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 | 18 | 5% | 22% | 16.8% |
| 0.10-0.20 | 14 | 14% | 29% | 14.4% |
| 0.20-0.30 | 6 | 24% | 0% | 23.7% |
| 0.30-0.40 | 6 | 34% | 33% | 0.2% |
| 0.40-0.50 | 6 | 45% | 67% | 21.9% |
| 0.50-0.60 | 10 | 53% | 40% | 12.7% |
| 0.60-0.70 | 14 | 65% | 29% | 36.7% |
| 0.70-0.80 | 6 | 77% | 0% | 77.3% |
| 0.80-0.90 | 6 | 85% | 33% | 51.4% |
| 0.90-1.00 | 24 | 95% | 0% | 94.9% |
Skill measures calibration quality (0-100). Variance measures return volatility (0-100, higher = more volatile).
On-chain verification: wallet age 84 days, 2000 txs, provenance grade C. Bot score: 0/100, wash trading score: 0/100.
Polymarket on-chain coverage: $2,096 in / $0 out across 76 withdrawal tx since 2026-02-05.
3100 total trades across 101 markets.
110 bets on resolved markets available for calibration scoring.
Calibration error: 40.6% — needs improvement.
Skill: 16/100 (calibration quality). Variance: 100/100 (higher = more volatile returns).
Brier Skill Score: -146.8% vs naive baseline (>0% = better than always predicting base rate).
Brier decomposition: REL=0.2751 RES=0.0322 UNC=0.1706.
Log loss: 1.3479 (skill: -156.9% vs naive). Lower log loss = better calibration on rare events.
This trader shows some skill signal, but not enough to clearly distinguish from luck. More data needed.
Confidence: C/51 [CI95: D→C, 48-54] (110 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.