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 | 1294 | 1.048 | 45% |
| Base rate | 1294 | 1.075 | 44% |
| Recent form | 1294 | 1.062 | 44% |
| Betting market | 1284 | 1.034 | 48% |
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 |
|---|---|---|---|---|---|
| Bookmakers (Pinnacle, DraftKings, bet365) | 20 Jan 2026 | 417 | 1.054 | 1.031 | −2.2% |
| Polymarket | 17 Oct 2025 | 617 | 1.052 | 1.034 | −1.7% |
| Kalshi | 31 Oct 2025 | 594 | 1.052 | 1.033 | −1.9% |
A draw is its own outcome: the model has a band around an even contest in which a draw is the likeliest single result, fit on the tuning seasons.
The market comparison uses the bookmakers' closing odds with their margin taken out.
Parachute payments give relegated Premier League clubs a head start the ratings only learn from results.
Team news and injuries are not in the model.
Expected goals are not in the rating yet; it learns only from results.
Questions and answers
How accurate are PodiumOracle's EFL Championship predictions?
On 1,294 games from held-back seasons (2024–25 to 2026–27), which the model never saw while it was tuned, the side it favoured won 45% of the time. Its log loss was 1.048, against 1.075 for the base rate and 1.062 for recent form, and 1.034 for the closing betting odds. Against the betting markets on the same finished games, its log loss was 1.054 against the bookmakers' 1.031 over 417 games, 1.052 against Polymarket's 1.034 over 617 games and 1.052 against Kalshi's 1.033 over 594 games (lower is better).