Why most copy traders lose money even when the trader wins

DarkTrade Research

TL;DR — A profitable trader and a losing follower are not a contradiction — they are two different trades. The one configuration we have measured end to end (stop at the risk line, exit at the whale's exit) came out negative after exchange fees while the whales themselves came out positive. That is a finding about that exit rule, not about following. The exit is the part we deliberately leave to you: we publish what the trader did and where the risk line sits, and you decide when to close.

There is a version of this question that sounds like an accusation and a version that is just arithmetic. This is the arithmetic one.

The configuration we measured end to end was a specific one: stop at the risk line, exit at the whale's exit. The follower opens when the trader opens, places a stop where a stop-out costs exactly the percentage of the account they chose to risk, and closes when the trader closes. Measured that way, across our own pool, every follower configuration we have tested comes out negative after exchange fees while the whales themselves come out positive. We publish that because it is what we found, and because a research note that only reports the flattering half is not research.

Read it carefully, though. It is a result about one exit rule, not a verdict on following a good trader. The rest of this page is why those are different sentences.

The trader and the follower are not in the same trade

The trade a whale is in and the trade a follower is in share a ticker and a direction and almost nothing else.

They do not enter at the same price. By the time a position is visible on-chain, confirmed, and delivered, the price has moved. On a position the whale holds for days that gap is noise. On one it holds for minutes it can be most of the move. This is exactly why our own entry rule requires a minimum hold: a whale whose median position closes faster than the delivery takes is not a whale a follower can be in the same trade as.

They do not hold the same size relative to their account. A whale putting a small slice of its capital into a position and a follower putting a large slice of theirs into it are running different risks, and will make different decisions under pressure — which is a polite way of saying the follower will close it first.

They do not pay the same fees, proportionally. The whale's costs are a rounding error against its position. The follower's are not. Fees are charged on both sides of every trade, and a strategy with a small average edge per trade and a high trade count is a strategy where fees decide the sign.

The exit is where the money actually is

Entry is the part everyone copies. It is also the part that matters least.

Take a position that goes up, keeps going up, then gives most of it back before the trader closes. Both parties saw the same chart. The peak is a real number that the whale did not capture either — it is the most a position was ever up, not money anybody received, which is why we label it as a peak everywhere it appears and never add it to anything.

Now change only the exit rule and nothing else:

• Exit at the whale's exit. You get the whale's result, minus your entry gap, minus your fees. If the whale gives back a run before closing, so do you — and you paid for the privilege on both sides.
• Stop at the risk line and otherwise hold. Your losses are bounded at a number you chose, and your winners run as long as the whale's conviction does.
• Take something off at a fraction of peak. You convert part of an unrealized number into a realized one, and you give up the tail.

Same entries, same trader, three different outcomes. Nothing about the trader's skill changed between those three lines. That is what "the exit rule is the lever" means, and it is why "this configuration lost" is not the same statement as "following this trader loses".

What we do about it

We are intelligence, not execution. We never place an order and we never hold funds — not as a policy we could change, but because there is no mechanism here to do either. So the exit is yours. What we can do is make sure the inputs to that decision are real:

• The risk line is computed, not guessed. Position size is derived from the stop rather than alongside it, so a stop-out costs exactly the percentage you chose. Not "about". Exactly.
• The trader's own worst case sets that line. The stop sits outside the worst loss the whale has genuinely sat through, plus a point — so it is not placed inside the ordinary noise of how that trader already trades.
• A position can't need a move bigger than the whole entry price. Beyond that, a "stop" is not a stop, it is total loss with extra steps, and the model says so instead of quoting a number that cannot be placed.
• A package holds as many traders as fit its signal budget, not as many as qualify. Concurrent risk adds up — every open position is spending the same account — so how many names you follow is a risk decision before it is a selection decision, and we size it deliberately rather than filling the card.
• We publish the losing measurements. No follower-basis performance number appears anywhere on this site, because the ones we have measured are negative and putting them on a page would be publishing something we cannot stand behind. When one measures positive, it will go up with its method attached.

So should you follow a trader at all?

The honest answer is that the question is badly formed. "Following" is not one strategy; it is an entry rule plus an exit rule plus a sizing rule, and the last two are doing most of the work. A profitable trader is a source of entries. What you do after the entry is a separate set of decisions, and it is the set that decides whether you are profitable.

What we would say is: judge the trader on their drawdown before their return, insist on knowing the denominator under every rate you are shown, and treat any exit rule — ours, theirs, yours — as something to be measured rather than assumed. We have measured one of them. It lost. We said so.

---

Every figure above is a whale's own realized result on its own account, over closed trades only. Peak is the most a trade was up before it closed — nobody captures it, and it is never added to anything. A colour is a measured drawdown band, not a promise. How we measure.