A copy trading setup has three parts: a master account that trades, a copier engine that watches it, and follower accounts that receive the orders. The engine subscribes to the master's trade events, and when a position opens, closes or has its stop moved, it translates that event into an order for each follower and sends it. On a single server the whole cycle can complete in well under a second. Across brokers and continents it takes longer, and the difference lands in the follower's fill price.
That is the entire product. Everything sold as social trading, mirror trading or signal following runs some version of it. The variables worth understanding are how size is calculated, how fast the copy happens, and what the leaderboard is hiding.
How size gets decided
The master trades 1 lot. What should a follower with a 3,000 unit account trade? Copier engines answer that in one of three ways, and the choice is the largest single source of unpleasant surprises.
| Allocation mode | How the size is set | Where it fails |
|---|---|---|
| Fixed lot | Every follower trades a size the follower chose, ignoring the master's size | The master's own risk sizing is discarded, so a small stop and a wide stop get identical exposure |
| Proportional by balance | Follower size scales by the ratio of the two account balances | Small accounts round down to the minimum lot and end up over-risked, or round to zero and skip trades |
| Risk multiplier | Follower size scales by balance ratio times a chosen factor | Users set the factor above 1 after a good month, which is exactly when it hurts |
The rounding problem is real and underrated. If the minimum tradable size is 0.01 lots and proportional sizing calls for 0.004, the engine either skips the trade or rounds up to 0.01, which can be two or three times the risk the master took. Followers with small accounts often run far hotter than the strategy they are copying, then blame the strategy.
The gap between the master fill and yours
Every copy arrives late by construction. The engine cannot send the follower order until the master order is confirmed. Add the network hop, the size calculation and the follower broker's own processing, and the copy enters the market after the price that triggered it has already moved. On positions held for hours, a fraction of a pip of difference is noise. On a scalping strategy that targets six pips, it can consume most of the edge.
Different brokers make it worse. If the master runs on a raw spread account and the follower on a wider retail spread, the follower pays more on every entry and exit before latency is even considered. This is why a strategy with a published record can look mediocre in a follower's account without anybody misbehaving. The mechanics of the transport layer are covered properly in the guide to trade copiers.
Copying does not transfer risk. The follower owns every position, pays every loss and is responsible for the margin. The only thing that moved is who decides when to click.
What a leaderboard does not show
Ranked lists of providers are marketing surfaces. They are usually sorted by return over a recent window, which quietly selects for whoever took the most risk in that window and survived. Providers who blew up disappear from the list, so the average visible record is better than the average real one.
Three things are worth more than the headline percentage. The worst peak to trough decline, expressed as a percentage of equity. The average holding time, which tells you whether latency will hurt. And the shape of the position sizing across a losing streak: sizes that grow while the account is down are the signature of martingale, and a martingale master will produce a flawless leaderboard entry until the day it does not. The checklist in evaluating signal services covers the rest of the due diligence.
Verified track records help, but only if the verification covers the account that is actually being copied. A screenshot of one account and a live link to another is a common trick, and reading the account number on the record is enough to catch it.
Fees change the arithmetic
Copy trading is usually paid for in one of three shapes: a flat subscription, a performance fee on profits, or a rebate the provider earns from the broker on the volume followers generate. The third is the one to inspect. When a provider is paid per lot traded, trading more is profitable for them regardless of the outcome for followers, and there is no rule of nature that stops a strategy from drifting towards higher frequency once the rebate arrives. None of this is hidden if you ask how the provider is paid, and a provider who will not answer that question has answered it.
The firm side of copy trading
For a broker, copy trading is a retention feature and a risk concentration event at the same time. When one master triggers four hundred identical orders, the firm's book takes four hundred correlated positions in the same second. Risk desks handle that by treating copied flow as a single exposure rather than many small ones, and by capping how much aggregate volume any one master can generate. The build considerations sit in the guide to copy trading and signal providers.
Prop firms have the opposite problem. A trader copying their own strategy across their own accounts is usually fine within limits. Dozens of accounts held by different people mirroring one master is a different situation, because the firm ends up paying out on one strategy many times over while carrying the identical drawdown many times over. Detection is a correlation problem: same symbol, same direction, entries clustered inside a few seconds, repeated across accounts that have no other connection. Firms that run copy trading abuse detection catch the cluster early, which is fairer than catching it at the payout window.
"A leaderboard shows you the survivors. Ask for the worst month and the account number, and half the conversation is finished."
— Alex Onta, Executive Director, SINGUARD
Key Takeaways
- Allocation mode decides your real risk: fixed lot throws away the master's sizing, and proportional sizing rounds small accounts up to sizes the master never intended.
- Copies always arrive after the master fill, so short holding times and wide follower spreads eat the edge before the strategy gets a chance.
- Judge a provider on worst drawdown, holding time and size behaviour during losses, not on the ranked return that put them on the list.
- For firms, copied flow is one correlated exposure repeated many times, and it needs an aggregate cap rather than per-account limits.
Frequently Asked Questions
Is copy trading passive income?
No. Copying moves the decision to somebody else, but the money and the loss stay with the follower. Leveraged trading carries a high risk of loss whether the orders are placed by hand or copied automatically, and a follower who never checks the account still owns every position in it.
Why does my copied trade fill at a different price from the master?
The copier has to see the master fill, calculate a size for the follower and send a new order, which takes time. Followers may also sit on different servers or different brokers with different spreads. Fast entries suffer most, which is why scalping strategies copy poorly while positions held for hours barely notice the delay.
Do prop firms allow copy trading?
Policies vary. Most firms permit a trader to copy their own strategy across their own accounts within stated limits, and most prohibit many accounts owned by different people mirroring one master, because that concentrates identical risk on the firm. Firms run correlation checks on entry times, symbols and direction to find it.