Copy trading sells a simple idea: find someone who trades well, connect your account, and every trade they place is repeated in yours, scaled to the money you allocate. On eToro, the largest platform built around it, the minimum to copy one trader is $200 and there is no separate fee for copying. MetaTrader's signal service, ZuluTrade and most crypto exchanges sell the same idea in slightly different packaging.
The number that sells a trader to you is usually the one on top of the profile: the win rate, sitting next to a return chart that climbs steadily. This guide is about why that number tells you the least, and what to check instead before your money follows someone else's.
Who this is for: people choosing whom to copy, or already copying, on any platform. It is not a review of eToro or any other service, it does not cover running a signal service yourself, and the strategies below are modelled, not real traders — so you can see the mechanics without the noise of one person's luck.
The trapA 97% win rate that loses money
Take two hypothetical traders. Each places about 20 trades a month, and each result is a percentage of the account.
- "Smooth" wins 97% of trades, for +0.8% each. Three times in a hundred, a position goes wrong and it is held until it costs −30% — the shape of an averaging-down, grid or martingale strategy, where losers get more size instead of a stop.
- "Bumpy" wins only 42% of trades. Its winners make +1.5% and its losers are cut at −1%.
On a leaderboard, Smooth looks like the professional and Bumpy looks like a coin flip. The arithmetic says the opposite. Expectancy — the average result per trade — is 0.97 × 0.8% − 0.03 × 30% = −0.12% per trade for Smooth and 0.42 × 1.5% − 0.58 × 1% = +0.05% for Bumpy. One loses money slowly with very high confidence; the other makes money slowly with constant small losses. (Our expectancy guide explains the formula in detail.)
3 months
- Smooth: no losing trade yet
- 16.0% of accounts
- Smooth: median result
- −22.2%
- Bumpy: median result
- +2.1%
6 months
- Smooth: no losing trade yet
- 2.4%
- Smooth: median result
- −12.9%
- Bumpy: median result
- +4.2%
12 months
- Smooth: no losing trade yet
- 0.1%
- Smooth: median result
- −47.3%
- Bumpy: median result
- +11.3%
24 months
- Smooth: no losing trade yet
- 0.0%
- Smooth: median result
- −72.2%
- Bumpy: median result
- +23.8%
| After | Smooth: no losing trade yet | Smooth: median result | Bumpy: median result |
|---|---|---|---|
| 3 months | 16.0% of accounts | −22.2% | +2.1% |
| 6 months | 2.4% | −12.9% | +4.2% |
| 12 months | 0.1% | −47.3% | +11.3% |
| 24 months | 0.0% | −72.2% | +23.8% |
Source: CalculatorAI · calculatorai.app · CalculatorAI Monte Carlo model; see 'Where these numbers come from'
The table hides the part that matters most for copiers. After three months, 16% of Smooth accounts have never lost a single trade, and every one of them is up 61%. Sixty clean wins at +0.8% compound to that. Put a thousand traders like Smooth on a platform and about 160 of them will sit at the top of the leaderboard with a perfect record and a straight-line chart — which is exactly when new copiers arrive.
SurvivorshipWhy the leaderboard shows you the lucky ones
A leaderboard ranks results that already happened. For a strategy with rare, large losses, a short track record mostly measures whether the loss has arrived yet. The probability that Smooth gets through 33 trades without a loss is 37%; through 60 trades, 16%; through a year, almost zero.
One loss, two months of gains gone
You start copying Smooth after a clean streak of 33 wins, which took the account up 30.1%. Your own copy starts at zero. If the next trade is the −30% one, your copy is down 30%, while the leader is down only 8.9% from where they started, because they banked the streak before you arrived. To get back to even, your copy needs +42.9% — about 45 more wins in a row at +0.8%.
That last detail is why copiers often lose on traders who look profitable over their lifetime: the returns on the profile were earned before most of the copying money arrived. Money flows in after good runs and leaves after drawdowns, so the average copier gets a worse result than the trader they copied.
The checksFive numbers that matter more than the win rate
Every serious copy platform shows more than the win rate, usually one click down. These are the figures to open first.
- 01Average loss vs average win. If the average loss is many times the average win, a high win rate is buying you a rare large loss. Smooth's ratio is 37.5 to 1 — a red flag no win rate can fix.
- 02Largest single loss. A trader who has never lost more than 2% on a trade is a different animal from one whose worst trade cost 25%, even with identical returns.
- 03Maximum drawdown — and how long the record is. A drawdown figure is only meaningful against a long enough history. Six months of a strategy with rare losses can show a drawdown near zero. Our guide to what drawdown is normal gives ranges by strategy type.
- 04Profit factor. Gross profit divided by gross loss. Above 1 means the record made money; what counts as a good profit factor depends on trade frequency, and very high values on short records are usually just a missing loss.
- 05How long losing trades are held. If winners close in hours and losers stay open for weeks, the trader is avoiding stops. Look at open positions, not only closed ones: a big floating loss does not show in the win rate at all.
ExecutionWhat copying itself costs you
Even if you pick a genuinely good trader, your results will not match theirs exactly. Two mechanical effects are worth putting numbers on.
Delay and spread. Your copy is placed after the leader's trade, at whatever price is available then, and you pay the spread on every trade. For a long-term investor that difference is negligible. For a scalper it can be the whole edge. Take a scalper who wins 60% of trades at +0.30% and loses 40% at −0.35%: an expectancy of +0.04% per trade, or about 0.8% a month at 20 trades.
