The CopyGrade Score model is tuned continuously against real wallet outcomes. Every material change to how a score is computed — the scoring math, the farming detectors, the realistic-return and simulator arithmetic, the validation reads, and the trade data that feeds them — is recorded here, dated, so the way a score is computed is never a moving target you can't see. Scoring changes only: pricing, notification and product changes are not logged here.
The daily cross-wallet pass had been re-checking a different sliver of wallets each day and discarding every finding it did not re-check, so a decoy or coordination flag on a wallet could lift one morning and return the next. A finding now stands until the pass re-evaluates that wallet with all of the cluster's members loaded.
The forward-return table that has lived on the methodology page since June now has its own page, /track-record, led by the honest headline (three 30-day reads, no rank correlation yet) and joined by a second ledger no one else publishes: whether the wallets we graded into each band 30 and 90 days ago are still trading and still in that band, and whether our farming-risk flags held.
A wallet whose recorded trades span fewer than 30 days from first to last fill is capped at the caution band (64) — the same ceiling the gate already applies under 20 recent trades. A short record cannot read as a copy candidate however clean its numbers.
An ingestion defect had been writing duplicate copies of the same Polymarket fill, leaving roughly a quarter of stored trades as phantom rows. Every wallet has been re-scored on the corrected tape, and scores moved — mostly downward in farming risk.
A rename, not a model change: the 0–100 metric formerly called the Copy Score is now the CopyGrade Score. Same inputs, same weights, same numbers — only the name is new.
The fee-adjusted (“realistic”) edge previously used a simplified fee assumption. It now applies Polymarket's actual published rates, which makes realistic edge lower — and more honest — for most wallets.
A wallet whose stored history does not reach back across a full window used to have its windowed stats rendered as if it did. Those figures are now labelled as floors, from the wallet's real tape start.
The first 30-day forward cohort matured. The honest read: no measurable rank relationship yet between CopyGrade Score and the realized 30-day return of a simulated copier. We committed to publishing this either way, before we saw it.
Farming flags are heuristic assessments of public on-chain data, not verified findings — the wording now says so, and wallet operators get a visible way to dispute one.
The first pass over real Polymarket data showed the v1 model — tuned on synthetic patterns — over-flagged farming and over-rewarded thin, no-edge records. Recalibrated.
The simulator's equity curve was a deterministic synthesis that never touched the wallet's trades. It now replays the wallet's actual fills.
Farming risk moved from a flat count of signals to a severity-weighted rollup, and gained two single-wallet detectors plus a cross-wallet pass.
The CopyGrade Score's return and risk factors were rebuilt to reflect what a copier actually keeps, not the headline number.
How the score is built is documented in full on the methodology page, and how it has performed on the track record page.