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 | 275 | 0.679 | 58% |
| Base rate | 275 | 0.693 | 53% |
| Recent form | 275 | 0.694 | 55% |
Box lacrosse is played indoors, in hockey arenas, so there is no weather.
The playoffs are simulated in the current format: the top eight in one table, single-game quarter-finals, then best-of-three semi-finals and finals. Earlier seasons used division brackets; their games all count for the ratings.
Live scores come from ESPN's scoreboard, matched to the league's games by date and teams.
Goalies and injuries are not in the model.
Questions and answers
How accurate are PodiumOracle's National Lacrosse League predictions?
On 275 games from held-back seasons (2024–25 to 2025–26), which the model never saw while it was tuned, the side it favoured won 58% of the time. Its log loss was 0.679, against 0.693 for the base rate and 0.694 for recent form.