Here is the part of portfolio trading almost nobody teaches: the same strategies should not run the same way on every account. A prop firm EA strategy lives or dies by a daily drawdown rule. A Darwinex track is scored on the relationship between profit and drawdown, with no daily guillotine. Personal capital answers to nobody but compounding math. Same method underneath, three different referees, three different configurations. Deciding what runs where, at what risk, is the fourth stage of my methodology, and in practice it is half the job. This post is how that decision actually gets made across the four account types I run in public.
Everything described here is auditable: the funded accounts, the challenge batch, the DARWIN with its €30,000 allocation and the personal accounts all sync live to my track record page. The architecture above this layer (why the accounts themselves form a portfolio) is in the portfolio of portfolios; this post is the floor below: per-account tuning.
And if you have ever taken a strategy that worked beautifully on your own account, dropped it unchanged into a funded challenge, and watched a daily drawdown rule execute it in week two: the strategy was fine. You were fine. What failed was an assumption the industry never warns you about, because the industry mostly sells accounts and strategies as if they were interchangeable containers and contents. They are not. The container has rules, and the rules change what the contents should look like.
The Principle: The Account’s Rules Are Part of the Market
Traders think of “the market” as price. But operationally, your market is price plus every rule that can end your participation. A 5% daily drawdown limit is as real as a spread. A rating formula that punishes volatility is as real as slippage. Personal capital’s rule (your own psychology’s breaking point) is the most real of all. So the tuning question for any account is never “what is the best configuration of my portfolio?” but “what is the best configuration that survives this referee?”.
Concretely, three dials move between accounts while the strategies themselves stay identical: risk per trade, which strategies from the stable get a slot, and how much simultaneous exposure the account may carry. Tuning those three to the account’s rulebook is the whole discipline. What never changes: the strategies’ logic. The moment you rewrite a strategy’s behavior per account, you no longer have one method with configurations; you have untested mutants everywhere, and none of your testing evidence applies to any of them.
Account Type One: Prop and Funded Accounts (The Strict Referee)
Funded accounts have the harshest rulebook: daily drawdown limits that do not care that your week was fine on average. I have written about how daily drawdown rules kill EAs; the allocation answer is defensive tuning. Risk per trade drops well below what the raw portfolio could carry. Strategy selection favors the smoothest members of the stable: steadier profiles earn slots, streaky ones stay home even when their long-run numbers are better. Simultaneous exposure gets capped so that a bad day across correlated positions cannot approach the daily limit.
On top of tuning sits structure: challenges are a batch, never a bet. I run FundedNext challenges three or four at a time, expecting casualties, requiring only that the group comes out positive. That is the honest arithmetic of evaluation-based funding, and it changes the psychology completely: an individual failed challenge is a line item inside a positive expectancy, not a catastrophe. My core funded capital, though, is where the incentives are cleanest: Axi Select, no challenge fee, scaling built on continuity (the full comparison with Darwinex is in Axi Select vs Darwinex Zero).
Account Type Two: Darwinex Zero (The Mathematician Referee)
Darwinex Zero has no daily guillotine. Instead, everything funnels into a rating driven mostly by profit against drawdown, with the bar for guaranteed allocations at 75 (mechanics in how Darwinex Zero works). That referee rewards exactly one personality: smooth and patient. So the DARWIN gets the configuration a funded account would call conservative and a mathematician would call optimal: moderate risk, the steadiest strategy selection, exposure managed for curve quality over speed. Ironically, the tuning that satisfies a strict prop rulebook and the tuning that maximizes a Darwinex rating end up cousins; both referees, in their own language, are pricing the same thing: control.
What is different is tempo. A prop account defends itself day by day. A Darwinex track is built over months (mine ran from April to the July rating that triggered the €30,000 allocation), so the configuration optimizes for consistency across a long window, and no single day matters much. Same strategies. Different clock.
Account Types Three and Four: Personal Capital (The Honest Referee)
Personal accounts carry the fullest expression of the portfolio: every proven strategy gets its slot, risk sits at the level long-term compounding math actually supports, and drawdowns are absorbed rather than feared. This is where the method runs closest to its theoretical shape, because no external rule distorts it. It is also where new survivors of the testing funnel do their live incubation at small size, before they are trusted anywhere stricter.
The honest referee still has a rulebook: yours. The correct risk level for personal capital is not what the math supports; it is the minimum of what the math supports and what your nerves can hold through the worst stretch without touching the system. Set it wrong and you become your own daily drawdown rule, executed emotionally at the exact bottom. Most traders should tune personal accounts more conservatively than any formula suggests, for exactly this reason.
The Decision Table (Steal This)
For every account you run, answer four questions in writing. What ends my participation here? (Daily rule, rating floor, my own panic threshold.) Which strategies fit this referee? (Smooth ones for strict rules; the full stable for patient capital.) What risk per trade keeps the worst realistic day far from the limit? What is this account’s single job in the structure? If two accounts produce identical answers, one of them is redundant. If an account’s answers contradict its configuration today, you have found this week’s homework. That is the entire discipline, and it fits on an index card.
The Honest Close
Allocation per account is unglamorous, which is why nobody sells it: there is no thumbnail for “I risked less on the account with stricter rules”. But walk through any graveyard of failed funded accounts and most tombstones read the same: right strategy, wrong configuration for the referee. The method series this post belongs to (generation, robustness, sequence risk, and now allocation) is being published piece by piece, with each installment landing in the newsletter first. And as always: every account described here is live and public on the track record page, so you can check whether the man recommending defensive tuning survives his own referees. Audit first, believe later, and apply that order to everyone.
Frequently Asked Questions
Should an EA run the same settings on a prop firm account and a personal account?
Almost never. Prop accounts add rules (daily drawdown limits, overall loss caps) that function like additional market conditions, so the same EA portfolio typically needs lower risk per trade, fewer simultaneous positions and a bias toward its steadiest strategies. Personal accounts can carry the fuller configuration. The strategies’ logic should stay identical everywhere; only risk, selection and exposure should change per account.
What is the best EA strategy for prop firm accounts?
Whatever your tested portfolio’s smoothest subset is, configured so a realistic bad day stays far from the daily drawdown limit. Prop referees reward equity-curve stability, not peak performance, so streaky strategies with great long-run numbers are often wrong for challenges even when they are right for personal capital. The evaluation is won by the configuration that cannot be disqualified, not the one with the highest theoretical return.
How do you approach prop firm challenges without gambling?
Treat challenges as a batch with positive expectancy rather than a single bet. Take several, trade them with the defensive configuration, expect some to fail, and require the group to come out ahead. This converts an emotional pass/fail exam into arithmetic, and it only works with challenge fees you can afford to lose in multiples, honest per-challenge risk, and the discipline not to revenge-trade a failed attempt.
What configuration works best on Darwinex Zero?
The rating that drives allocations is dominated by the relationship between profit and drawdown, so the optimal configuration is moderate risk with maximum smoothness, sustained over months. Aggressive months tend to hurt the rating rather than help it. In practice the Darwinex-optimal setup looks similar to a well-tuned prop configuration on a longer clock: both referees are pricing control, just with different formulas.
How much should risk differ between account types?
There is no universal number, but the ordering is consistent: strict-rule accounts (challenges, funded) run the most defensive risk, rating-based tracks run moderate, and personal capital runs whatever long-term math supports, capped by your honest psychological tolerance. The written test: for each account, the worst realistic losing day should remain comfortably clear of whatever ends your participation there. Size backward from that line, not forward from ambition.
