01In plain English
Copy trading is the idea that if someone else is consistently good, you can ride along by doing what they do, automatically. On Polymarket, every trade a wallet makes is recorded on the Polygon blockchain in public. A copy-trading bot watches one or more target wallets, and the instant a target buys YES on some market, the bot places the same trade in your account - scaled to your own size limits.
The appeal is obvious: you don't have to find the edge yourself, you rent someone else's. The catch is equally obvious once you say it out loud - you are always a step behind the wallet you copy, and you can only see the trades that already happened, not the reasoning or the exits they may make faster than you.
02How it works on a prediction market
Because Polymarket settles through the CLOB API and the Conditional Token Framework on Polygon, a wallet's fills are visible almost in real time through the order stream and on-chain logs. A copy-trading bot turns that visibility into a pipeline:
- Watch. Subscribe to the target wallet's fills via the WebSocket feed or an on-chain indexer.
- Translate. Map the target's trade (market, side, price, size) onto your own sizing rules - a fixed fraction of their size, a flat per-trade cap, or a risk-scaled amount.
- Fire. Place your order through your own proxy wallet, respecting slippage and rate-limit budgets.
- Exit. Mirror the target's sells too - or run your own exit logic if you trust your entries more than their timing.
03Why sizing decides the outcome
The single biggest mistake in copy trading is sizing blindly. A $2M wallet taking a $40,000 position is risking 2% of its book; if you copy that notional onto a $20,000 bankroll, you've just bet 200% of yours. The bot has to translate, not photocopy.
04Constraints that limit the strategy
- Survivorship bias. The wallet that turned $5k into $2M is visible precisely because it won. For every one of those, dozens blew up and left no leaderboard trace. Copying the winner after the fact is not the same edge they had.
- Latency. You see the trade after it fills. On thin books, the target may have already moved the price you wanted - you copy into a worse fill, or none at all.
- Exit asymmetry. Entries are easy to mirror; exits are where alpha lives. A target who scratches a losing position in seconds can leave a slow copier holding it.
- Style drift. A wallet's edge is tied to a regime - an election cycle, a crypto run. When the regime ends, last quarter's hero becomes this quarter's drag.