Today's NBA predictions
This table updates by itself, showing Sage's win chance and projected score for the free half of upcoming NBA games. The full list is on the NBA predictions page.
Updated Friday 30 October. Times are UK time. 6 of the 13 games in the next 36 hours are free. Pro opens the rest, plus over/under and handicap reads.
| Tip-off | Game | Sage favours | Win chance | Projected score |
|---|---|---|---|---|
| Fri 23:30 | Boston Celtics v Chicago Bulls | Boston Celtics | 84% | 125 – 109 |
| Fri 23:30 | Toronto Raptors v Orlando Magic | Toronto Raptors | 68% | 116 – 108 |
| Fri 23:30 | Brooklyn Nets v Detroit Pistons | Detroit Pistons | 76% | 105 – 116 |
| Sat 00:00 | Minnesota Timberwolves v Los Angeles Clippers | Minnesota Timberwolves | 64% | 115 – 109 |
| Sat 00:00 | Dallas Mavericks v Houston Rockets | Houston Rockets | 69% | 110 – 118 |
| Sat 19:00 | Utah Jazz v Portland Trail Blazers | Portland Trail Blazers | 64% | 113 – 119 |
Most NBA games tip off after midnight UK time, so a game on tonight's card can sit on tomorrow's date.
Below is what five seasons of results say about how to read those numbers. All figures come from 5,670 finished NBA games in our data since 2022-23, read on 30 October 2026, and include pre-season and play-off games where the feed lists them.
Five seasons of the NBA
| Season | Games | Points a game | Home wins | Home margin |
|---|---|---|---|---|
| 2022-23 | 1,386 | 228.3 | 57.2% | +2.5 |
| 2023-24 | 1,377 | 227.2 | 54.9% | +2.2 |
| 2024-25 | 1,386 | 226.5 | 54.8% | +2.0 |
| 2025-26 | 1,384 | 229.8 | 55.4% | +1.7 |
| 2026-27 so far | 137 | 229.4 | 51.8% | +1.2 |
The home lean keeps shrinking
The home margin has dropped in every season in the table: 2.5, 2.2, 2.0, 1.7. This season's +1.2 comes from just 137 games and includes pre-season, so ignore that figure for now. The four full seasons are the real signal.
At 55.5% over the whole sample, the NBA has one of the weakest home leans we track. Hosts in the EuroLeague win 62.6%. Only Japan's B.League, Korea's KBL and Poland's top league are lower among our 24.
What it means in practice: in the NBA, team strength decides who the favourite is, and venue is a small nudge on top. A good side on the road will often be the likelier winner, which is much rarer in Europe. We look at this in depth in home advantage: NBA vs Europe.
Scoring has barely moved
For all the talk about offence, the NBA has sat between 226.5 and 229.8 points for four straight seasons. Compare that with the EuroLeague, up 12 points in the same time.
Eight in ten NBA games finish with a combined score between 203 and 254. That is a 51-point band, which is why a projected total is only ever a central estimate.
The quarters show a clear pattern. Across 5,374 games they average 57.4, 57.2, 57.6 and then 54.9. The fourth is the lowest-scoring quarter by about 2.5 points, so simply doubling the half-time score tends to overshoot a little.
Close games and blowouts, at the same time
The NBA is two leagues in one:
| Result | NBA | EuroLeague |
|---|---|---|
| Within 3 points | 14.9% | 17.7% |
| Within 5 points | 25.6% | 32.0% |
| Decided by 20+ | 19.7% | 11.8% |
| Overtime | 5.2% | 4.5% |
A quarter of games go down to the wire, yet almost one in five is a 20-point blowout. Forty-eight minutes gives a stronger team more time to pull clear, and a side resting players can collapse quickly.
This matters for reading a win chance. It tells you who is likelier to win. It says little about how close it will be. For that, look at the projected score, then remember how wide real results run.
The teams that break the pattern
Over the last three full seasons, a few teams have made nonsense of the 55% average at home:
| Team | Home games | Home wins | Home margin |
|---|---|---|---|
| Oklahoma City Thunder | 155 | 81% | +12.8 |
| Boston Celtics | 151 | 77% | +10.8 |
| Houston Rockets | 135 | 70% | +6.6 |
| New York Knicks | 156 | 69% | +7.1 |
| Cleveland Cavaliers | 148 | 68% | +5.6 |
| Denver Nuggets | 146 | 68% | +6.1 |
These are strong teams who are also at home, so don't read them as pure home-court effects. Sage picks this up through team ratings, game by game, and fades it when form turns. See the NBA table for where each team stands today.
What the model cannot see
In our test on games the model had never seen, Sage's favourite won 66.0% of 1,518 NBA games, and its typical miss on the margin was 11.6 points. That is a good guide to what to expect from any single night.
The NBA is also a hard league for a results-only model, because team news can move a game a long way. A superstar sitting out on the second night of a back-to-back can change a game by several points, and Sage does not know.
We do not have injury, rest or line-up data, so check them before you rely on any chance here. The method post has the full list of limits, and how the model did in testing on 9,440 games.
Early-season chances also lean on last season's ratings. Expect them to sharpen by December.
Frequently asked questions
How often does the home team win in the NBA? 55.5% of the time across 5,670 games, by 2.1 points on average. The margin has fallen each season, from 2.5 to 1.7.
How many points are scored in an NBA game? 228.0 on average, both teams combined. Eight in ten games land between 203 and 254.
How often does an NBA game go to overtime? 5.2% of the time, or about one game in 19.
Are these NBA predictions free? The strongest half of each day's games shows win chance and projected score free. Pro opens the rest, plus over/under and handicap reads. Settled calls are free on the record page.
Do you cover other leagues? Twenty-four in all. See the basketball predictions page, or how they differ in points per game by league.
Predictions are chances, not promises. Only bet what you can afford to lose. 18+. BeGambleAware.org.