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 | 15% | 0% | 15.4% |
| 0.30-0.40 | 6 | 34% | 0% | 33.6% |
| 0.40-0.50 | 52 | 45% | 0% | 45.2% |
| 0.50-0.60 | 4 | 51% | 0% | 50.8% |
| 0.70-0.80 | 1 | 74% | 0% | 74.1% |
Skill measures calibration quality (0-100). Variance measures return volatility (0-100, higher = more volatile).
On-chain verification: wallet age 23 days, 2999 txs, provenance grade C. Bot score: 0/100, wash trading score: 50/100.
Polymarket on-chain coverage: $11,103 in / $0 out across 91 withdrawal tx since 2026-03-30.
3100 total trades across 63 markets.
64 bets on resolved markets available for calibration scoring.
Calibration error: 44.5% — needs improvement.
Skill: 7/100 (calibration quality). Variance: 12/100 (higher = more volatile returns).
Brier Skill Score: 19.0% vs naive baseline (>0% = better than always predicting base rate).
Brier decomposition: REL=0.2020 RES=0.0000 UNC=0.0000.
Log loss: 0.5968 (skill: -7447.0% 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/53 ± 8 (medium confidence, 64 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.