VIGIL Trust Score for the AI Agent Economy
Extra-Small-Implement
0xf27d40745542dc871e127acff3a1c9d3910d9a88
Polymarket Trader
DANGER
High probability of further losses.
Calibration Analysis
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 | 538 | 1% | 0% | 1.2% |
| 0.90-1.00 | 462 | 99% | 0% | 98.7% |
Calibration Error: 46.3%
Reliability (CAL)
0.4502
Lower = better calibrated
Resolution (RES)
0.0000
Higher = stronger opinions
Brier Skill Score
-80.1%
vs naive baseline
Log Loss
2.0300
Skill: -25572.5% vs naive — sensitive to rare events
Skill & Variance Analysis
Skill measures calibration quality (0-100). Variance measures return volatility (0-100, higher = more volatile).
Signals
✓ Well-diversified: 102 unique markets
✓ On-chain: human-like trading patterns
✓ On-chain: $11,805 total USDC.e received from Polymarket contracts across 209 tx (Polygon)
⚠ Weak Brier score: 0.4502
⚠ Overconfidence bias: 0.463
⚠ Brier Skill Score: -80.1% — performing worse than naive baseline
⚠ Low resolution: forecasts cluster near base rate — no genuine opinions
⚠ Log loss penalty: severe overconfidence on wrong bets detected
⚠ Penny-lottery strategy: 54% of bets at sub-$0.10 — score penalized by 15 pts
⚠ Net loss: -$44164 total PnL
⚠ Low win rate: 0%
⚠ Concentrated portfolio: 100% in single market
⚠ PnL divergence: $44327 gap between API and on-chain USDC flows
On-Chain Verification (Polygon)
Protocols: USDC.e (Bridged)
PnL divergence: $44326.79 gap between API and on-chain USDC
✓ Heavy on-chain activity: 2000 txs
✓ Significant USDC flow: $32,921
✓ Human-like transaction patterns
⚠ Limited counterparties: only 0 unique addresses
⚠ No contract interactions despite activity — possible EOA-only transfers
Reasoning
On-chain verification: wallet age 97 days, 2000 txs, provenance grade C. Bot score: 0/100, wash trading score: 0/100.
Polymarket on-chain coverage: $11,805 in / $0 out across 209 withdrawal tx since 2026-04-18.
3100 total trades across 102 markets.
1000 bets on resolved markets available for calibration scoring.
Calibration error: 46.3% — needs improvement.
Skill: 5/100 (calibration quality). Variance: 15/100 (higher = more volatile returns).
Brier Skill Score: -80.1% vs naive baseline (>0% = better than always predicting base rate).
Brier decomposition: REL=0.4502 RES=0.0000 UNC=0.0000.
Log loss: 2.0300 (skill: -25572.5% vs naive). Lower log loss = better calibration on rare events.
What Does F/18 Mean?
No demonstrated forecasting skill. This trader performs at or below random chance based on available data.
Confidence: F/18 [CI95: F→F, 16-20] (1000 resolved bets). This score is highly reliable — enough resolved bets to be confident.
Methodology: Brier Score Decomposition (Murphy 1973), Log Loss, On-Chain USDC Verification. Same approach used by IARPA to identify superforecasters.