Prop-Firm Copier Sizing: The Arithmetic When Funded Accounts Differ

A copier is configured to mirror the master's lot size across four funded accounts. The master takes 0.60 lots on gold. The 200,000 dollar account risks 0.3 per cent, the 100,000 account risks 0.6 per cent, the 50,000 account risks 1.2 per cent, and the 25,000 account risks 2.4 per cent on the identical trade. Four accounts, one decision, four different pieces of risk — and the smallest account, which is usually the newest and the one you can least afford to lose, is carrying eight times the exposure of the largest.

The architecture of running one strategy across several prop-firm accounts is covered elsewhere. This post is only about the arithmetic: how to size so that every account experiences the same thing, where that arithmetic breaks, and how the drawdown maths changes when the accounts are not the same.

The rule: each account sizes itself

The only sizing mode that survives mixed account balances computes the lot size independently on each receiving account, from that account's own equity and the trade's stop distance. The master's volume is not copied at all — only the decision is.

For a EURUSD trade with a 34-pip stop, at 0.5 per cent risk, with a pip value of 10 dollars per standard lot:

account      equity     risk 0.5%    lots = risk / (34 * 10)   rounded
A           200,000      1,000.00          2.941                2.94
B           100,000        500.00          1.470                1.47
C            50,000        250.00          0.735                0.73
D            25,000        125.00          0.367                0.36

Every account now takes an identical percentage loss if the stop is hit and an identical percentage gain at the target. The accounts move together in percentage terms regardless of their sizes, which is exactly the property you want: one set of statistics describes all four.

The alternative, balance-proportional sizing, scales the master's lot by the ratio of receiver equity to master equity. It produces the same answer as long as the master trades a constant percentage risk. It silently breaks the first time you vary risk on the master, or the first time the master's balance drifts from its reference, and the breakage is invisible until you audit it. Use percentage sizing and the question never arises.

Where the arithmetic breaks: the lot step

Rounding is not a rounding error at small account sizes. Every broker enforces a minimum volume and a volume step — commonly 0.01 on forex, sometimes 0.1 on index CFDs, and occasionally 1.0 on some instruments.

Take the same trade at 0.5 per cent risk on a tight 8-pip stop:

account      equity     risk 0.5%    ideal lots    rounded (0.01 step)   actual risk
A           200,000      1,000.00       1.250          1.25                0.500%
D            25,000        125.00       0.156          0.15                0.480%

Tolerable. Now the same accounts on an index CFD with a 0.1 lot step:

account      equity     risk 0.5%    ideal lots    rounded (0.1 step)    actual risk
A           200,000      1,000.00       2.500          2.5                 0.500%
D            25,000        125.00       0.312          0.3                 0.480%
                                                       0.4                 0.640%

Rounding down costs the small account 4 per cent of its intended risk, which is fine. Rounding up, which some copiers do by default so that the trade is not skipped, costs it 28 per cent more risk than intended. Across a losing run that difference compounds into a materially different drawdown on one account than on the others, and the statistics you thought described all four now describe none of them.

Three rules follow:

• Always round down, never up. A slightly smaller position is a rounding artefact. A slightly larger one is an unrequested risk increase.
• Skip rather than floor. If the computed size rounds to below the broker's minimum, the correct behaviour is to skip the trade on that account and log it, not to place the minimum. Placing the minimum on a very tight stop can be several times your intended risk.
• Track the skips. An account that silently skips 15 per cent of trades is running a different strategy from the others, and its results are not comparable.

Drawdown maths when the structures differ

Normalised percentage risk keeps the accounts moving together. It does not make their limits the same, because the limits are set by the firms, not by you.

Consider three funded accounts running the same strategy at 0.5 per cent per trade:

A losing run of six trades takes every account down 3 per cent. Account A and Account C are comfortable. Account B is three quarters of the way to its daily limit — and if those six trades straddle 17:00 New York, they fall across two of B's days and none of A's.

Two principles come out of this.

The weakest constraint governs the group. If you want one set of settings across all accounts, the per-trade risk has to be sized for the account with the tightest limits, not the average. Sizing for the average means the tightest account breaches first and repeatedly, and you will mistake a configuration problem for a strategy problem.

Correlated drawdown is the structural risk of the whole arrangement. Copying does not diversify anything. The same strategy on five accounts is one strategy with five times the exposure. Every account hits its worst day on the same day, by construction. Size as though that is true, because it is.

Account B's trailing maximum deserves separate attention. After a good month it sits at, say, 108,000 with a floor at 99,360. Accounts A and C, on static floors, still have their original floors at 90,000 and 45,000. The identical subsequent drawdown that costs A and C a fifth of their room can end B entirely. Running a trailing-drawdown account alongside static ones in the same copy group means one member of the group is always the fragile one.

Per-account guards, not a portfolio guard

Each firm measures its own account against its own floor on its own clock. Nothing you do aggregates across firms, and nothing a firm does considers your other accounts.

That means the guard configuration is per receiver, not global:

1. 01A daily equity halt on each account, set to a fraction of that firm's own limit, using that firm's own reset hour.
2. 02A maximum-lot clamp per account, sized to that account's balance.
3. 03A maximum-notional check per account, so that a tight stop cannot produce an oversized position.
4. 04A halt that stops copying to one account without stopping the others. This is the feature that matters most and the one most often missing. When Account B approaches its limit, B should stop receiving and A and C should continue untouched.
5. 05A record of which accounts were halted when, so the equity curves can be compared honestly afterwards.

What divergence is normal and what is not

Even with correct sizing, the accounts will not produce identical returns. Some divergence is expected and some is a signal.

Expected: small differences from fill prices at different brokers, from rounding to the lot step, and from spread differences on the same symbol at different venues.

Worth investigating: one account consistently filling worse than the others in the same direction, which usually means a feed or spread problem at that venue. One account skipping trades others take, which means a symbol mapping gap or a minimum-volume floor. One account's percentage result drifting steadily from the group, which means a sizing input is wrong somewhere.

Measure it deliberately: percentage return per account per week, side by side, with the skip count next to it. A table you look at every Friday will find a misconfiguration in a week. An equity curve you glance at monthly will find it in a quarter.

Adding an account without breaking the group

1. 01Run the new account as a receiver on demo for a week first, with the real settings.
2. 02Verify symbol mapping with minimum-size trades on every instrument you actually trade at that broker.
3. 03Record the broker's lot step and minimum volume for each symbol, and check what the copier does when the computed size falls below the minimum.
4. 04Enter the firm's daily limit, maximum drawdown structure and reset hour into the guard configuration before the first live trade.
5. 05Compare the new account's percentage returns against an existing account daily for the first week, and explain any gap above a few tenths of a percentage point before adding capital.

The week is not wasted. Every configuration mistake you are going to make on that account shows up in it.

Where this fits

For the architecture around this arithmetic — master and receiver setup, the scaling rationale, symbol mapping, and what the terms of service allow — read scale-strategy-multiple-prop-accounts, which this post deliberately does not repeat. If the per-account drawdown structures are the part you need to pin down, what-is-daily-drawdown covers how the floors are calculated.