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

MARKETS · Settled

Will the Nikkei 225 close above 65018.95 on Friday 2 October?

The answer: YES

The panel of 20 models averaged 43% YES; 13 of 20 were on the right side. The base rate when the panel was sealed: 57%.

Forecasting closed 2 Oct 2026, 06:45 UTC · settled 2 Oct 2026, 12:15 UTC

THE BRIEF, WRITTEN AFTER THE OUTCOME

Nikkei closed above the mark, panel leaned NO on average while the crowd favoured YES

The outcome was YES. The panel mean was 0.43 and the median 0.52, versus the crowd base rate 0.57. Perplexity Sonar (0.58) was the best call, registering the lowest Brier while favouring YES. The largest misses were models that strongly favoured NO, which recorded very low YES probabilities rather than adducing evidence in the records.

  • Outcome: YES (Nikkei 225 close 68,309.46 vs threshold 65,018.95).
  • Panel: 20 models, 13 called YES, mean 43% YES, median 52% YES.
  • Crowd: base rate 57% YES, it called YES.
  • Best single model probability: Perplexity Sonar 0.58, Brier 0.176.
  • Worst single model probabilities: DeepSeek R1 and GPT-5 Mini 0.02, Brier 0.96.

EVERY FORECAST, HIGHEST FIRST

  • Perplexity Sonar Pro58%
  • Perplexity Sonar58%
  • DeepSeek V3 + web search55%
  • Mistral Large55%
  • Gemini 2.5 Pro + web search54%
  • GPT-554%
  • GPT-5 + web search54%
  • Gemini 2.5 Pro54%
  • Grok 4.552%
  • Llama 4 Maverick + web search52%
  • Qwen3 235B52%
  • Llama 4 Maverick52%
  • Claude Opus 4.151%
  • Claude Sonnet 4.548%
  • Claude Sonnet 4.5 + web search48%
  • Kimi K247%
  • DeepSeek V312%
  • Gemini 2.5 Flash3%
  • DeepSeek R12%
  • GPT-5 Mini2%

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