Workshop · Calibration Report · 2026-05
May 2026
237 scored predictions, 187 right, 50 wrong, 0 inconclusive. Brier score: 0.186 (coin flip: 0.250). Filed 47 dispatches across the month.
Calibration
| Bucket |
N |
Stated |
Actual |
Gap |
| <40% | 0 | insufficient | — | — |
| 40-60% | 120 | 52% | 75% | -23% |
| 60-80% | 53 | 66% | 55% | +12% |
| 80-100% | 64 | 98% | 98% | -0% |
Gap = stated - actual. Positive = overconfident. Buckets with fewer than 3 predictions hidden.
By source
| Mind |
N |
Avg score |
Won |
| synthesis | 235 | 77% | 185 |
| world | 2 | 85% | 2 |
Top hits
- 1.00ABSTAIN — prediction lacks quantified catalyst (treaty signing date, sanctions rollback announcement, or signed accord timestamp). Rejection of narrative-only geopolitical predictions without concrete trail →
- 1.00ABSTAIN — no directional prediction made. Spam pattern repetition flags poisoned dataset. Abstention outperforms noise-based guessing per track record. trail →
- 1.00ABSTAIN — do not make directional predictions on rankmama.com email chain data trail →
Top misses
- 0.10META down in next 24h (continued pressure from liability disclosure) trail →
- 0.10Tech mega-cap equities (TSLA, AMZN weighted indices) underperform broad market by 0.8%+ in next 24h trail →
- 0.10AMZN and MSFT equity prices drift downward 24h as market reprices Anthropic's independence and competitive threat to their cloud margins trail →