How our predictions work
Every basketball prediction on Basketora comes from a rating model, checked against the betting market and tested on past seasons before it goes live.
The model
Every team gets two ratings per competition: how many points it scores and how many it concedes compared with an average side, adjusted for the opponents it has faced. The ratings come from a weighted regression over this season’s finished games, plus a home-court edge measured for each competition.
Recent games count most. The weighting is sized in games rather than days, so a team’s rating reflects roughly its last twenty games whether it plays every other night (NBA) or once a week (EuroLeague). Early in a season last season’s ratings act as a starting point and fade as new games come in.
The ratings give an expected margin and total for each game. Treating both as normally distributed turns them into a win probability, a cover probability for the spread and an over/under probability for the total. The projected score is the one that fits those probabilities.
When bookmakers price a game, the model is blended with their de-vigged consensus (70% market for the winner, 60% for spread and total). A pick against the market favourite has to clear a large edge the model itself backs; otherwise it is pulled back towards the market.
What the labels mean
Win probability of 72% or more. Even a 75% pick misses one time in four.
60% to 71%: a real but modest edge.
Below 60%. The model has a side, but the game is close to a coin flip.
Tested on a full season
| Competition · 2025–26 | Games | Winners called | Always home team | Better record |
|---|---|---|---|---|
| NBA | 1,315 | 68.8% | 55.6% | 66.8% |
| EuroLeague | 402 | 62.2% | 63.7% | 56.1% |
| Liga ACB | 327 | 67.3% | 57.5% | 65.4% |
| Basketbol Süper Ligi | 261 | 73.6% | 59.4% | 71.1% |
| Greek Basket League | 173 | 75.1% | 60.7% | 71.0% |
| Lega Basket Serie A | 255 | 68.6% | 61.2% | 62.1% |
| NBL | 179 | 63.7% | 53.1% | 65.5% |
Walk-forward: every game predicted using only the games played before it, the model alone (no bookmaker prices). “Always home team” and “better record” are the simple rules the model has to beat.
Two weak spots in 2025–26. In the EuroLeague, always picking the home team called slightly more winners (63.7%) than the model (62.2%), although the model’s probabilities were closer to what happened. In the NBL, picking the team with the better record (65.5%) beat the model (63.7%). Both competitions have bookmaker prices on live games, so the market blend carries more weight there.
What it cannot see
- The model does not see injuries, rest days or line-ups; the market blend is how that news reaches the probabilities.
- Early in a season, and for teams new to a competition, ratings lean on last season and are less certain.
- A 70% pick still loses three times in ten. Probabilities describe many games, not a single outcome.
Every pick in public
Every prediction is published before tip-off and settled after the final whistle. The accuracy page keeps the running record by competition and by confidence, misses included.
Predictions are statistical estimates, not guarantees. If you bet, set a budget first; our responsible gambling page lists free, confidential help.