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

MARKETS · Settled

Will above 70000 jobs be added in September 2026?

The answer: NO

The panel of 20 models averaged 67% YES; 3 of 20 were on the right side. The crowd price when the panel was sealed: 59%.

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

THE BRIEF, WRITTEN AFTER THE OUTCOME

Panel leaned YES but outcome NO; a few low forecasts picked the result

The question resolved NO. The panel mean was 67% and median 69%, both above the crowd Polymarket price of 59%. Three models correctly called NO; Llama 4 Maverick made the best call at 25%. High forecasts missed broadly, with DeepSeek R1 the strongest YES at 88%.

  • Outcome: NO, resolved 2026-10-02.
  • Panel mean 67% and median 69%, versus the crowd Polymarket price 59%.
  • Three models called the NO outcome, 17 did not.
  • Best Brier was Llama 4 Maverick at 0.063 with a 25% YES probability.
  • Worst Brier was DeepSeek R1 at 0.774 with an 88% YES probability.

EVERY FORECAST, HIGHEST FIRST

  • DeepSeek R188%
  • GPT-5 + web search86%
  • Claude Sonnet 4.585%
  • Kimi K282%
  • GPT-582%
  • GPT-5 Mini75%
  • Claude Opus 4.175%
  • Gemini 2.5 Flash72%
  • Grok 4.571%
  • DeepSeek V370%
  • Mistral Large68%
  • Gemini 2.5 Pro68%
  • Perplexity Sonar67%
  • Perplexity Sonar Pro66%
  • Qwen3 235B62%
  • DeepSeek V3 + web search55%
  • Llama 4 Maverick + web search52%
  • Claude Sonnet 4.5 + web search45%
  • Gemini 2.5 Pro + web search44%
  • Llama 4 Maverick25%

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