Kazakhstan Kurultai Election Forecast: Adilet Majority Estimate Rises
The Adilet majority forecast rose from 62.6% to 71.2% over 83 minutes, while supplied market signals and model estimates remain divided.
The live forecast for whether Adilet will win a majority in the Kazakhstan Kurultai election increased from 62.6% to 71.2% over 83 minutes. The recorded change is 8.57 percentage points, with a velocity of 6.177 percentage points per hour.
The question has a resolution date of 2026-08-23. The increase favors the Adilet-majority outcome, but the supplied forecasts are live estimates rather than a settled result.
polymarket signals sit above the forecast movement
Four supplied polymarket signals list probabilities of 0.9995, 0.9985, 0.997 and 0.995. Their accompanying notes give market prices of 100.0%, 99.9%, 99.7% and 99.5%, respectively.
By comparison, the ai_consensus_run records a score of 0.3948, confidence of 0.4961 and dispersion of 0.265 across 2 judges. The difference between these records shows that the supplied estimates do not present a single common probability.
Model estimates diverge over the 73-seat threshold
The openai/gpt-4.1-nano estimate assigns a probability of 0.65 with confidence of 0.7. Its reasoning describes a moderate likelihood that Adilet will secure at least 73 seats, while identifying political dynamics and reporting delays as uncertainties.
The deepseek/deepseek-v4-flash estimate is lower at 0.12, with confidence of 0.65. Its supplied reasoning calls 73+ seats in a 98-seat Kurultai a high threshold and says recent polls show Adilet with around 60-65 seats.
ai_openai_nano: "Given the election date and typical reporting timelines, there's a moderate likelihood Adilet will secure at least 73 seats, but uncertainties remain due to political dynamics and reporting delays."
What the forecast does and does not show
The upward move indicates that the tracked forecast became more favorable to an Adilet majority during the recorded 83-minute period. It does not remove the disagreement between the polymarket signals, the ai_consensus_run and the two supplied model estimates.
The stored reasoning focuses on the election date, typical reporting timelines, political dynamics, reporting delays and the difficulty of reaching at least 73 seats. Until the question resolves, each probability should be read as a live estimate rather than confirmation of the election outcome.