When this trader buys at $0.70, they imply 70% probability. Perfect calibration = the event happens 70% of the time.
| Bucket | Bets | Expected | Actual | Error |
|---|---|---|---|---|
| 0.00-0.10 | 1 | 6% | 0% | 6.0% |
| 0.20-0.30 | 3 | 26% | 0% | 25.6% |
| 0.30-0.40 | 3 | 37% | 0% | 36.7% |
| 0.40-0.50 | 31 | 47% | 0% | 46.8% |
| 0.50-0.60 | 56 | 54% | 0% | 54.2% |
| 0.60-0.70 | 10 | 64% | 0% | 63.7% |
| 0.70-0.80 | 7 | 76% | 0% | 75.7% |
| 0.80-0.90 | 2 | 81% | 0% | 80.6% |
| 0.90-1.00 | 2 | 98% | 0% | 97.5% |
On-chain verification: wallet age 8 days, 2000 txs, provenance grade D. Bot score: 0/100, wash trading score: 0/100.
Polymarket on-chain coverage: $0 in / $0 out across 0 withdrawal tx since never.
3100 total trades across 110 markets.
115 bets on resolved markets available for calibration scoring.
Calibration error: 53.9% — needs improvement.
Skill: 0/100 (calibration quality). Variance: 26/100 (higher = more volatile returns).
Brier Skill Score: -22.4% vs naive baseline (>0% = better than always predicting base rate).
Brier decomposition: REL=0.3053 RES=0.0000 UNC=0.0000.
Log loss: 0.8432 (skill: -10564.0% vs naive). Lower log loss = better calibration on rare events.
Timeliness: avg entry 1.2 days before resolution, 0% early mover.
Below average. The data shows poor calibration, thin evidence, or both. When this trader expresses high confidence, events don't happen at the rate they imply.
Confidence: D/48 [CI95: D→D, 47-49] (115 resolved bets). Moderate confidence — score may shift as more markets resolve.
Methodology: Brier Score Decomposition (Murphy 1973), Log Loss, On-Chain USDC Verification. Same approach used by IARPA to identify superforecasters.