The top of the leaderboard is mostly bots you can't copy.
We decoded the wallets Polymarket ranks as "smart money." Their profits are real. Their strategies are uncopyable — and by displayed PnL alone, you can't tell them apart from a lucky coin flip.
Why "who's winning?" is the wrong question
Every leaderboard and copy-trading tool ranks wallets by profit. But profit doesn't tell a copier the one thing that decides whether they'll make money: is this edge reproducible, or is it a bot doing something you structurally cannot?
Flagship decode — "debased"
Verdict: artifact — uncopyable
A leaderboard sees a high-volume, seven-figure-profit "smart money" wallet. The full fill history says bilateral market-maker + near-certain-outcome harvester, not a directional trader:
- 65% of buy volume at ≥ $0.99 — buying outcomes already priced as near-certain, to capture the last fraction of a cent to $1.00.
- 67% of its markets traded on both sides (YES and NO) — the signature of merge/spread arb, not a directional call.
- Average buy price $0.78 across geopolitical markets (Russia–Ukraine ceasefire, US–Iran, Trump nominations) — expensive, near-resolved entries.
market: "Will Portugal win the 2026 FIFA World Cup?"
BUY 126,195 NO @ 0.9980 = $125,943 · 2026-07-06 20:58
BUY 8,609 NO @ 0.9980 = $8,592 · 20:58
BUY 2,804 NO @ 0.9900 = $2,776 · 20:54
→ 151,227 NO shares @ avg 0.9975 for $150,856, all inside a 4-minute window (three largest fills shown)
market: "Will USA win the 2026 FIFA World Cup?"
BUY 2,864 NO @ 0.9950 = $2,850
It isn't predicting Portugal or the USA lose. It's buying an all-but-settled outcome for a locked ~0.25¢ spread, at size. A copier mirroring this pays $0.9975 for $1.00 — no edge, all tail risk.
Correction, added 2026-07-16 — the arithmetic on this receipt. Every dollar amount here is shares × price, rounded half-up to the dollar; each line is now checked against that rule by tools/audit.mjs and the build fails if one drifts. Two were wrong when this page went up. The largest fill printed $125,950 where 126,195 × 0.9980 is $125,942.61 — $7.39 high, and reachable by no rounding model, while its two sibling lines rounded correctly to the dollar. The USA line printed $2,849 where 2,864 × 0.9950 is $2,849.68, which truncates to 2,849 but rounds to 2,850 — so the block was using two different rules at once. The subtotal line is not an error and is left alone: 151,227 × 0.9975 appears to miss by $7.07, but 0.9975 is a 4-decimal display of 150,856/151,227 = 0.99754673, and a 4dp average carries enough slack ($7.56 across this size) to cover it — the product is the wrong relationship to check, the quotient is the right one.
What this receipt cannot prove. These fills come from a 2026-07-11 pull that retained no artifact in this repo, and the wallet is published here only as 0x24c8cf69…823e1 — a truncated address that cannot be looked up. So the three numbers on each line are internally consistent and not independently verifiable: correcting the arithmetic does not make them sourced. The gap is logged in the unreconciled-items table rather than explained away, per DOCTRINE rule 4. Also unchanged: debased's exact net PnL is not decision-grade — at 444 markets / 74K fills / heavy merges, reconstruction needs deeper accounting. The strategy classification rests on the prices and sizes, not on these dollar totals.
Full-history decode — the rest of the leaderboard
We pulled the fill history for each of the next five — 14,000 to 469,000 fills apiece — and ran the same classification debased got. The story holds: the biggest names are market-makers, arbitrageurs, and lottery players whose edge is structure or variance, not calls a retail account could follow.
| Wallet | Displayed · Polymarket full history, pulled 2026-07-11 | Full history as of the 2026-07-11 pull | Signature (full fills) | What it really is | Retail-copy verdict |
|---|---|---|---|---|---|
| swisstony | $18.6M | withdrawn 2026-07-16 — a re-fetch counts 2,918,281 fills, >200× this. See the correction above. | 35% bilateral · 11% ≥$0.99 | Mixed market-maker / arb flow | not copyable |
| RN1 | $11.1M | 10,062 mkts · 324K fills | 56% bilateral | Pure bilateral market-maker — spread capture | not copyable |
| ImJustKen | $3.3M | 2,140 mkts · 135K fills | 59% bilateral · 14% ≥$0.99 | Bilateral MM + redemption-arb tail | not copyable |
| debased | $1.67M | 444 mkts · 74K fills | 67% bilateral · 65% ≥$0.99 | Near-certain-outcome harvester (full decode, §02) | not copyable |
| tripping | $89.5K | 656 mkts · 469K fills | 80% of buys sub-cent ≤$0.02 | Sub-cent longshot lottery — pure variance | variance |
| suntori | −$4.6M | 1,484 mkts · 53K fills | 26% bilateral · high volume | High-volume net loser | avoid |
Why no reconstructed-PnL column here: at hundreds to 10,000+ markets, exact cashflow reconstruction off the public feed is not decision-grade — merges, redemptions, and per-market API limits dominate the cashflow view. That limitation doesn't touch the copy verdict: every signature above is computed from the fill history we pulled and is enough to classify whether a retail user could realistically copy the wallet. Displayed profit is Polymarket's own leaderboard figure, pulled 2026-07-11 — this line said 2026-07-12 until 2026-07-16, contradicting the five other places on this page that date the same pull, including the column header directly above it.
The contrast — what real, copyable edge looks like
Not everyone is a bot. When we reconstruct a genuinely directional wallet, our engine ties out to Polymarket's own number — proving the method, and letting the difference between "real" and "artifact" show cleanly.
- Theo4 — real edge — the 2024 election whale. $13.8M+ into Trump/Harris presidency & popular-vote markets, held to resolution. Directional conviction. Copyable in principle (if you have the conviction and the bankroll).
- myzbsq — one lucky bet — reconstructed to 99.6% of its displayed $1.22M. But that $1.22M is a single bet: bought "No Argentina" on one World Cup match at ~$0.15, rode it to $0.99, sold. Not a strategy — a coin flip that landed. Copying "myzbsq +$1.2M" copies the outcome, not an edge.
For an operator reading this
Your leaderboard ranks these wallets identically. Your users copy the number and lose. The closest anyone has measured this is in crypto, not prediction markets: across 1,486 copy-trading leaders on Binance, Bybit and MEXC, 97.04% were green on their own PnL and only 43.61% produced positive PnL for their followers.1 Different venue, different instrument — but it is the same gap this page is about, and nobody has published the prediction-market version.
MSR Decode is a Replicability Score: a per-wallet API that labels each one real / artifact / farming with the on-chain evidence behind it. You embed it as a column next to your PnL ranking. Your leaderboard stops misleading, your copiers stop torching money, and you have a feature no competitor ships.
Want the full decode of your current #1 wallet, free? We'll send the receipts. If it's useful, it's a feed you embed.
1. Yieldfund, “Is Copy Trading Profitable? A 90-Day Multi-Exchange Study” (Vlad Hatze, published 24 Nov 2025): 100,236 copier outcomes across 1,486 lead traders on Binance, Bybit and MEXC, over a 90-day window to 24 Oct 2025. Both percentages are over all leaders in that sample, which is the study's own denominator — an earlier version of this page said “~56% of even profitable leaders,” a figure the study does not support: it is 100 − 43.61 re-labelled onto a smaller population, and the honest range for that population is 55.1–58.1%, not a point estimate. Corrected 2026-07-16. This is self-reported vendor research and the leader-level data is not published, so we cannot recompute it; we cite it as the closest measurement that exists, over a population that is not ours.