Position Size & Risk Calculator
Work out how many units a fixed slice of the account can risk, what the trade pays right against wrong, and what full, half or quarter Kelly staking would be. Runs entirely in your browser.
About this tool
Two people can trade the same setup and get different results purely because of size. Risking 2% of the account instead of 0.5% does not make a strategy four times better — it makes four bad trades in a row cost 8% instead of 2%, and it is the run of losses, not the average trade, that ends accounts. This calculator starts from the opposite end: decide what one trade may cost, and let the distance to your stop determine the size.
The stop has to be on the correct side, and the page says so rather than quietly computing a number. Enter a stop above the entry on a long trade and most calculators will still return a positive, plausible-looking size — because the formula uses the absolute distance and never checks the direction. Requiring the direction explicitly turns a silent wrong answer into a message.
It also refuses to give a size you cannot actually trade. Positions come in whole units, and often in whole lots: a 10,000 account risking 1% with a two-point stop wants 50 units, which is impossible if your market trades in hundreds. The size rounds down to zero and the page says why — the alternative is an answer that looks precise and cannot be placed.
The second half of the page answers a different question with the same numbers. Win rate alone says nothing: 45% with a 2:1 payoff is profitable, 45% with 1:2 is not, and both look like "a coin flip". Expectancy per trade in R and the break-even win rate make the comparison honest, and Kelly turns the same two inputs into a stake size — reported alongside its half and quarter versions, since full Kelly assumes your inputs are exact and they never are.
What happens when the stop sits at the entry price
Nothing is risked per unit, so no size follows and the tool says so instead of dividing by zero. The equal-price case is tested before the direction check, since typing one price twice is what actually happened.
The size formula is the risk budget divided by the distance between the entry and the stop. Make those two prices the same and the distance is zero, a division with no answer — and the honest result is that there is no size, not an enormous one.
Which message to give is a smaller judgement worth naming. When the prices are equal, “the stop is on the wrong side” and “there is no risk here” are both technically true. The second is the one the user needs, because they typed a single price rather than reversing the two.
A distance that only looks non-zero is treated the same way. Prices are rounded to ten decimal places before the difference is taken, so a subtraction that should give zero is not reported as a risk of 0.0000000000000005 per unit.
What to do when the size you want needs more leverage than you allow
The page keeps the size the risk budget produced, flags that it exceeds your leverage, and also works out the largest position the cap permits — so you see both the size you asked for and the one you may take.
Leverage does not change what you lose at the stop; the stop distance does. What leverage limits is how much position the account may carry, and the two constraints can disagree: a tight stop produces a large notional that margin rules will not accept.
Reporting only “over the limit” would leave you to find the alternative yourself. The tool instead multiplies the account by your maximum leverage, divides by the entry price and rounds down to your lot size, giving the largest position that fits.
The margin figure follows the same rule — notional divided by the leverage you entered, which at a leverage of one is simply the position value. A maximum leverage above a hundred is refused outright, because beyond that it is not position sizing but wishful thinking.
Why the page answers when only some of the fields are filled in
Sizing needs the account, the risk, the direction and the two prices; expectancy needs the account, the risk and the win statistics. The groups are validated separately, so a blank win rate never blanks out the position size.
The page asks two questions that share only a couple of inputs. How large a position may be comes from the account, the risk per trade and the distance to the stop. Whether a setup is worth taking comes from the win rate and the average win and loss in R.
Checking everything as one block would tie the two together: leave the win rate empty and the size panel would complain about a field it never reads. So the checks are split into an account group, a trade group and a statistics group, and each half of the page runs only the groups it needs.
The practical effect is that a half-filled page still gives half an answer. Enter the account, the risk, the direction and the two prices and the position size is complete, whether or not you have measured a win rate yet.
Whole-lot rounding: how 1% of risk turns into 2%
Size is rounded down to the smallest tradable unit, so the loss at the stop can only land below the risk you set and never above it. On a small account the rounding can fall all the way to zero, and then the question is no longer 1% but whether one lot is affordable at all.
The lot rule is a constraint, not a suggestion. The raw size is divided by the lot, rounded down, then multiplied back. A 10,000 account risking 1% with an entry at 50 and a stop at 48 has a budget of 100 and a risk of 2 per share, which is 50 shares — rounded down to zero in a market whose smallest order is 100.
Zero is a real answer rather than an error: with that account and that stop, a single lot is out of reach. The page reports it as its own case instead of showing a plausible-looking number. The only ways out are raising the risk percentage or using a tighter stop, and both change the plan you started with.
The direction of the rounding does guarantee that the actual loss never exceeds the figure you set. Buying one lot anyway — 100 shares at 50 is 5,000 of notional — risks 200 at the stop, which is 2% of the account and twice the plan. That is the real price of a market that sells in hundreds, and it is the part small accounts miss most often.
Frequently asked questions
What is R?
One R is the amount you risk on a single trade — the account size multiplied by your risk per trade. With a 10,000 account risking 1%, one R is 100. Expressing results in R rather than currency is what makes a strategy comparable across account sizes: a trade that made 2R made twice the risk, whatever the R happened to be worth.
Why does the position size come out as zero?
Because the risk budget is smaller than one tradeable unit. A 10,000 account risking 1% will risk 100; with a stop two points away that is 50 units of theoretical size — but if your market trades in lots of 100, the smallest real position risks more than 100 (in fact, two points on 100 units is 200). The page shows the arithmetic rather than a number you could not act on.
Should I use full Kelly, half or a quarter?
Full Kelly maximises long-run growth only if your win rate and payoff are exactly right. They are estimates from a finite sample, and the formula is unforgiving: overstate the win rate slightly and the full-Kelly stake jumps. Half Kelly gives up roughly a quarter of the growth for much less of that sensitivity, and quarter Kelly is the usual choice when the estimate comes from fewer than a few hundred trades. Many traders size by fixed fractional risk and use Kelly only as a ceiling to check against.
Does the payoff ratio account for commission and slippage?
Only if you put them in. The average win and average loss should be the amounts you actually realise — after fees, and with the assumption that your stop fills at the price you set. If stops tend to slip, lowering the average win or raising the average loss in this calculator is the quickest way to see how much of an edge survives.
Does it store or send my account size?
No. Every figure is computed in the page, with no request, no storage and no logging. It works with your network disconnected, which is the simplest way to confirm that.
Related tools
Work out what regular investing adds up to — initial sum, monthly or yearly contributions, a chosen rate basis, and inflation-adjusted buying power, year by year. Nothing you type leaves your browser.
Combine any legs — long or short calls, puts and stock — and see the payoff at expiration: breakeven points, max profit and max loss, drawn to scale.
How many years does your savings rate buy you? Or what rate does a chosen deadline demand? Real, inflation-adjusted terms, with the 4% rule built in. Nothing you type leaves your browser.
Your watchlist and the whole A-share market board in one place: live quotes, intraday and candlestick charts, index levels, breadth, the limit-up ladder and capital flows.