Forex Compounding Calculator

Where this page starts from a different place

Every other compounding calculator in this search asks the same three things: how much are you starting with, what percentage do you make per period, and how many periods. Checked against the pages this keyword returns, the ones whose input fields could actually be read take precisely those three — a starting balance, a return percentage, and a count of years or months. None of them takes a risk percentage per trade.

This page runs from the other end, and that difference is not cosmetic. You do not know your return per period, so a page that asks you to supply one is asking you to invent the answer first. What a trader actually has is three quantities: the share of the balance staked on each trade, how many trades are coming, and — if the journal is long enough to say — what share of them tend to win and how large those wins run against the losses. Nothing here requires a forecast return, because the return per trade is not an input. It is two numbers already implied by how you trade: plus your average winner when the trade works, and minus exactly the amount staked when it does not.

So there is no interest-rate field, no "monthly gain" box, and no compounding-frequency dropdown. Put in the quantities above and this returns the whole set of places those trades can leave you — the middle one, the ones either side of it, and the averaged figure most pages quote, marked with its true rank among the outcomes. That last column is the thing this page has that nothing above it in this search does, and the rest of the write-up is about it.

How the balance is compounded

Two multipliers, applied over and over. Nothing else happens anywhere on this page.

1. The order does not matter for where you finish, only for what you have to survive getting there. Twenty losers first and one hundred winners after lands in exactly the same place as the same two hundred shuffled, because multiplication commutes. This is the single most useful thing about working in multipliers — and it is also why this page cannot tell you how deep the hole gets on the way. That is the drawdown calculator's job.
2. Every loser costs 1R and nothing else. That is what "risking 1%" means: the stop is placed so a loss takes out that share. It is why there is no average-loss field here — at this layer it is fixed at one by construction.
3. The stake is re-computed before every trade. After a winner the balance is larger, so 1% of it is a larger stake, and after a loser the same arithmetic runs the other way. That is the entire mechanism of compounding, and switching it off is one line of arithmetic away — see the next-but-one section.
4. Trades are treated as independent with a fixed rate. Real sequences cluster more than that allows: losers arrive in bunches when conditions change. The distribution here is therefore the optimistic reading of the spread, and the tables below are the floor of how wide it gets.

The percentile column is exact, not simulated. Where you finish depends only on how many of the N trades were winners, that count has a binomial distribution, and this page sums that distribution in full rather than drawing outcomes out of a hat — so the same inputs always return the same numbers.

Worked example: 10,000.00 at 1%, 200 trades, winning half at 1.5R

Everything typed into the fields above, everything repeatable on a phone. No number below was fetched, estimated or borrowed from a strategy.

1. The two multipliers. A loser takes 1% off: ×0.99. A winner pays 1.5R on a 1% stake, so it adds 1.5%: ×1.015.
2. Expectancy per trade, in R: 50% × 1.5 − 50% × 1 = 0.75 − 0.50 = 0.25R. Risking 1% of the balance, that is 0.25% of the balance expected per trade — and this is the one figure a fixed-rate calculator could be fed.
3. The averaged curve, which is what those pages print: 10,000.00 × (1 + 0.0025)^200 = 10,000.00 × 1.0025^200 = 16,476.93.
4. Typical growth per trade is not the same thing as expected growth per trade: 50% × ln(1.015) + 50% × ln(0.99) = 0.0074443 − 0.0050252 = 0.00241914, which compounds to 0.24% a trade rather than 0.25%.
5. The median run wins exactly half its trades — 100 winners out of 200. Its finish: 10,000.00 × 1.015^100 × 0.99^100 = 10,000.00 × (1.015 × 0.99)^100 = 10,000.00 × 1.00485^100 = 16,222.72.
6. The gap between the two: 16,476.93 − 16,222.72 = 254.21, or 1.57%. Small over two hundred trades. It does not stay small — the table further down runs the same input out to a thousand.
7. Where 16,476.93 actually ranks: summing the binomial from zero winners upward, the point where the accumulated probability passes it is 52.82%. It is a number above the middle of the range, and it is the one most-quoted figure in this whole category.
8. The tenths. Ninety-one winners out of two hundred is where the low tenth falls: 10,000.00 × 1.015^91 × 0.99^109 = 12,961.23. One hundred and nine winners is the high tenth: 20,304.92.
9. Compounding switched off. Keeping the stake at the original 1% — 100.00 per trade, never resized — and taking the same hundred winners: 10,000.00 + 100.00 × (1.5 × 100 − 100) = 10,000.00 + 100.00 × 50 = 15,000.00. Reinvesting was worth 1,222.72.
10. How long this takes. Two hundred trades at five a week is 40 weeks — about 9.2 months.

