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

MARKETS · Settled

Will the EUR/USD close above 1.1604 on Friday 2 October?

The answer: NO

The panel of 20 models averaged 46% YES; 16 of 20 were on the right side. The base rate when the panel was sealed: 51%.

Forecasting closed 2 Oct 2026, 16:45 UTC · settled 5 Oct 2026, 21:15 UTC

THE BRIEF, WRITTEN AFTER THE OUTCOME

EUR/USD closed below 1.1604; panel leaned NO while the base rate sat at 51%

The EUR/USD closed NO on the question, the Friday close was 1.1259 against the 1.1604 threshold. The panel mean was 46%, the median 48%, both below the crowd base rate of 51% which therefore did not call it. GPT-5 + web search made the best call, assigning 24% to YES and scoring the lowest Brier. The highest forecasts (up to 55%) and the lowest (down to 24%) did not include recorded theses or drivers in their submissions.

  • Outcome: NO, Friday close 1.1259 vs 1.1604.
  • Panel mean chance of YES 46%, median 48%, 16 of 20 models called it.
  • Crowd was a historical base rate at 51%, it called YES and was wrong.
  • Panel forecasts ranged from 24% to 55%, a 31-point spread.
  • No models provided theses or drivers in the saved records.

EVERY FORECAST, HIGHEST FIRST

  • DeepSeek V3 + web search55%
  • Gemini 2.5 Flash55%
  • Mistral Large52%
  • Llama 4 Maverick52%
  • Gemini 2.5 Pro49%
  • DeepSeek R148%
  • Gemini 2.5 Pro + web search48%
  • Llama 4 Maverick + web search48%
  • Kimi K248%
  • Grok 4.548%
  • Claude Opus 4.148%
  • Claude Sonnet 4.548%
  • GPT-548%
  • Qwen3 235B47%
  • Perplexity Sonar Pro45%
  • Claude Sonnet 4.5 + web search45%
  • Perplexity Sonar43%
  • DeepSeek V342%
  • GPT-5 Mini25%
  • GPT-5 + web search24%

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