Ten accounts can still be one bet

Two things we could not find written down anywhere, after going looking for how serious traders actually construct a multi-firm book. Plus four problems we have not solved.

We went looking for how serious traders construct a book across several prop firms and found almost nothing. The published material is firm marketing, copier and VPS vendors, or affiliate comparison pages. There is an execution-operations literature. There is no portfolio-construction literature.

These two points came out of that search. Take them and use them, whoever you end up subscribing to.

1. Ten accounts can still be one bet

Portfolio theory has a precise statement for this. Grinold and Kahn's fundamental law says your information ratio scales with the square root of breadth — where breadth counts independent decisions. Not accounts. Not contracts. Not firms.

Copy one signal to ten accounts and your breadth is one. You have multiplied the evaluation fees by ten, multiplied the monthly costs by ten, and arranged for every account to fail on exactly the same day. What you have not done is diversify.

Ask yourself: if my worst trade of the year happened tomorrow, how many of my accounts would still be alive?

If the answer is “none”, you own one account with extra paperwork.

This is not a beginner's mistake. The same blind spot is documented at institutional level: the CFA Institute found that much of the apparent difference between managed-futures managers is the same trend bet running at different speeds.

2. The drawdown type should pick the strategy

Trailing and static drawdown are not two flavours of the same rule.

A static limit sits at your starting balance and only moves when you lose. A trailing limit follows your highest-ever balance — so it tightens when you make money and give some back, and it can end an account on a trade that finished profitable.

Which means compatibility with a drawdown regime is a property of the strategy, not of the firm. A mean-reversion engine that routinely gives back open profit is structurally mismatched to intraday trailing. A strategy that takes money off the table quickly barely notices it. Same firm, same rules, opposite outcomes — decided by something you can measure in advance.

Ask yourself: am I choosing firms by profit split and payout speed — or by whether their drawdown rule is compatible with how my strategy actually makes money?

Almost everyone does the first. The second is what ends accounts.

We could not find anyone mapping strategy type to drawdown regime. As far as we can tell the argument is simply not made anywhere, which is strange given how mechanical it is. If you know of a source, we would genuinely like to read it.

Four things we have not solved

Stated plainly, because pretending to have answers is how this industry got its reputation.

Position sizing across prop accounts. The Kelly literature is entirely single-account. A prop account is a different shape: the bankroll is not yours, your downside is capped at the evaluation fee, and there is a hard barrier well above zero. Nobody has written the sizing rule for that.

Risk of ruin across N accounts. The arithmetic is not hard, but it only holds under independence — which copy-trading destroys. We found no source stating this in a prop context.

How many accounts is optimal. Every number in circulation — three, five, twenty — comes from what a human can physically manage, or from firm caps. None is derived from edge, variance, fee structure or pass probability.

Payout sequencing as design. First withdrawals are capped and rise with each consecutive payout, so account age has real value. Yet nobody writes about deliberately staggering start dates so the ladders desynchronise and the income smooths.

If you work any of these out, we would like to hear it.