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 | 2 | 5% | 0% | 5.0% |
| 0.10-0.20 | 3 | 17% | 0% | 17.3% |
| 0.20-0.30 | 11 | 26% | 0% | 26.2% |
| 0.30-0.40 | 15 | 36% | 0% | 35.6% |
| 0.40-0.50 | 25 | 46% | 0% | 45.9% |
| 0.50-0.60 | 43 | 56% | 0% | 55.5% |
| 0.60-0.70 | 30 | 63% | 0% | 63.5% |
| 0.70-0.80 | 26 | 76% | 0% | 76.5% |
| 0.80-0.90 | 22 | 85% | 0% | 85.3% |
| 0.90-1.00 | 18 | 93% | 0% | 93.5% |
On-chain verification: wallet age 44 days, 2576 txs, provenance grade B. Bot score: 100/100, wash trading score: 0/100.
Polymarket on-chain coverage: $1,009 in / $0 out across 33 withdrawal tx since 2026-04-17.
3100 total trades across 195 markets.
195 bets on resolved markets available for calibration scoring.
Calibration error: 60.9% — needs improvement.
Skill: 0/100 (calibration quality). Variance: 13/100 (higher = more volatile returns).
Brier Skill Score: -65.0% vs naive baseline (>0% = better than always predicting base rate).
Brier decomposition: REL=0.4119 RES=0.0000 UNC=0.0000.
Log loss: 1.1394 (skill: -14309.4% vs naive). Lower log loss = better calibration on rare events.
Timeliness: avg entry 1.3 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/37 [CI95: D→D, 35-39] (195 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.