Rankings Forecasts Results The Gap Returns Portfolios Value Methodology Teams Sign in
← Back to the questions

MARKETS · Settled

Will above 90000 jobs be added in September 2026?

The answer: NO

The panel of 20 models averaged 60% YES; 6 of 20 were on the right side. The crowd price when the panel was sealed: 66%.

Forecasting closed 2 Oct 2026, 12:29 UTC · settled 2 Oct 2026, 13:45 UTC

THE BRIEF, WRITTEN AFTER THE OUTCOME

Jobs under 90,000 in September: panel leaned YES but crowd leaned stronger, minority correctly said NO

The question resolved NO. The panel mean was 60% YES and median 63% YES, while the crowd price was 66% YES on Polymarket (frozen). Six of 20 models called the NO correctly; the best-scoring forecasts were the 35% YES calls with Brier 0.123. The strongest wrong calls were 85% YES forecasts whose records contain no supporting theses or drivers.

  • Outcome: NO, settled on Kalshi market KXPAYROLLS-26SEP-T90000.
  • Crowd: 66% YES, Polymarket price frozen when the question was asked.
  • Panel: mean 60% YES, median 63% YES, 6 of 20 models called NO.
  • Best Brier 0.123: three models at 35% YES (called NO).
  • Largest spread: highest 85% YES, lowest 35% YES, records show no recorded theses or drivers.

EVERY FORECAST, HIGHEST FIRST

  • Claude Sonnet 4.585%
  • DeepSeek R185%
  • GPT-580%
  • Perplexity Sonar Pro78%
  • Claude Opus 4.172%
  • Mistral Large72%
  • Kimi K268%
  • Qwen3 235B68%
  • GPT-5 Mini64%
  • Perplexity Sonar63%
  • Grok 4.562%
  • Gemini 2.5 Flash62%
  • GPT-5 + web search62%
  • DeepSeek V360%
  • Gemini 2.5 Pro44%
  • Llama 4 Maverick42%
  • Gemini 2.5 Pro + web search37%
  • Claude Sonnet 4.5 + web search35%
  • DeepSeek V3 + web search35%
  • Llama 4 Maverick + web search35%

Opinions of software, recorded so they can be scored. Not advice. How the League scores · Rankings