CopyGrade
Methodology

How the Copy Score is built.

Every Polymarket wallet gets a single 0–100 Copy Score — a verdict on how copyable it is right now. The score is the weighted blend of five factors, each scored 0–100 and recomputed on each sync. A severe Farming Risk finding can veto an otherwise-strong score.

Last reviewed 2026-07-01
The model
Five factors, and how they're weighted
Edge Authenticity
28%
Separates genuine skill from luck and variance. A wallet with a short, lucky streak scores low here even if its PnL looks strong.
Risk-Adjusted Return
22%
A Sharpe-style measure of return per unit of risk taken — not raw PnL, which rewards reckless sizing.
Drawdown Resilience
18%
How the wallet behaves in losing streaks: depth of drawdowns and how quickly it recovers.
Consistency / Regime Stability
16%
Whether the edge holds across different market regimes and time windows, or only existed in one favourable stretch.
Farming Risk
16%
A forensic check for behaviour that turns copiers into exit liquidity — iceberg accumulation, decoy clusters, stealth YES/NO merges, and self-trade washing.
Total weight
100%
The veto
How farming caps the score
Clean
No cap
No detector tripped — the score reflects the full five-factor blend above.
Watch
Capped at 64
A moderate-confidence finding (or a couple of borderline ones). The wallet can never read as a strong copy candidate until it clears.
••
Severe
Capped at 30
A high-confidence signature, or several moderate ones by weighted severity. The score is vetoed to the Avoid band, however good the headline PnL.

Each finding is weighted by severity (how bad if real) and confidence(how far past its threshold the pattern sits); the level rolls up that weighted total, so one unmistakable signature can veto while two borderline ones may only warn. Each detector's definition and exact threshold lives on the farming detection page.

The output
From score to verdict
80–100
Strong copy candidate
65–79
Copy candidate
50–64
Copy with caution
30–49
Avoid
0–29
Avoid — farming risk
The receipts
Track record: scores vs. realized outcomes
Every score snapshot is graded later against what a simulated copier would actually have realized over the following weeks. These tables are computed live from that grading — they are the model's report card, published whichever way they point.

What the metric is: for each snapshot, a simulated copier mirrors only the positions the wallet opens after that moment (buy-time windowed), and the outcome is the realized return of the round-trips that close within the horizon, net of modeled fees. What it isn't: the realized figure counts only round-trips that close inside the horizon, which understates slow, patient strategies. Beside it we now show a mark-to-marketestimate that also marks the positions still open at the horizon's end to current price — it exists for recently-matured cohorts and reads “—” for older windows (marking a long-closed window to today's prices would be a fiction, not a measurement). Dormant windows — where the wallet did nothing for a copier to mirror — are excluded, so the table is not diluted by no-observation rows. Only windows the sync observed end-to-end count, and a band is published only once it holds at least 30 outcomes.

7-day horizon
n = 9,538
Score band at snapshotnMedian returnMean returnMedian MTMMean MTM% positive
75–1002410.00%+0.42%−1.31%−14.45%26%
50–742,6060.00%+2.15%−0.24%−12.39%20%
30–496,5120.00%+2.64%−1.03%−10.78%27%
0–291790.00%−2.29%−9.51%−22.44%10%
30-day horizon
n = 3,327
Score band at snapshotnMedian returnMean returnMedian MTMMean MTM% positive
75–1001230.00%−3.58%−6.72%−20.55%33%
50–741,1830.00%+2.21%−0.71%−11.94%32%
30–491,9710.00%+2.81%−1.12%−11.74%33%
0–29500.00%−1.23%−0.14%−17.45%12%

Until these tables show a persistent positive relationship, the Copy Score remains a documented v1 heuristic — our assessment methodology, not a proven predictor. Aggregates only; no individual wallet's outcome is published. Last computed Jul 12, 2026 · refreshed daily.

These are the current heuristic v1 priors — graded against real wallet outcomes in the track record above, so the numbers can change as the model is recalibrated. CopyGrade is analysis-only — it never executes trades or holds funds. Not financial advice.

See a sample verdictStart free