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 |
|---|---|---|---|---|
| 0.10-0.20 | 1 | 18% | 100% | 81.8% |
| 0.20-0.30 | 1 | 25% | 100% | 75.0% |
| 0.30-0.40 | 1 | 36% | 100% | 64.2% |
| 0.40-0.50 | 6 | 45% | 67% | 22.1% |
| 0.50-0.60 | 4 | 56% | 75% | 19.3% |
| 0.60-0.70 | 3 | 66% | 100% | 34.0% |
| 0.70-0.80 | 6 | 76% | 83% | 7.0% |
| 0.80-0.90 | 6 | 84% | 100% | 15.5% |
| 0.90-1.00 | 1 | 95% | 100% | 5.0% |
Skill measures calibration quality (0-100). Variance measures return volatility (0-100, higher = more volatile).
On-chain verification: wallet age 9 days, 146 txs, provenance grade D. Bot score: 0/100, wash trading score: 0/100.
90 total trades across 72 markets.
29 bets on resolved markets available for calibration scoring.
Calibration error: 23.2% — needs improvement.
Skill: 35/100 (calibration quality). Variance: 38/100 (higher = more volatile returns).
Brier Skill Score: -56.7% vs naive baseline (>0% = better than always predicting base rate).
Brier decomposition: REL=0.0899 RES=0.0183 UNC=0.1189.
Log loss: 0.5475 (skill: -36.5% vs naive). Lower log loss = better calibration on rare events.
Timeliness: avg entry 6.2 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 ± 15 (low confidence, 29 resolved bets). Low confidence — take this score with a grain of salt until more markets resolve.
Methodology: Brier Score Decomposition (Murphy 1973), Log Loss, On-Chain USDC Verification. Same approach used by IARPA to identify superforecasters.