Mixed martial arts · 2024 season
Bellator forecasts
Win probabilities for every fight from a rating model fit only on past results.
Recent results
| Date | Fight | Score | Pre-game chance of the winner |
|---|---|---|---|
| 14 Sep 2024 | Marc Diakiese vs Tim Wilde | 1–0 | 31% |
| 14 Sep 2024 | Sara Collins vs Leah McCourt | 1–0 | 43% |
| 14 Sep 2024 | Luke Trainer vs Laurynas Urbonavicius | 1–0 | 82% |
| 14 Sep 2024 | Dmytrii Hrytsenko vs Darragh Kelly | 0–1 | 61% |
| 14 Sep 2024 | Joseph Luciano vs Steven Hill | 1–0 | 42% |
| 14 Sep 2024 | Manoel Sousa vs Archie Colgan | 0–1 | 76% |
| 14 Sep 2024 | Simeon Powell vs Rafael Xavier | 1–0 | 66% |
| 14 Sep 2024 | Eslam Abdul Baset vs Mike Shipman | 0–1 | 77% |
| 14 Sep 2024 | Tuomas Grönvall vs Ciaran Clarke | 0–1 | 86% |
| 8 Sep 2024 | Usman Nurmagomedov vs Alexander Shabliy | 1–0 | 55% |
| 8 Sep 2024 | Levan Chokheli vs Lorenz Larkin | 0–1 | 68% |
| 8 Sep 2024 | Aaron Jeffery vs Douglas Lima | 1–0 | 21% |
Questions and answers
How accurate are PodiumOracle's Bellator predictions?
On 433 games from held-back seasons (2022 to 2024), which the model never saw while it was tuned, the side it favoured won 66% of the time. Its log loss was 0.649, against 0.693 for the base rate and 0.653 for recent form.
How are the Bellator predictions made?
Every fighter has a rating that moves after each result, by more when the result was a surprise. The gap between two ratings, plus home advantage, gives each game's chance. The settings are chosen on older seasons and tested on the most recent ones.