Results
NBA results Monday 16 November 2026: AI predictions review
NBA results for Monday 16 November 2026: the AI models went 3 of 3, with Houston, Atlanta and Oklahoma City all delivered as called.
Three games, three winners, no arguments. Monday's NBA slate went exactly as the TuringStats panel of ten frontier models called it: Washington Wizards against Houston Rockets, Milwaukee Bucks against Atlanta Hawks and Sacramento Kings against Oklahoma City Thunder all landed on the correct side, giving the desk a clean 3 of 3.
The biggest surprise was not a wrong pick but a wrong margin. Houston won by 28 in a 147-point night in Washington, well beyond the 116-117 scoreline the models most commonly projected, and Atlanta's seven-point win in Milwaukee came against a consensus that had the game far tighter than it finished. Confidence was steady rather than spectacular — 68%, 66% and 72% average across the panel — but unanimity did the heavy lifting.

Full-time scenes from Monday's three-game slate, where the panel went 3 of 3.
Where the AI got it right
The Wizards were never expected to stay with Houston, and they did not. All ten models backed the Rockets, with an average confidence of 68% and a most common score of 116-117 — a projection the actual 119-147 result blew past in a high-tempo game. The direction was right, the margin wildly understated. Washington Wizards vs Houston Rockets prediction
Milwaukee against Atlanta was the least certain call of the night at 66% average confidence, and the models still got it. The Hawks won 113-106 on the road, a seven-point margin against a most common projected score of 116-117. A unanimous panel on a moderate-confidence game is the kind of result that flatters the consensus, and the desk will take it. Milwaukee Bucks vs Atlanta Hawks prediction
The firmest call of the slate was Oklahoma City, backed by all ten models at 72% average confidence with a most common score of 112-113. The Thunder delivered 125-103 in Sacramento, a 22-point win that made the projection look conservative. Of the three, this was the one where the panel's conviction matched the performance most closely. Sacramento Kings vs Oklahoma City Thunder prediction
Where it got it wrong
Nothing. For the first time on this slate there are no misses to report: all three pre-match consensus picks — Houston, Atlanta and Oklahoma City — were correct, and every one of the ten models listed on the day recorded a perfect card.
The honest caveat is calibration rather than accuracy. The panel called all three winners but understated the margins in two of them, most obviously in Washington, where a projected one-point game became a 28-point rout. A perfect record on winners should not be read as a perfect read on how these games would be played.
Models of the day
There is no separating the field. GPT-6.1 Sol, Claude Opus 5.5, DeepSeek V4 Pro, Qwen 3.8 Max, MiMo V2.6 Pro, Grok 4.7, Gemini 4 Argon, Kimi K3, GLM 5.3, Meta Muse Spark 1.3, NVIDIA Nemotron 3 Ultra and Tencent Hy4 each finished 3 of 3.
With twelve models on the list and no incorrect calls between them, there is no worst performer to name. The day belongs to the panel collectively, and the differentiation will have to wait for a slate with a genuine split.
Full results
- NBA: Washington Wizards 119–147 Houston Rockets — consensus Houston Rockets win (100%) RIGHT
- NBA: Milwaukee Bucks 106–113 Atlanta Hawks — consensus Atlanta Hawks win (100%) RIGHT
- NBA: Sacramento Kings 103–125 Oklahoma City Thunder — consensus Oklahoma City Thunder win (100%) RIGHT
A 3 of 3 card is a clean result for the models, but the next slate will test whether unanimity holds when the games are closer on paper. Today's predictions, with the full model-by-model breakdown, are live at TuringStats NBA predictions for today.
Next up: today's nba predictions from the same panel, and the results archive with every graded call. Forecasts are informational only.