 62 for Matthews |
| Motta v Fan | 62 for Fan | 62 for Fan | 63 for Fan | 61 for Fan |
| Nurgozhay v Lorenz | 64 for Nurgozhay | 63 for Nurgozhay | 66 for Nurgozhay | 62 for Nurgozhay |
| Yousaf v Mitchell | 55 for Yousaf | 55 for Yousaf | 59 for Yousaf | 51 for Mitchell |
How the AI crowd works
The crowd is 10 AI forecasters, played by Claude Opus 5.5, Claude Sonnet 5.5 and Claude Fable 5.2 from Anthropic. None of them was told how to weigh the fights: each got only the fact sheet and the question.
All of them read the same fact sheet: each fighter's record, size, career stats and UFC fights from our database, plus this week's news (the reports, and what the fighters and their coaches said), gathered once for each fight. None of them saw the betting odds, our model or each other's picks, and none could search the web. Half of them saw each fight with the fighters in the opposite order, so the order they're listed in can't tilt the result.
Each forecaster picked a winner in every fight and gave a chance. The crowd's number is the average of those chances. It's an experiment in how AI reads fights, not betting advice.
The AI crowd: every card so far, and how it works →
UFC Fight Night: Prochazka vs. Stirling: the full card, odds and results →