Results
NBA results Tuesday 10 November 2026: AI predictions review
NBA results Tuesday 10 November 2026: the AI models went 4 of 8, with a 100% Cleveland consensus and a unanimous Toronto call both sunk by road upsets.
Eight games, ten frontier models, and a split verdict. Tuesday's NBA slate produced a 4 of 8 record for the TuringStats consensus, with the models calling exactly half the board correctly and missing on the other half. The averages were not shy, either: three of the eight matches drew a unanimous 100% pick from the ten-model panel, and two of those unanimous calls went down.
The biggest surprise came in Toronto, where all ten models backed the Raptors and the Grizzlies left with a 118-102 win. Charlotte's home defeat of Cleveland, a 100% consensus pick in the other direction, ran it close for the day's most costly miss.
Eight games, ten models, four correct calls on Tuesday night.
Where the AI got it right
The panel's strongest correct call of the night was in Houston, where all ten models picked the Rockets and the average confidence sat at 62%. The consensus scoreline was 118-109; the actual result was Houston Rockets vs Denver Nuggets prediction, 110-100. The margin landed a little wider than forecast, but the direction was never in doubt.
Golden State delivered the most emphatic return of the evening. A 100% consensus, an average confidence of 61% and a most common score of 118-109 gave way to a 138-87 demolition of New Orleans, the largest winning margin on the slate. See the Golden State Warriors vs New Orleans Pelicans prediction for the pre-match numbers.
Phoenix was a more measured hit. Nine of ten models backed San Antonio at 58% average confidence, with a most common score of 114-115 — essentially a coin-flip margin. The Spurs instead won 122-96, comfortably covering a call the panel had made only narrowly. Full detail sits on the Phoenix Suns vs San Antonio Spurs prediction page.
Detroit rounded out the correct calls. The Pistons drew 80% of the panel at 57% average confidence, with 116-112 the most common forecast score. They beat Washington 122-111, a result that matched the pick and roughly matched the projected margin. The Detroit Pistons vs Washington Wizards prediction records the consensus in full.
Where it got it wrong
The costliest miss, by consensus weight, came in Charlotte. All ten models picked Cleveland, at an average confidence of 64% — the highest on the day — with 112-113 the most common score. The Hornets won 124-112. A unanimous panel, and the day's most confident one, was beaten by twelve points. The Charlotte Hornets vs Cleveland Cavaliers prediction shows how one-sided the pre-match view was.
Toronto was the other unanimous failure. Ten of ten backed the Raptors at 58% average confidence, projecting 116-111. Memphis won 118-102, a sixteen-point swing against a call no model questioned. The Toronto Raptors vs Memphis Grizzlies prediction page carries the numbers.
Dallas was a near-miss in consensus terms and a clear miss in outcome terms. Nine of ten models favoured Indiana at 57% average confidence, with 116-117 the most common score — a one-point margin. The Mavericks won 125-122. The Dallas Mavericks vs Indiana Pacers prediction shows a panel that expected a tight game and got one, but with the wrong winner.
Atlanta completed the set of misses. Eight of ten models backed the Hawks at 57% average confidence, forecasting 121-117. The Lakers won 124-117. The projected margin was close; the projected victor was not. See the Atlanta Hawks vs Los Angeles Lakers prediction for the breakdown.
Models of the day
Three models shared top spot on 5 of 8: DeepSeek V4 Pro, MiMo V2.6 Pro and Meta Muse Spark 1.3. Each found one match the wider consensus missed, which on a day of two unanimous defeats was enough to separate them from the pack.
A cluster of five models finished on 4 of 8: Claude Sonnet 5.5, Mistral Medium 3.5, Qwen 3.8 Max, Grok 4.7 and GLM 5.3. All five landed on the consensus record exactly.
The bottom of the table was shared three ways. GPT-6 Luna, Gemini 3.8 Flash and Kimi K3 each finished on 3 of 8, one below the consensus. None of the eleven models on the day cleared six correct calls.
Full results
- NBA: Atlanta Hawks 114-124 Los Angeles Lakers — consensus Atlanta Hawks win (80%) WRONG
- NBA: Charlotte Hornets 124-112 Cleveland Cavaliers — consensus Cleveland Cavaliers win (100%) WRONG
- NBA: Detroit Pistons 122-111 Washington Wizards — consensus Detroit Pistons win (80%) RIGHT
- NBA: Toronto Raptors 102-118 Memphis Grizzlies — consensus Toronto Raptors win (100%) WRONG
- NBA: Dallas Mavericks 125-122 Indiana Pacers — consensus Indiana Pacers win (90%) WRONG
- NBA: Houston Rockets 110-100 Denver Nuggets — consensus Houston Rockets win (100%) RIGHT
- NBA: Phoenix Suns 96-122 San Antonio Spurs — consensus San Antonio Spurs win (90%) RIGHT
- NBA: Golden State Warriors 138-87 New Orleans Pelicans — consensus Golden State Warriors win (100%) RIGHT
The lesson from Tuesday is a familiar one for anyone tracking model consensus: unanimity is not certainty. Three 100% picks produced two defeats, and the highest average confidence on the slate, Cleveland at 64%, was the most expensive miss of the night. Today's slate is already loaded, with fresh ten-model consensus numbers and projected scorelines for every game on the today's NBA predictions page.
Next up: today's nba predictions from the same panel, and the results archive with every graded call. Forecasts are informational only.
