Realistic ROI on a Funded Prop Account

The screenshots show a five-figure payout on a six-figure account and imply that’s your monthly income. The math that survives contact with real fees, profit splits, and drawdown looks very different — and it’s the only math worth planning your life around.

Why advertised prop returns are misleading

Marketing quotes the notional account size, not your capital, and returns are framed against the whole balance as if it were yours. It isn’t. A funded account is a mandate to trade someone else’s money under strict rules, and what reaches your bank is a slice of the profit — after the firm takes its share.

Three things get quietly dropped from the headline number:

The challenge fee you paid to get funded, which is a real cost even when it’s later refunded.
The profit split, which hands a meaningful chunk of every winning month back to the firm.
The survivorship framing — you only ever see the accounts that paid out, never the far larger pool that breached.

So “ROI” on a funded account is not return on the account size. It’s return on your money at risk: fees, resets, and the time you spent earning the payout. Model that instead.

Modeling monthly profit from your expectancy

Start from the one number that actually generates money: your expectancy per trade, measured in R (units of risk). If you don’t have a firm figure, derive it from your real trade history with the expectancy calculator, and read trading expectancy if the concept is fuzzy.

A rough monthly profit model looks like this:

Monthly R gained = Expectancy per trade × Trades per month

Multiply that by the dollar value of 1R (your risk per trade in currency) to get gross monthly profit on the account. Two honest constraints keep this grounded:

Use your historically observed expectancy, not an aspirational one. A number pulled from twenty trades is usually luck, not edge.
Use a realistic trade count. Overtrading to hit a bigger number typically lowers expectancy, so the two inputs aren’t independent.

This gives you gross profit on the account. It is not yet income.

Netting out fees, splits and reset spend

Now subtract everything the headline number ignored. Work through it in order:

Profit split — the firm keeps its percentage of every profitable payout. Apply it to gross profit first; confirm the exact split with your firm since these change.
Challenge fee — amortize what you paid to obtain the account across the months you expect to hold it. Even a refunded fee is capital you fronted and can’t spend elsewhere.
Reset and retry spend — money paid to reset a breached evaluation, or to buy a fresh one, belongs in the cost column whether or not this particular account is the one that paid.

The payout calculator turns a gross monthly profit and a split into the number that actually lands in your account, which is the figure to anchor on. Everything above the split line is the firm’s; everything below the fees is yours.

Annualizing a realistic payout figure

Resist multiplying one good month by twelve. Real funded trading has losing months, flat months, and payout cycles that don’t line up neatly with the calendar. A defensible annual figure accounts for:

Variance — some months you clear the split threshold, some you don’t reach a payout at all.
Payout cadence — many firms pay on a fixed cycle, so profit sitting in the account isn’t cash until the window opens. Firms like Topstep publish their cadence; confirm the current terms directly.
Account mortality — a breach resets your realized-profit clock to zero, and the odds of that over a year are not small.

A grounded annualization multiplies a median month — not your best — by the number of months you realistically expect the account to survive and pay. That’s usually well under twelve.

Stress-testing the model against drawdown

The final step is asking what a bad run does to the whole picture. A funded account can be net-positive on paper and still end the year negative for you once a breach wipes out unpaid profit and forces new fees.

Pressure-test with pessimistic inputs:

Shave your expectancy down to the low end of its confidence interval, not the point estimate.
Assume one breach during the year and add the cost of getting funded again.
Assume one flat month where you clear nothing.

If the model still shows positive net ROI on your capital under those assumptions, funded trading is a business for you. If it only works when every input is optimistic, you’re planning on luck.

This is where continuous measurement earns its keep. Shibiki auto-journals every trade, tracks your live expectancy with a Wilson confidence interval so you know whether your edge is real or a lucky streak, and enforces hard risk limits at the broker so a single bad day can’t breach the account and reset your ROI to zero. Model the economics once — then let the numbers keep themselves honest.