Step 7 is the one to take away. A compounding page that reports one number is reporting the arithmetic average, and the arithmetic average of a right-skewed distribution is not where anyone lands — it is the centre of mass, dragged upward by the runs that went well.

Why the averaged curve is not the answer

Put the same expectancy into a plain compound-interest tool and it will happily return 16,476.93. The arithmetic is not wrong. What is wrong is treating that as a destination, because the distribution behind it is lopsided: a winner is capped at 1.5R, a loser is capped at 1R, and the runs that string their winners together early go on compounding a larger stake, so the upper tail stretches further than the lower one ever can.

The consequence shows up in one number. The averaged figure lands at the 52.82nd percentile — 47.18% of runs finish above it, and the rest do not. Nobody quotes that alongside the headline, because the headline is the average.

The band is the point of the chart. It starts as nothing — before the first trade there is one possible balance — and widens with every trade taken, faster than most people expect, because each trade is a fresh chance to be on the wrong side of a multiplication that then has two hundred more trades to compound.

The further out you push it, the more the average lies

Same inputs as the worked example, only the trade count moves. The fourth and fifth columns are the ones worth reading together.

Over fifty trades the averaged figure is within half a percent of the middle and nobody should care. By a thousand trades it is 9,083 above it — and, more to the point, one run in ten still finishes at 68,234.00, roughly half what the average promised, from an edge that never changed. Nothing about the method deteriorated. The horizon did the damage.

Where risking more stops helping

The same two hundred trades, varying only the percentage staked. Every row below has the identical edge — 0.25R expectancy per trade.

Two things happen at once as the stake grows, and they pull in opposite directions. The median climbs — from 12,788.00 to 38,797.00 across those rows — because more of a positive edge is being deployed. And the gap between the averaged curve and the median widens from 0.39% to 14.87%, because the drag scales with the square of the stake. At 3% the averaged figure promises 44,567.00 while the typical run delivers 38,797.00, and the low tenth is sitting at less than half the average.

The right-hand column is the one to read before raising a stake. At 1% it takes 69 losers in a row to halve the account; at 3% it takes 23. Neither is remotely likely at a 50% win rate, but they are not the same kind of unlikely, and only one of them is the sort of thing a bad month actually produces. The win rate calculator publishes a streak column built on the same multiplication — twenty losers at 1% leaves 81.79%, which is 0.99^20 here as well.

Which stake to pick is deliberately not answered on this page. That is a question about maximising growth for a given edge, and it belongs to the Kelly criterion calculator. This page takes your stake as given and reports what it does.

Compounding against flat sizing

Switching compounding off means the stake stops tracking the balance — 100.00 per trade, every trade, whatever the account does. The comparison is worth putting in front of people who assume reinvesting is where the money is.

On the worked example it is worth 1,222.72 over two hundred trades, or 8.15% of the flat result. Real, and much smaller than the folklore about compounding suggests — because the effect is second-order at these sizes. To grow the stake you first have to have grown the balance, and at 1% a trade that growth arrives slowly.

The honest way to read it: compounding's contribution scales with how far above break-even the method is and how long it runs. Extending the horizon moves the final number far more than raising the stake does, which is visible in the last column of the table above — compare it against the median across rows.

