What Is Copy Score & How to Use It
Copy Score tells you whether a Polymarket trader is worth copying after the cost of following them. What the number means, what a dash means, and how to act on it.
Last updated August 21, 2026
Great traders are not always great to copy. Copy Score is the one number that tells you which ones are: it takes a trader's finished trades, subtracts what it costs to follow them, and reports what was left. It is free on every trader, everywhere. Read it as a reason to rule someone out rather than as a ranking to buy off the top.
Understand what Copy Score measures
Copy Score is about the trader, not the trade in front of you. It looks at their finished trades, subtracts what it costs to copy them (you fill worse than they did, Polymarket charges a fee on taker orders, and a Trader Bot includes our 0.5%), and asks whether anything is left. Win rate and P&L tell you how they did. Copy Score tells you whether following them has been worth the toll. It is based on their finished trades to date, not a forecast, and it assumes every one of their trades was copied exactly — your filters, budget and timing will all change it. Copying anyone costs something. What changes is whether their edge clears it.
Read the number
The score is a leftover, not a rating. On a feed card or beside a trader's name it renders as a signed whole number with no unit — +6 means what was left after copy costs was around six percent, −4 means it was negative. On the Discover board the same score is shown as a word and a magnitude together, "Clears 16%" or "Losing 6%", because a column of bare numbers taught nobody what good looked like. It is not a win probability and not a 0 to 100 rating. We round on purpose: tenths would pretend at a precision we do not have.
Read a dash
A dash is the absence of a score, not a middling one, and there are two cases. "Not comparable" means they close out early by selling rather than holding to the result, so we can warn about them but will not rank them alongside hold-to-the-end traders. "Not scored" means we do not have a score for that wallet yet. That is not a verdict. A third case is not a dash: a trader with very little finished history still gets a real score, with a "Thin history" badge beside it, because a big return on a handful of trades is not yet a track record.
Treat it as a veto, not a ranking
Start with traders you would consider on P&L, volume, and category anyway, then use Copy Score to rule people out. "Caution" and "Losing" are the useful no, and they mean different things: Caution is a trader who makes money whose edge does not survive being copied, Losing is a trader down on their own book before copying costs start. Do not sort strangers by the number and copy the top row. "Ahead" is a positive read, but only just. Treat it as thin rather than as a recommendation.
Prefer a category they have a record in
A trader can clear copy costs overall and still be ordinary in the market on this card. Prefer a category they already trade. Some traders also carry a red flag for being hard to mirror, which is about how they trade rather than what they earned.
Know what it costs
Nothing. Copy Score is free on every trader, on Discover, on trader profiles, and on feed cards, with or without Premium. There is nothing to unlock and no score filter to buy. Any screen telling you to upgrade to see a Copy Score is a bug — tell us at help@polycopy.app.
| State | What it means |
|---|---|
Clears | Copying them has worked on their finished trades. |
Ahead | They finish ahead of what copying costs — but only just. |
Caution | They make money. It doesn't survive being copied. |
Losing | They've been losing on their own trades, before copying costs even start. |
Not comparable | They close out early, so we can warn about them but never rank them alongside hold-to-the-end traders. |
Not scored | We do not have a score for this wallet yet. That is not a middling score; it is the absence of one. |
What it cannot tell you
It is not a forecast, and it says nothing about the market on the card in front of you. It also assumes every one of a trader's trades was copied exactly — your filters, your budget and the seconds between their fill and yours will all move the real result away from it. And the absence of a score is never reassurance: a trader with no verdict has not been checked and passed, they have not been checked.