The calculation
Expectancy per trade, as a multiple of risk, is:
E = (win rate × reward-to-risk) − (1 − win rate)
A strategy winning 45% of the time at 1.8R has an expectancy of (0.45 × 1.8) − 0.55 = 0.26R per trade. To make 10% of the account you need 10% ÷ 0.26 ≈ 38R of accumulated gain, which at 1% risk per trade means roughly 38 trades of net progress — realistically 80 to 120 trades taken.
That is the half everyone does. It answers "can I reach the target", and on its own it is misleading, because it says nothing about what happens on the way.
Now check it against the drawdown
With a 45% win rate, a run of eight consecutive losses is entirely ordinary over 100 trades — its probability is around 1 in 3 across a sample that size, which makes it something to plan for rather than something to hope against. At 1% risk that run costs 8%, which breaches a 6% drawdown and survives a 10% one.
So the question is not "what risk reaches the target" but "what risk reaches the target without a normal losing run ending the account". Those two numbers are often incompatible, and when they are, no amount of discipline fixes it — you are being asked to be lucky rather than good.
| Risk per trade | 8-loss run costs | 12-loss run costs | Trades of net progress to 10% |
|---|---|---|---|
| 0.25% | 2% | 3% | ~154 |
| 0.5% | 4% | 6% | ~77 |
| 1% | 8% | 12% | ~38 |
| 2% | 16% | 24% | ~19 |
Read that table against your own drawdown. On a 10% limit, 1% risk survives eight losses and dies on twelve. On a 6% limit, only 0.5% and below is defensible. The row you can afford is the row you trade, regardless of how long it makes the evaluation.
A worked case
Target 10%, total drawdown 6% trailing, win rate 45%, reward-to-risk 1.8.
- At 1% risk: about 38 trades of net progress, but an eight-loss run costs 8% and breaches. Not viable, however good the strategy is.
- At 0.5% risk: the same run costs 4% and survives. You now need roughly 77 trades of net progress — a long evaluation, though where there is no time limit that is entirely fine.
- At 2% risk: you reach the target quickly if the sequence cooperates, and breach on any four-loss run. This is the choice most failed evaluations made.
The viable answer is 0.5% and patience. See how long a challenge really takes for what that means in calendar terms, and note that on a trailing drawdown the 6% is not even fully available — the trail moves up behind you, which is the correction covered in position sizing against a drawdown.
The third constraint: the daily limit
Two constraints are usually discussed and three usually apply. The daily loss limit binds sooner than the maximum drawdown, and it binds on a shorter sample where variance is larger.
At 1% risk against a 5% daily limit, five losing trades ends your day — and four ends it where the limit is measured on equity and one position is still open. If your strategy takes six to eight trades a session, you will hit that limit on an ordinary bad day rather than an unusual one, and a stopped day is not a breach but it is a day you cannot make progress.
Work out how many losses your session can absorb before the daily rule stops you, and if the answer is fewer than your typical trade count, your size is wrong for your frequency regardless of what the drawdown arithmetic says.
Test the assumption, not just the arithmetic
Every number above rests on a win rate and a reward-to-risk you supplied, and those are the least reliable figures in the calculation. Backtests overstate both, live records under small samples are noisy, and traders remember their winners.
So run the calculation twice: once with your believed numbers and once with the win rate five points lower and the reward-to-risk 0.2 lower. If the conclusion holds in the pessimistic version, you have a plan. If the strategy only works at exactly the numbers you hope for, you have a forecast.
At 40% and 1.6R, the earlier example's expectancy falls from 0.26R to 0.04R — a strategy that needs roughly 250 trades of net progress to make 10%, which is a completely different evaluation from the one you thought you were buying. That sensitivity is why sizing conservatively is not caution; it is the correct response to genuine uncertainty about your own inputs.
When the maths says no
Sometimes there is no risk level that both reaches the target and survives a normal losing run. A 10% target against a 4% intraday trailing drawdown is close to that for most strategies: the risk needed to survive the streak is so small that the trade count required becomes unrealistic, and the risk needed to finish in reasonable time breaches on any bad week.
The correct response is to choose a different firm, not to accept worse odds and hope the sequence is kind. The target-to-drawdown ratio is a documented field on every firm profile precisely so this comparison is possible before you pay rather than after — and it is the first thing compared in how to choose a prop firm.
Fixed fractional or fixed dollar
A detail that matters more under an evaluation than on a personal account. Risking a fixed percentage of the current balance means your size falls automatically as the account draws down, which protects the tail of a losing streak — the eighth loss is smaller than the first. Risking a fixed dollar amount set at the start does not.
Under a static drawdown, fixed fractional on the balance is the better default. Under a trailing drawdown, neither is right on its own: the correct base is the buffer to the fail level, which can shrink while the balance grows. Recomputing from the buffer each morning is the only version that stays correct in both regimes.
Improving the inputs
Only three things change the arithmetic: a higher win rate, a better reward-to-risk, or lower risk per trade. The first two require actual strategy improvement, which is slow and uncertain. The third is available immediately and costs only time.
Under a drawdown constraint, more trades at lower risk is almost always the better trade-off, because the constraint punishes variance rather than slowness. A strategy that reaches the target in 150 trades at 0.4% and one that reaches it in 40 at 1.5% have the same expectancy and very different survival probabilities — and the evaluation only pays the one that survives.
Do it with your own numbers
Take your real win rate and reward-to-risk from your own record — not from a backtest, not from the strategy's advertised performance — and run the calculation before buying anything. Then run it again five points lower.
The position size calculator covers the sizing half; the expectancy half is the formula at the top and a spreadsheet. Fifteen minutes of it settles whether an evaluation is a purchase or a lottery ticket, and it is the same fifteen minutes whether the account is $10,000 or $200,000.