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 |
|---|---|---|---|---|
| 0.30-0.40 | 1 | 31% | 0% | 31.4% |
| 0.40-0.50 | 30 | 46% | 0% | 46.5% |
| 0.50-0.60 | 7 | 52% | 0% | 51.7% |
| 0.60-0.70 | 2 | 67% | 0% | 66.8% |
Skill measures calibration quality (0-100). Variance measures return volatility (0-100, higher = more volatile).
On-chain verification: wallet age 6 days, 2989 txs, provenance grade C. Bot score: 0/100, wash trading score: 50/100.
Polymarket on-chain coverage: $10,120 in / $0 out across 59 withdrawal tx since 2026-04-14.
3100 total trades across 39 markets.
40 bets on resolved markets available for calibration scoring.
Calibration error: 48.0% — needs improvement.
Skill: 2/100 (calibration quality). Variance: 11/100 (higher = more volatile returns).
Brier Skill Score: 6.4% vs naive baseline (>0% = better than always predicting base rate).
Brier decomposition: REL=0.2337 RES=0.0000 UNC=0.0000.
Log loss: 0.6617 (skill: -8267.9% 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 ± 15 (low confidence, 40 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.