How good are these forecasts?
Every forecast scored on seasons the model never saw while its settings were chosen.
Held-back seasons
| Forecast | Forecasts | Log loss | Favourite won |
|---|---|---|---|
| The model | 11010 | 0.501 | 75% |
| Base rate | 11010 | 0.684 | 57% |
| Recent form | 11010 | 0.582 | 69% |
Against the betting markets
Each market's price before the start, scored on the same finished games as the model. A negative figure means the market was more accurate.
| Market | Since | Priced | Model log loss | Market log loss | Model vs market |
|---|---|---|---|---|---|
| Kalshi | 10 Sep 2026 | 697 | 0.468 | 0.463 | −1.1% |
The margin of victory is sets (3–0 counts for more than 3–2); points within sets are not in the file.
The title odds use an approximate 64-team tournament of conference leaders and the best of the rest, seeded by simulated wins and rating, on neutral courts. The real bracket gives the top 16 seeds home matches in the first two rounds.
Volleyball is indoors, so there is no weather.
The season runs from late August to mid-December; outside it the pages show how the last one finished.
Questions and answers
How accurate are PodiumOracle's NCAA women's volleyball predictions?
On 11,010 games from held-back seasons (2024 to 2026), which the model never saw while it was tuned, the side it favoured won 75% of the time. Its log loss was 0.501, against 0.684 for the base rate and 0.582 for recent form. Against the betting markets on the same finished games, its log loss was 0.468 against Kalshi's 0.463 over 697 games (lower is better).