0.00% (perfect copy)
- Your expectancy
- +0.04%
- Your month
- +0.80%
0.02%
- Your expectancy
- +0.02%
- Your month
- +0.40%
0.04%
- Your expectancy
- 0.00%
- Your month
- 0.00%
0.06%
- Your expectancy
- −0.02%
- Your month
- −0.40%
| Extra cost per trade | Your expectancy | Your month |
|---|---|---|
| 0.00% (perfect copy) | +0.04% | +0.80% |
| 0.02% | +0.02% | +0.40% |
| 0.04% | 0.00% | 0.00% |
| 0.06% | −0.02% | −0.40% |
Source: CalculatorAI · calculatorai.app · CalculatorAI arithmetic; 20 trades a month
A few hundredths of a percent — two or three ticks on a fast market — turns a winning strategy into a losing copy. That is why slower, longer-holding traders copy better than fast ones.
Minimum position size. eToro states a minimum of $1 for each copied position. If the leader opens a position worth 0.25% of their account and you copy with $200, your share would be $0.50 — below the minimum, so that trade is not opened in your copy. With $500 it would be $1.25 and is opened. A small copy can quietly skip the leader's smaller trades, and your result drifts from theirs in a way neither of you sees on the profile.
The other sideWhat the trader you copy is paid for
Copy trading works for some people, and it helps to know how the incentives run. Leaders are typically paid by the platform based on assets copying them, or take a share of profits on services that charge one. Either way, a long, smooth record attracts more money than a volatile honest one — which rewards exactly the strategies that hide their losses. Regulators have noticed: in 2023 the European Securities and Markets Authority told firms that copy trading is an investment service under MiFID II, that traders being copied must have adequate knowledge and experience, and that costs and conflicts must be disclosed clearly.
The disclosure you will see most is about CFDs. eToro's own page states that 51% of retail investor accounts lose money when trading CFDs with it. That figure covers all CFD trading, not copying specifically — but copying a leveraged CFD strategy puts you inside it.
The methodTrack your copy like your own trading
The fix is to stop judging a copied trader by their profile and start judging them by your account. Treat each trader you copy as one strategy in a journal:
Log every copied trade under the leader's name
One strategy tag per trader you copy, so their results never blend into each other or into your own trades.
Record your fill, not theirs
Your entry and exit prices are the ones that decide your result. The difference from the leader's prices is your copying cost.
Watch average loss and largest loss weekly
If one loss is larger than several weeks of wins, you are copying the Smooth pattern, whatever the win rate says.
Set your exit in advance
Decide the drawdown at which you stop copying — and set the platform's copy stop-loss to it — before the first trade, not during a losing week.
Give it enough trades
Judge after dozens of your own copied trades, not after a good first week.
The Trading Journal handles this without extra setup: log each leader's trades under their own strategy name, filter the dashboard to that strategy, and the win rate, average win and loss, largest loss, profit factor, expectancy and maximum drawdown you see are for that leader alone. Before you allocate, the Risk/Reward Calculator shows what a leader's average win and loss imply at their win rate — and the risk of ruin guide shows how quickly a deep single loss compounds.
Where these numbers come from
- The two traders are models, not real accounts. Smooth: 97% wins of +0.8%, 3% losses of −30%. Bumpy: 42% wins of +1.5%, 58% losses of −1%. Twenty trades a month, results as a percentage of the current balance and compounded, 20,000 simulated accounts each with a fixed random seed. Script:
drafts/copy-trading-numbers.mjs. Real strategies vary their trade sizes and do not have fixed outcomes; Smooth's single −30% figure stands in for the blow-up of an averaging-down position, which in practice can be smaller or much larger. - Bias of the model: it ignores fees, spreads and slippage, which would make both traders look worse; and it treats each trade as independent, while real losses tend to cluster in volatile markets, which would make Smooth's tail worse still.
- Copying cost table: a scalper with 60% wins of +0.30% and 40% losses of −0.35% (+0.04% expectancy), minus a flat extra cost per trade.
- Platform facts: eToro's CopyTrader page — $200 minimum to copy, $1 minimum per copied position, no additional charge for copying (spreads and transaction fees still apply), "51% of retail investor accounts lose money when trading CFDs with this provider" — read October 6, 2026. Platform rules change; check the current page before you allocate. ESMA's supervisory briefing on copy trading services, 2023.
FAQFrequently asked questions
Is copy trading profitable?
It can be, but your result depends on the trader's real expectancy, on what copying costs you in delays and spreads, and on when you start. Many copiers lose money on traders whose lifetime record is positive because they join after a strong run.
What is a good win rate for a trader to copy?
There is no good win rate on its own. A 40% win rate with winners larger than losers can be excellent; a 97% win rate with rare large losses can be a losing strategy. Compare the average loss with the average win, and look at the largest single loss.
How do I spot a martingale or grid strategy?
Very high win rates, a smooth equity curve, winners closed quickly, losers held for days or weeks, and large open floating losses. Position sizes growing after losses is the clearest sign.
How much money do I need to copy a trader on eToro?
eToro's minimum is $200 per copied trader, with a $1 minimum per copied position. Small copies may skip some of the leader's smaller trades.
Should I stop copying after a drawdown?
Decide in advance. Set a copy stop-loss or a personal drawdown limit before you start, based on the trader's historical maximum drawdown, and follow it — changing the rule during a loss is how copiers end up exiting at the bottom.
Does copy trading count as my own trading for taxes?
In most countries, yes: copied trades are executed in your account, so their gains and losses are yours to report. Keep a full record of each trade, not just the platform's summary.
Copy the record, not the reputation
The profile tells you what a trader earned for the people who were there first. Your journal tells you what you are earning now. Check the average and largest loss before you copy, give each leader a strategy tag in the Trading Journal, and judge them by your own fills — that number cannot hide a loss that has not happened yet.