"Compound interest calculator forex" — what people mean, and what changes

People arriving from a compound-interest background are looking for three fields: a deposit, a rate, and a period. There is no rate field here, so it is worth saying plainly why, and what to do instead.

You can borrow this page's own number if you insist on a rate. On the worked example the expectancy is 0.25R per trade and the stake is 1%, so the arithmetical expectation is 0.25% per trade. Feed that into any compound-interest tool alongside a starting balance and a trade count and it returns this page's 16,476.93 — the averaged-curve figure, which is to say the figure sitting at the 52.82nd percentile. If you use that route, use it knowing what it is: the average of a lopsided distribution, quoted as though it were a destination.

Daily compounding is a banking convention, not a trading one. A savings product compounds daily because interest accrues whether you act or not. A trading account compounds by trade, because the balance only changes while there is a position on — so the unit here is one trade, not one day. Translating is one division: five trades a week is roughly one a day, and the tool's third pane converts your horizon into months so the number has a shape you recognise.

Nothing here accounts for overnight financing. Swap charged on positions held past rollover is a cost against the balance, not against the arithmetic done here, and on a carry trade it can be the dominant term — in either direction. That side is handled on the forex profit calculator, where the financing line sits alongside the price move rather than inside it.

What happens when there is no edge to compound

Compounding does not create an edge, it re-applies one. Hand it a method breaking even and it re-applies that too, and every row below loses money at the median despite two of them showing a dead-flat average.

The first two rows are exactly break-even by expectancy and both still finish below where they started at the median. This is not a paradox and not a rounding effect: it is the same drag the tables above measure, running against a zero rather than against a positive edge. In the third row the edge is real but thin, and it takes two hundred trades to be worth less than the ten-percent spread around the result.

It is also why the high-tenth column deserves suspicion. Every row has a version where things went well — that is what the top of a distribution looks like — and a journal where the method is still floating because of four good weeks is looking at the wrong end of the range. Whether an edge exists at all before it gets compounded is the profit factor calculator's question.

What people call this

The search that leads here gets typed four ways, and they land on the same arithmetic:

• Compound interest calculator forex — the same idea in banking language. There is a conversion above: turn expectancy in R into a percentage of the balance per trade and any compound-interest tool will match this page's averaged column, minus the distribution.
• Forex compounding — the practice rather than the tool. Everything above describes what it does to a balance that is re-sized trade by trade.
• Forex compound interest — usually means the same thing, occasionally means overnight swap specifically. If it is the second, the financing line on the forex profit calculator is where that gets priced.
• Forex compound calculator daily — asking for a daily step. Compounding here happens per trade; the third pane converts the trade count into weeks and months so a daily cadence can be entered either way.

This page and its neighbours, side by side

Six pages on this site start from the same two quantities — a percentage of the balance and an outcome — and they split by which question they answer.

• This page projects a balance forward. It takes the stake, the horizon and the shape of a trade, and returns the set of places those land, ranked. Nothing here picks inputs for you.
• The kelly criterion calculator picks the stake. Same win rate and payoff, different output: it maximises the growth rate and returns a fraction of the balance. This page deliberately takes that fraction as given — it is the input the whole projection runs off.
• The win rate calculator looks backwards. Counts of wins and losses become a rate and a break-even. It also publishes the streak column this page cross-checks against: twenty losers deep, 1% leaves 81.79%.
• The profit factor calculator asks whether there is anything to compound. Two totals divided, and the win rate that ratio demands. Compounding a method whose ratio sits at 1.00 is a fast way to lose slowly.
• The risk of ruin calculator asks whether the account survives long enough to collect any of this, and the drawdown calculator prices the deepest point on the way — both of which this page skips, because it only looks at where the run finishes.
• The position size calculator on the home page goes the other direction. Give it a stop distance rather than a win rate and it converts a percentage into lots, which is the step immediately before every trade this page counts.

Questions traders ask

How much can I make compounding 1% a day?

