VIGIL Trust Score for the AI Agent Economy
mnr17
0x0208b1b43b10f3c6b75cd9da18619180d72ac2c2
Polymarket Trader
SOLID
Process is working. Metrics hold up under scrutiny.
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 | 968 | 1% | 0% | 1.5% |
| 0.90-1.00 | 10 | 96% | 0% | 96.2% |
Calibration Error: 2.5%
Reliability (CAL)
0.0097
Lower = better calibrated
Resolution (RES)
0.0000
Higher = stronger opinions
Brier Skill Score
96.0%
vs naive baseline
Log Loss
0.0537
Skill: -578.5% 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
✓ Strong Brier score: 0.0099
✓ Brier Skill Score: 96.0% better than naive baseline
✓ Strong live edge: 22 open positions trending profitable
✓ Well-diversified: 806 unique markets
✓ On-chain: human-like trading patterns
⚠ Low resolution: forecasts cluster near base rate — no genuine opinions
⚠ Log loss penalty: severe overconfidence on wrong bets detected
⚠ Receive-only wallet: zero outbound transactions — possible proxy/settlement address
⚠ Net loss: -$362 total PnL
⚠ Low win rate: 0%
⚠ On-chain: wallet only 4 days old on Polygon
⚠ PnL divergence: $808 gap between API and on-chain USDC flows
⚠ On-chain: wash trading signal — 100% of txs with single counterparty
On-Chain Verification (Polygon)
Protocols: USDC.e (Bridged)
PnL divergence: $807.78 gap between API and on-chain USDC
✓ Heavy on-chain activity: 2001 txs
✓ Human-like transaction patterns
⚠ Very new wallet: 4 days old — possible sybil
⚠ Limited counterparties: only 1 unique addresses
⚠ Receive-only wallet: never sent a transaction
⚠ Wash trading signal: 100% of txs with single counterparty
Reasoning
On-chain verification: wallet age 4 days, 2001 txs, provenance grade D. Bot score: 0/100, wash trading score: 80/100.
1115 total trades across 806 markets.
978 bets on resolved markets available for calibration scoring.
Calibration error: 2.5% — excellent.
Skill: 57/100 (calibration quality). Variance: 42/100 (higher = more volatile returns).
Brier Skill Score: 96.0% vs naive baseline (>0% = better than always predicting base rate).
Brier decomposition: REL=0.0097 RES=0.0000 UNC=0.0000.
Log loss: 0.0537 (skill: -578.5% vs naive). Lower log loss = better calibration on rare events.
What Does B/70 Mean?
This trader shows genuine forecasting skill with a meaningful edge across multiple markets.
Confidence: B/70 ± 3 (high confidence, 978 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.