How accurate are UFCalendar's MMA fight predictions?
How the model's pre-fight probabilities actually fared, graded against every decided fight in the database. Nothing is cherry-picked: these are the same numbers shown on bout pages before each fight, recomputed live.
UFCalendar's Power Index has correctly picked the winner in 63% of the 26259 MMA fights graded so far, with a Brier score of 0.232. It is well calibrated: when the model gives a fighter a 50%-55% chance, that favorite wins 54.8% of the time.
- Favorite accuracy
- 63%
- Fights graded
- 26259
- Brier score
- 0.232
When the model says 70%, does the favorite actually win about 70% of the time? Predicted vs. observed, by confidence bucket.
| Model says | Avg predicted | Favorite won | Fights |
|---|---|---|---|
| 50%–55% | 52.3% | 54.8% | 8732 |
| 55%–60% | 57.4% | 60.2% | 7137 |
| 60%–65% | 62.3% | 65.8% | 5100 |
| 65%–70% | 67.2% | 72.9% | 3037 |
| 70%–75% | 72.2% | 80.8% | 1798 |
| 75%–80% | 77.1% | 82% | 762 |
| 80%–100% | 83.1% | 83.7% | 300 |
| Year | Fights | Accuracy |
|---|---|---|
| 2026 | 1824 | 60.6% |
| 2025 | 2138 | 63.8% |
| 2024 | 2082 | 64.1% |
| 2023 | 2232 | 63.4% |
| 2022 | 1868 | 64.5% |
| 2021 | 1736 | 63.5% |
| 2020 | 1303 | 63.9% |
| 2019 | 1843 | 62.3% |
| 2018 | 1562 | 61.1% |
| 2017 | 1319 | 63% |
| 2016 | 1188 | 61.8% |
| 2015 | 964 | 61.5% |
| 2014 | 995 | 63.6% |
| 2013 | 742 | 65.4% |
| 2012 | 650 | 60.5% |
| 2011 | 670 | 61.6% |
| 2010 | 629 | 61.8% |
| 2009 | 470 | 58.9% |
| 2008 | 372 | 60.2% |
| 2007 | 320 | 63.4% |
| 2006 | 357 | 67.8% |
| 2005 | 211 | 71.1% |
| 2004 | 161 | 60.9% |
| 2003 | 100 | 64% |
| 2002 | 121 | 64.5% |
| 2001 | 90 | 65.6% |
| 2000 | 93 | 74.2% |
| 1999 | 65 | 66.2% |
| 1998 | 40 | 60% |
| 1997 | 35 | 60% |
| 1996 | 31 | 67.7% |
| 1995 | 27 | 59.3% |
| 1994 | 18 | 77.8% |
| 1993 | 3 | 66.7% |
Every number on this page is recomputed live from the stored pre-fight probabilities, the same ones printed on bout pages before each fight happens. Draws are excluded (no verdict to grade) and dead-even 50/50 prices grade no pick. Ratings and probabilities span every organization we track, back to 1993.