FILE 07 · RESEARCH ARCHIVEAs of Aug 19, 2026

Research Archive & Corrections.

Published research, material revisions, and corrections behind MSR Decode’s July 2026 Polymarket wallet forensics, dated and permanently accessible. The recurring monthly leaderboard audit and forward-tested ledger described below were retired on August 26, 2026; material corrections were recorded here as they were identified, dated, without altering the original entry. Nothing below is edited after the fact — corrections are appended, never erased.

30 wallets audited · 4 profit claims checked · 1 confirmed · 3 refuted · 16 items in our public register of unreconciled numbers and corrections

Each figure is computed from a published artifact rather than kept as a tally, and the pre-publish gate refuses this page if the two disagree: the scorecard, the weather-bot reconciliation, the BNB ledger, and the register. Two of the sixteen are resolved and stay listed; entries leave that file by being reconciled, never by being deleted. Not every one is a mistake of ours — some are numbers we rely on that we cannot reconcile.

Cycle July 2026 Corrections dated & append-only As of Aug 19, 2026

Historical record · snapshot through August 19, 2026

Shipped

What actually went live this cycle, in order.

Jul 15

The Leaderboard Scorecard — July 2026 cycle

Audited Polymarket’s top 25 wallets by 30-day profit. 10 of 25 earned a Real-edge, copyable rating at launch; the dated Jul 16 re-stamp entry in Section 02 below moved four of them to Insufficient evidence, and the standing count is 6 of 25. Re-run monthly. Read the scorecard →

Jul 15

Reconstruction Workpaper published — methodology v1 frozen

Every per-wallet reconstruction error published, spec recon-error-v1 frozen the same day. Headline: median gap 5.1 points, aggregate 98.7% of displayed dollars, across the 10 wallets in the structural set — and 15 of 25 refused rather than guessed. Read the workpaper →

Jul 2026

Two case files published

The Leaderboard Autopsy — a top wallet’s celebrated “edge” decomposed, fill by fill, into redemption arbitrage a copier could never have captured. Sample Decode: myzbsq — a $249K → $1.22M World Cup bet reconstructed to 99.6% of Polymarket’s displayed profit (MSR’s cleanest case, a closed single-market wallet) and classified a one-shot, not a repeatable edge. Autopsy → · myzbsq →

Jul 2026

Evidence-confidence gate shipped

A second, independent label — Verified / Provisional / Insufficient data — grades fetch completeness separately from the six-way verdict, so a copyable stamp can no longer imply a complete fetch it doesn’t have. Of the 25 as shipped: 6 Verified, 4 Provisional, 15 Insufficient data. (Correction, Aug 4, 2026: three of those labels have moved to Provisional under the gate’s own rule — the split now reads 3 / 7 / 15; see U12 below.) See the rule →

Jul 19

Prospective Copyability Ledger launched

Every Real-edge verdict now gets one ledger entry: follower-accessible prices frozen at entry, outcomes published at 24h / 7d / 30d — gains and losses alike. Cohort 1 froze 6 entries on Jul 19; the d7 checkpoint published on schedule Jul 26, 2026, and the d30 checkpoint followed Aug 19, 2026, about 1.2 days late — cohort 1’s ledger is now complete. Read the ledger →

Jul 24

Case file: the BNB 5-minute wallet

The viral wallet that turned $449K into $1.37M in 3.1 hours, reconstructed to the cent: net realized $917,311.53, matching the viral “~$917k” claim to within $311.53. Read the case →

Jul 25

Method report: an endpoint default that undercounted fills, corrected

An integrity check initially flagged three top-leaderboard wallets for $926,462.58 of profit on positions they never paid for. Polymarket’s /trades endpoint defaults to takerOnly=true, so it silently returns only part of a wallet’s fills — and in markets with no order book that hides nearly all of them. Re-fetched properly, the 50 largest flagged markets within each of the three wallets — 150 of the 4,725 flagged, not a pooled ranking — reconciled 150 of 150. All three traders were ordinary. Read the corrected method report →

