AI crowd · UFC Fight Night: Cortes Acosta vs. Walker
All 10 AI forecasters back Cortes Acosta at UFC Fight Night: Cortes Acosta vs. Walker
Every AI forecaster picked Waldo Cortes Acosta to beat Johnny Walker, with an average chance of 63%. The crowd also backed Gabriel Lorenco, the betting market's underdog.
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 5.1Anthropic · 3 forecasters
All 10 AI forecasters picked Waldo Cortes Acosta to beat Johnny Walker in Saturday's five-round heavyweight main event. Their average chance was 63%, just under the betting market's 66% and our model's 65%. The crowd is 10 AI forecasters from Anthropic: Claude Sonnet 5.5 played four, and Claude Opus 5.5 and Claude Fable 5.1 three each.
On the last card, UFC Fight Night: Moicano vs. Nolan, the crowd got 9 of 12 winners right. The betting favourites also went 9 of 12, and our model went 9 of 11. The crowd's two biggest misses were both close calls. Renato Moicano, at 54%, was stopped by Tom Nolan in round 3, and Farman Hasanov, at 52%, lost to Jean-Paul Lebosnoyani the same way.
Where the crowd went against the odds
Gabriel Lorenco vs Thomas Petersen is the one fight where the crowd's pick is the betting market's underdog. All 10 forecasters took Lorenco, at 57%. The betting market has Petersen at 60%, and our model has him at 63%.
The forecasters kept coming back to how Petersen got here. A Claude Fable 5.1 forecaster wrote: "Petersen has been knocked down three times recently and steps in on 13 days' notice against a power puncher with eight straight wins."
The featherweight bout between Jose Delano and Murtazali Magomedov is even in the betting market, at 50%. The crowd leaned Magomedov at 57%, with 8 of 10 forecasters on him. Our model also has Magomedov, at 54%.
Claude Sonnet 5.5 stood apart twice
Opus and Fable forecasters were unanimous in every fight. Claude Sonnet 5.5's four forecasters split 2-2 twice.
In the bantamweight fight, all six Opus and Fable forecasters took Elijah Smith, at 66% and 64%. Sonnet averaged 51% for Smith, and one of its forecasters went the other way: "Grant's experience and high volume should beat a 24-year-old coming off hand surgery and a long layoff, though Smith's grappling is a threat to a 40-year-old." The betting market has Smith at 72%.
Sonnet did the same in Delano vs Magomedov, averaging 52% for Magomedov while Opus sat at 62% and Fable at 61%. A Claude Sonnet 5.5 forecaster who picked Delano gave the fight almost no edge either way: "Delano has the higher volume and the Brazilian Top Team background. Magomedov is unbeaten and a finisher, so this is close."
Confident, but less than the odds
The crowd's most confident picks were Seokhyeon Ko at 73%, Nina Milosevic at 72% and Mantas Kondratavicius at 69%. All three are clear favourites elsewhere too: Ko is at 85% in the betting market and 89% with our model.
Whenever the crowd and the betting market picked the same fighter, the crowd's number was lower. The widest gap is Keiichiro Nakamura vs Ollie Schmid. The crowd has Nakamura at 62%, against 74% in the betting market and 81% from our model.
His hand kept coming up in the forecasters' reasons. A Claude Opus 5.5 forecaster wrote: "Nakamura is the Road to UFC champion against a 4-3 opponent, but his twice-broken hand limited his camp."
| Fight | AI crowd | Our model | Betting market |
|---|---|---|---|
| Acosta vs Walker | Acosta 63% | Acosta 65% | Acosta 66% |
| Suarez vs Jandiroba | Suarez 63% | Suarez 68% | Suarez 64% |
| Grant vs Smith | Smith 60% | Smith 78% | Smith 72% |
| Elekana vs Fernando | Fernando 61% | Fernando 76% | Fernando 72% |
| Cachoeira vs Milosevic | Milosevic 72% | Milosevic 86% | Milosevic 80% |
| Raposo vs Degli | Degli 63% | Degli 68% | Degli 67% |
| Nakamura vs Schmid | Nakamura 62% | Nakamura 81% | Nakamura 74% |
| Lorenco vs Petersen | Lorenco 57% | Petersen 63% | Petersen 60% |
| Silva vs Fernandes | Silva 61% | Silva 73% | Silva 65% |
| Delano vs Magomedov | Magomedov 57% | Magomedov 54% | Even |
| Kondratavicius vs Petroski | Kondratavicius 69% | Kondratavicius 76% | Kondratavicius 72% |
| Ko vs Turman | Ko 73% | Ko 89% | Ko 85% |
| Parsons vs Souza | Souza 64% | Souza 71% | Souza 67% |
Our model and the betting market as of Nov 4, 4:18 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 5.1 |
|---|---|---|---|---|
| Acosta v Walker | 63 for Acosta | 62 for Acosta | 64 for Acosta | 62 for Acosta |
| Suarez v Jandiroba | 63 for Suarez | 63 for Suarez | 62 for Suarez | 63 for Suarez |
| Grant v Smith | 60 for Smith | 66 for Smith | 51 for Smith | 64 for Smith |
| Elekana v Fernando | 61 for Fernando | 60 for Fernando | 63 for Fernando | 58 for Fernando |
| Cachoeira v Milosevic | 72 for Milosevic | 71 for Milosevic | 73 for Milosevic | 72 for Milosevic |
| Raposo v Degli | 63 for Degli | 62 for Degli | 65 for Degli | 62 for Degli |
| Nakamura v Schmid | 62 for Nakamura | 57 for Nakamura | 65 for Nakamura | 61 for Nakamura |
| Lorenco v Petersen | 57 for Lorenco | 55 for Lorenco | 60 for Lorenco | 55 for Lorenco |
| Silva v Fernandes | 61 for Silva | 61 for Silva | 62 for Silva | 59 for Silva |
| Delano v Magomedov | 57 for Magomedov | 62 for Magomedov | 52 for Magomedov | 61 for Magomedov |
| Kondratavicius v Petroski | 69 for Kondratavicius | 69 for Kondratavicius | 70 for Kondratavicius | 67 for Kondratavicius |
| Ko v Turman | 73 for Ko | 73 for Ko | 74 for Ko | 72 for Ko |
| Parsons v Souza | 64 for Souza | 65 for Souza | 63 for Souza | 63 for Souza |
How the AI crowd works
The crowd is 10 AI forecasters, played by Claude Opus 5.5, Claude Sonnet 5.5 and Claude Fable 5.1 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: Cortes Acosta vs. Walker: the full card, odds and results →