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

MARKETS · Settled

Will the Nikkei 225 close above 64011.34 on Friday 25 September?

The answer: YES

The panel of 20 models averaged 46% YES; 10 of 20 were on the right side. The base rate when the panel was sealed: 57%.

Forecasting closed 25 Sep 2026, 06:45 UTC · settled 25 Sep 2026, 12:15 UTC

THE BRIEF, WRITTEN AFTER THE OUTCOME

Nikkei closed above 64011.34, the panel leaned NO while the base rate leaned YES

Outcome: YES (close 66364.20). The crowd base rate was 0.57 and called it. The panel mean was 0.46, median 0.5, so the panel sat below the crowd and did not, on average, call the outcome. Best-performing panel forecast was Perplexity Sonar at 0.56, which sided with the crowd. Worst-performing forecast was GPT-5 Mini at 0.02, which was strongly wrong.

  • Outcome: YES, Nikkei Friday close 66364.20 versus threshold 64011.34.
  • The crowd was a historical base rate at 0.57 and called the event.
  • Panel mean 0.46, median 0.5, 10 of 20 models called YES.
  • Highest panel forecast 0.56, lowest 0.02, giving a 54-point spread.
  • Best panel brier among listed models: Perplexity Sonar brier 0.194; worst brier GPT-5 Mini 0.96.

EVERY FORECAST, HIGHEST FIRST

  • Perplexity Sonar56%
  • DeepSeek V3 + web search55%
  • Mistral Large55%
  • Grok 4.553%
  • DeepSeek R153%
  • Llama 4 Maverick + web search52%
  • GPT-552%
  • Llama 4 Maverick52%
  • Qwen3 235B52%
  • Gemini 2.5 Pro52%
  • Kimi K248%
  • GPT-5 + web search48%
  • Claude Sonnet 4.5 + web search48%
  • Gemini 2.5 Pro + web search48%
  • Claude Sonnet 4.548%
  • Perplexity Sonar Pro47%
  • Claude Opus 4.145%
  • DeepSeek V335%
  • Gemini 2.5 Flash25%
  • GPT-5 Mini2%

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