Jul 26

The d7 ledger checkpoint published on schedule

Cohort 1’s 7-day outcome came back identical to the 24-hour marks: hot2trot $1,475 (+47.5%), highnetworth $1,665 (+66.5%) on a simulated $1,000 each, the other four flat because they had nothing a follower could buy. Every copyable position had already resolved by the roughly 26-hour checkpoint (computed 2h late per the ledger) — so the copyable window on this cohort was days shorter than a week; this checkpoint bounds the window rather than measuring it, a limitation of the checkpoint interval published as one. Read the checkpoint →

Aug 16

Two forensic notes published

NYC Council opened a public marketing-practices review of four prediction-market platforms; MSR Decode has not independently reconstructed any specific wallet, trade or video named in the reporting that preceded it. Separately, a creator’s viral Polymarket win was reported staged, with more than 50 real accounts said to have made the same bet and lost — rebuilt fill by fill: 48 wallets, none profited. Note 01 → · Note 02 →

Aug 18

Forensic Note 03: a federal insider-trading complaint, reconciled

The CFTC’s complaint against Gannon Van Dyke states its figures as approximations and floors. MSR Decode pulled the named wallet’s complete Polymarket record: the approximations hold, the floors are exceeded, and press-reported prosecution figures reproduce to the dollar. Read the note →

Aug 19

The d30 ledger checkpoint completes cohort 1

Cohort 1’s 30-day outcome matches the 7-day mark exactly: hot2trot $1,475 (+47.5%), highnetworth $1,665 (+66.5%) on a simulated $1,000 each, the other four flat — no copyable position remained open past the 7-day mark to move. Computed about 1.2 days late (see the ledger for why). All three checkpoints (24h / 7d / 30d) are now on the record for every cohort-1 entry. Read the checkpoint →

Aug 26

Program-status update: the July leaderboard program moved to historical status

MSR Decode broadened from a wallet-copyability research program into a general independent prediction-market forensic research desk. As part of that repositioning: the July 2026 Scorecard remains frozen and public; the completed Prospective Ledger (cohort 1) remains public; the wallet-reconstruction Workpaper remains public; MSR Decode is not currently running a recurring monthly Top-25 cycle. This supersedes, from this date forward, the recurring-cycle commitment in Section 03 below (“Monthly scorecard refresh”) and the dataset/API concept referenced there (“Dataset access on request”) — that commercial dataset/API concept is not an active MSR Decode offering; current commissioned services are described on the Capabilities page. Both original entries remain visible below because this archive is append-only. Read the current investigative standard →

Corrections & refusals

Material corrections, dated and append-only, and the wallets refused a rating because the data didn’t support one.

15 of 25 wallets refused rather than guessed. “That 15 of 25 of the most-copied wallets on the leaderboard cannot be verified from public data is, in our view, the most important number” on the workpaper — a statement about what a follower can know, not an allegation about those traders. See every rationale, verbatim →

Jul 16, 2026 — incomplete-fetch disclosure. One audited wallet’s fetch held at most 1.7% of the fills later measured for that 30-day window. A sweep of the cycle’s other 24 wallets found 16 consistent with their published counts, 3 incomplete and rebuilt, and 5 inconclusive. See the method →

Jul 16, 2026 — four wallets re-stamped. CandleHammerDrums, asparagus2012, filthybetz and wco26 moved from Real edge to Insufficient evidence after closer review found their agreement with the displayed figure rested on Polymarket’s own open-book valuation, or fell outside the published ±15% band. Scored set: 8 of 25; refused: 15 of 25. Read the workpaper →

Jul 25, 2026 — phantom-positions finding corrected. An initial check flagged three top-leaderboard wallets for $926,462.58 of profit that did not reconcile against their trades. The cause was a method defect, not the wallets’ trading: Polymarket’s /trades endpoint defaults to takerOnly=true and silently returns only part of a wallet’s fills, which in markets with no order book hides nearly all of them. Re-fetched per-market from /activity, the 50 largest flagged markets within each of the three wallets — 150 of the 4,725 flagged, not a pooled ranking — reconciled 150 of 150, backed by $1,361,899.05 of real buying across those markets. All three traders were ordinary. Read the corrected method report →

