How to copy trade on Polymarket: the complete guide
Polymarket has no native copy-trading feature. To copy a trader you do four things yourself: build a shortlist of candidate wallets, vet them, point a third-party execution bot at the one you choose, and control your size and exit rules. Most guides cover step three and skip the rest — which is exactly backwards, because in our June 2026 snapshot of 1,297 actively-traded Polymarket wallets, only 28 (2.2%) passed every test a copier should apply. The bot is the easy part. Choosing who to copy is the whole game.
This guide walks the full loop, in order.
What is copy trading on Polymarket?
Copy trading means automatically mirroring another wallet's trades: when the wallet you follow buys YES in a market, a bot buys YES for you, scaled to your capital. Because every Polymarket trade settles on-chain (on Polygon) and is queryable through public APIs, any wallet's full trade history is public — which is what makes both copying and vetting possible.
Polymarket itself doesn't execute copies. Third-party bots do: they watch a target wallet's fills and replay them from your account, usually for a fee. CopyGrade sits on the other side of that loop — it's the analysis layer that grades wallets before you commit capital, and it never executes anything.
Step 1: build a shortlist — and don't use the leaderboard as one
The obvious place to start is Polymarket's profit leaderboard. Resist it. When we scored the 193 wallets on the profit and volume boards, 72% carried a farming flag and only 1.6% were worth copying — a worse hit rate than the wider active population. Raw profit rewards size, luck, and the exact volume-manufacturing behaviours a farmer uses to attract copiers.
A better shortlist filters on the things profit hides: post-fee edge, risk-adjusted return, drawdown depth, and a clean forensic record. That's what Wallet Scout ranks by, and our graded leaderboard shows every board name next to its verdict so you can see the rank-versus-reality gap directly.
Step 2: vet the wallet before anything touches your money
This is the step the official guidance and every bot vendor skip, and it's where the 2.2% separates from the rest. The full quantitative checklist is its own guide — how to vet a Polymarket trader — but the five questions are:
- Is the edge real, or variance? Enough resolved trades, across enough independent markets.
- What's the return per unit of risk? Two wallets with the same return are not equally copyable.
- How deep are the drawdowns? You will copy through one.
- Where does the edge live? A sports-sharp wallet may be a politics donor.
- Is it farming its copiers? The behavioural signatures — iceberg accumulation, self-trade wash, decoy clusters — are visible on-chain if you look.
Here's how the active population fares against those tests, applied cumulatively:
| Test | Wallets passing | Share |
|---|---|---|
| Scored | 1,297 | 100% |
| Enough recent history (≥20 trades / 90d) | 995 | 77% |
| Farming-clean | 606 | 47% |
| Realistic post-fee edge above 1% | 214 | 16% |
| Worth copying (Copy Score ≥ 75) | 28 | 2.2% |
The full distribution behind this funnel is in our scored-population report.
Step 3: choose an execution bot
Only after a wallet survives vetting does the bot question matter. Bots differ on the dimensions that decide whether a thin edge survives execution: latency, fee structure, sizing controls, and what happens when the leader does something you don't want to mirror. The selection criteria — and the fee math that quietly eats most copied edges — are covered in how to choose a Polymarket copy-trading bot.
The one-line preview: in our June 2026 snapshot only 16% of active wallets had a realistic post-fee edge above 1% — so a bot charging 1% per side can, on its own, consume more edge than the median wallet has.
Step 4: size it like the drawdown is coming
Never size off the headline return — a copier typically keeps 20–40% less than the leader's posted number after fees, slippage, and latency. Size off the wallet's worst historical drawdown instead, and assume you'll live through one like it early. The arithmetic of recovering from losses, per-category budgets, and stop rules are in position sizing for copy trading.
Step 5: monitor, and know what makes you stop
A wallet that was worth copying in June isn't guaranteed to be worth copying in September. Edges decay, behaviour changes, and a clean record is — sometimes by design — exactly what precedes a farming exit. Decide before you start what makes you stop: a score drop, a new farming finding, a drawdown beyond the historical worst. Watchlist & Alerts automates those triggers; the discipline of having them is what matters.
What should you actually expect to make?
Honestly: less than any headline suggests, and nothing at all if the vetting step is skipped. The median active Polymarket wallet has a negative fee-adjusted edge before a copier adds slippage or latency. Copying is not passive income; it's a selection problem where the base rate of good choices is about one in fifty — and the Copy Score exists to compress that selection work into a single, vetoable 0–100 verdict. One honest caveat: the score is a documented opinion built from trade history, and we recalibrate it in public as outcome data accumulates.
CopyGrade is independent and analysis-only — it never executes trades, holds funds, or takes referral fees from any bot. Not financial advice.