How often the favourite won, how often the home side won, and how often the total went over — across 20,339 played NBA games in AccuScore's archive. None of this is our model. The favourite is whichever side the bookmakers priced shorter, and the result is the result.
Seasons run October to June.
| Season | Games | Favourite won | Home won | Over |
|---|---|---|---|---|
| 2025-26 | 1,317 | 68.3% | 55.7% | 50.2% |
| 2024-25 | 1,321 | 67.7% | 54.5% | 52.3% |
| 2023-24 | 1,311 | 67.9% | 54.9% | 49.9% |
| 2022-23 | 1,307 | 69.8% | 58.2% | 47.2% |
| 2021-22 | 1,290 | 67.5% | 54.6% | 50.1% |
| 2020-21 | 1,205 | 64.5% | 54.4% | 49.1% |
| 2019-20 | 1,102 | 65.1% | 55.0% | 51.3% |
| 2018-19 | 1,311 | 66.6% | 59.1% | 48.6% |
| 2017-18 | 1,311 | 67.0% | 58.4% | 47.2% |
| 2016-17 | 1,308 | 66.8% | 58.3% | 49.7% |
| 2015-16 | 1,316 | 69.6% | 59.3% | 49.2% |
| 2014-15 | 1,311 | 69.8% | 57.6% | 47.5% |
| 2013-14 | 1,315 | 68.0% | 57.9% | 51.3% |
| 2012-13 | 1,229 | 69.0% | 61.2% | 49.6% |
| 2011-12 | 1,074 | 69.7% | 59.3% | 47.8% |
| 2010-11 | 1,311 | 68.6% | 60.8% | 48.1% |
Across 19,233 priced NBA games in AccuScore's archive, the favourite won 67.8% of the time. The favourite is whichever side the bookmakers made shorter, not whichever side we projected.
Home teams won 57.5% of 20,339 games in the archive.
Of 19,340 games with a posted total, 49.4% went over. A rate near half is what an efficient market looks like — the number is interesting when a season moves away from it.