Jul 26, 2026 — corrected fetch applied to the frozen headline set. Decision taken Aug 2. Run against the six Real-edge wallets: five hold under the corrected fetch; hot2trot moves from 110.8% to 133.7% of displayed profit — from 10.8 points inside the ±15% band to 33.7 points from parity, 18.7 beyond the band’s edge. The July cycle remains frozen on the methodology used at publication; the corrected fetch is the method of record going forward. The complete dated decision record is preserved in U7 and U8. Read U7 and U8 →

Aug 3, 2026 — sampling description corrected sitewide. A filed CFTC comment described the 150-market re-fetch as a pooled ranking of “the 150 largest of the 4,725 markets the check had flagged.” The selection was not pooled: it was the 50 largest within each of three wallets. The federal filing cannot be amended, so the register entry is the correction of record; the wording was corrected on the case page, this page, and the homepage. Count, reconciliation and conclusion are unaffected — only the description of how the sample was drawn. Read U9 →

Aug 4, 2026 — Real-edge verdict line scope corrected. The verdict line on the six Real-edge wallet pages, the homepage and the checker stated that whether the edge is copyable by you (“sizing, timing, liquidity”) was “the one this rating is about.” No order-acceptance, depth, or executable-quote test was ever run, so the line now states the rating does not answer that question. The definition, where the label appears →

Aug 4, 2026 — three Verified evidence labels moved to Provisional (U12). The evidence gate caps a wallet at Provisional after a measured, after-the-fact fetch truncation, regardless of how well its numbers reconcile. The Jul 26 corrected re-fetch found the selection-time fetch behind hot2trot, muchobliged and weatherman12 short by 22.9, 12.1 and 5.6 points of displayed profit; the same measurement confirms palegrit and highnetworth to within a twentieth of a point. No verdict, count, or frozen figure changed — only the evidence label. Read U12 →

Aug 5, 2026 — “through the order book” corrected to “by the wallets’ own purchases” (U16). The 150 re-fetched markets carry enable_order_book: false and are matched by Polymarket’s own operator, not an order book. The purchases and reconciliation are unchanged; the mechanism described was not accurate, and was corrected in four places, including a filed CFTC comment that cannot itself be amended. Read the register →

21 of 24 markets closing net-positive in the BNB 5-minute wallet, reconstructed fill by fill — not the viral “23 of 24” claim. The most charitable reading MSR could construct still tops out at 22 of 24; no reading in the dataset reaches 23. Read the correction →

4 of 6 — a different count, on purpose. On day one of the Prospective Copyability Ledger, 4 of the ledger’s first 6 Real-edge entries had zero follower-accessible positions the instant their prices were frozen — every position had already resolved or gone redeemable, or the wallet held nothing at all. This is an accessibility count about a reader acting after publication, not a re-statement of the 6 of 25 verdict count above — the two measure different questions and neither wallet re-uses the other’s figure. Read what day one shows →

What’s next

Only commitments already on the record — nothing new promised here. Both lines below were superseded Aug 26, 2026 by the program-status update in Section 01 above; they are preserved here, not deleted, because this archive is append-only.

Monthly scorecard refresh. The top-25 audit re-runs monthly; the next cycle’s accuracy rule (recon-error-v2) is already pre-registered and frozen ahead of the results. Superseded Aug 26, 2026 — no recurring monthly cycle is currently running; see the program-status update above. Scorecard →
Dataset access on request. Structured historical-and-prospective verdicts, including the append-only ledger, remain under evaluation for research, risk & compliance teams — ask about intended use. Superseded Aug 26, 2026 — this dataset/API concept is not an active MSR Decode offering; current commissioned services are described on the Capabilities page. Capabilities →

Nothing above is a new promise. Every line in this section is a date or a policy already published on the page it links to; this page adds no figure and no commitment that didn’t already exist there.