Not a question with a single answer, and the distance between the answers is the whole subject of this page. Ten thousand risked at 1% over two hundred trades with a half-win rate at 1.5R lands at a median of 16,222.72 — with one run in ten finishing below 12,961.23 and one in ten above 20,304.92. Change the horizon and the same edge produces different numbers; change nothing and the draw you get out of the distribution does.

Why does this page not have a monthly return field?

Because a trader does not have a monthly return to enter — that figure is the output, not an input. What is known is the share of the balance risked per trade, how many trades are coming, and how wins and losses tend to size against each other. Everything here is built out of those, so no forecast has to be invented before the calculation starts.

What growth rate do I get risking 2% per trade?

It depends on the edge rather than on the stake. At 50% winners and 2R average winners, 2% risk carries a typical growth of about 0.96% per trade — which over two hundred trades puts the median at 66,979.44 from 10,000.00. At the same 50% with 1.5R it is roughly 0.47% a trade, and the median falls to 25,488.00.

Is it better to compound or to keep my lot size fixed?

Compounding wins on the median, and by less than most people expect at ordinary stake sizes. On this page's example it is worth 1,222.72 over two hundred trades — around 8% against the flat result. Flat sizing also has the property nobody prices: after a losing run the next stake is unchanged, so recovery arithmetic does not have to climb out of a smaller position.

What win rate do I need for compounding to work?

One above the break-even rate for your payoff, which is 1 ÷ (1 + payoff) — 40% at 1.5R, 33.3% at 2R. Compounding does not lower that threshold; at exactly break-even this page still shows the account finishing below where it started, because re-applying a zero edge in percentage terms loses money in compounding terms.

Why is the averaged number higher than the number I should expect?

Because the distribution is right-skewed. A loser can only remove the stake, while a winner pays a multiple of it and then compounds off a larger balance. The arithmetic mean gets dragged toward those runs — it sits at the 52.82nd percentile on the default inputs, meaning 47.18% of runs finish above it — while the median is where the typical run actually lands.

How long until my account doubles at 1% risk per trade?

Watch the median column as you raise the trade count. At 50% winners and 1.5R the typical growth is 0.24% per trade, which needs a little under 290 trades to double; at 2R it is about 0.49% per trade and takes roughly 142. The collateral number is worth reading alongside: at 1% risk a halving drawdown needs 69 losers in a row.

Does this include spread, commission and swap?

No. Everything here runs on the risked amount and the multiple the winner pays, so costs have to be folded into your inputs rather than being deducted afterwards — trim the average winner by the round-turn cost expressed in R, or lower the stake. That decay is priced properly on the spread and commission page, and overnight financing separately on the forex profit calculator.

What this page will not tell you

Three gaps, and each one is a page rather than a footnote.

• How deep the hole gets getting there. This page only looks at where the run finishes, and multiplication commutes — the same two hundred trades land identically whatever order they arrive in. The path is the part that decides whether anyone is still there to collect it.
• Whether the edge is real. The projection assumes the inputs hold for all two hundred trades. A method measured over forty has a win rate wide enough that several rows of these tables overlap.
• What stake you should pick. Deliberately absent. This reads what a stake does, not which one to choose.

The distribution on this page is computed in full rather than simulated, so the same inputs always return identical figures. It assumes each trade is independent with a fixed win rate and a fixed payoff — an assumption that flatters the result, because real losing runs cluster. There is no dataset behind any table here: every number comes out of the values at the top of the page, and nothing leaves your browser.

Size the next trade from your risk

Related guides

• How much should you risk per trade? — where the 1% and 2% this page projects actually come from.
• Kelly criterion calculator — which stake maximises growth for your win rate and payoff.
• Win rate calculator — turning a journal into the rate this page projects with, plus the streak column.
• Profit factor calculator — whether there is anything worth compounding before you run the projection.
• Risk of ruin calculator — the chance the account reaches a drawdown you pick at 0.5%, 1% or 2% per trade.
• Drawdown calculator — the deepest point on the way, which this page skips entirely.
• Forex profit calculator — overnight swap and financing, priced alongside the price move.
• Adding to a position — the same budget split over several entries rather than one.