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 | 8 | 7% | 50% | 43.4% |
| 0.10-0.20 | 8 | 13% | 50% | 36.9% |
| 0.20-0.30 | 26 | 25% | 31% | 5.5% |
| 0.30-0.40 | 30 | 35% | 47% | 11.2% |
| 0.40-0.50 | 20 | 46% | 20% | 26.3% |
| 0.50-0.60 | 34 | 56% | 35% | 20.3% |
| 0.60-0.70 | 30 | 65% | 40% | 25.3% |
| 0.70-0.80 | 26 | 76% | 54% | 21.8% |
| 0.80-0.90 | 8 | 85% | 75% | 10.1% |
| 0.90-1.00 | 6 | 94% | 33% | 60.6% |
Skill measures calibration quality (0-100). Variance measures return volatility (0-100, higher = more volatile).
On-chain verification: wallet age 15 days, 2078 txs, provenance grade B. Bot score: 10/100, wash trading score: 0/100.
Polymarket on-chain coverage: $337 in / $0 out across 31 withdrawal tx since 2026-04-14.
3100 total trades across 53 markets.
196 bets on resolved markets available for calibration scoring.
Calibration error: 20.9% — needs improvement.
Skill: 37/100 (calibration quality). Variance: 100/100 (higher = more volatile returns).
Brier Skill Score: -17.0% vs naive baseline (>0% = better than always predicting base rate).
Brier decomposition: REL=0.0575 RES=0.0147 UNC=0.2416.
Log loss: 0.7994 (skill: -18.2% 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/58 [CI95: C→C, 56-59] (196 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.