Risk-Reward and Expectancy: The Only Two Numbers That Matter

Win rate on its own tells you nothing. Expectancy combines how often you win with how much you win and lose, and it is the single number that decides whether a strategy makes money.

Risk-Reward and Expectancy at a glance

What it is and why it works

Two numbers describe every trade you take: how much you stood to lose if you were wrong, and how much you actually made or lost. Express the second as a multiple of the first and you have the R multiple. Risk £100, make £300, and the trade was +3R. Risk £100 and hit your stop, and it was −1R. Working in R rather than in pounds lets you compare a gold trade with a EUR/USD trade and a trade from last year with one from today.

The risk-reward ratio is the version of this you decide before entering: the distance from entry to target, divided by the distance from entry to stop. If your stop is 20 pips away and your target is 60, that is a 1:3 ratio, and if it works you make 3R.

Expectancy is what happens when you combine that with how often you are right. The formula is straightforward:

Expectancy = (win rate × average win) − (loss rate × average loss)

It answers one question: on average, across many trades, what does each trade produce? A positive expectancy means the strategy makes money over a large enough sample. A negative one means it loses money over a large enough sample, no matter how it felt or how many individual trades went well. There is nothing else to appeal to.

How to trade it, step by step

1. Record every trade in R, not in currency. For each trade write down what you risked, then divide the result by that amount. A trade risking £50 that made £150 goes in the book as +3R. Do this for winners and losers alike, and include the trades that ended somewhere in between.
2. Calculate your win rate from at least 50 trades. Divide the number of winning trades by the total. Forty winners out of a hundred trades is a win rate of 0.40, or 40%. Your loss rate is simply one minus that: 0.60. Fewer than fifty trades and the figure will move around too much to rely on.
3. Calculate the average win and the average loss, separately. Add up all your winning R multiples and divide by the number of winners, then do the same for losses. Keep them apart; the whole point of expectancy is the relationship between them, and a combined average destroys it. Suppose the winners average +3R and the losers average −1R.
4. Put the four numbers into the formula. Expectancy = (0.40 × 3) − (0.60 × 1) = 1.20 − 0.60 = +0.60R per trade. That is the whole calculation. Every trade you take is worth, on average, six tenths of what you risk on it.
5. Translate it into money so it means something. With a £10,000 account risking 1% per trade, one R is £100, so expectancy is £60 per trade. A hundred trades has an expected value of £6,000 before costs. Check it the long way if you like: 40 wins × £300 = £12,000, minus 60 losses × £100 = £6,000, which leaves £6,000. The formula and the arithmetic agree.
6. Now compare that with a high win rate at a poor ratio. Take a strategy that wins 70% of the time but takes profit at half of what it risks: (0.70 × 0.5) − (0.30 × 1) = 0.35 − 0.30 = +0.05R per trade. It is still positive, but it produces £5 per trade against the other strategy’s £60. The 40% strategy is twelve times better, and it feels twelve times worse to trade.
7. Work out the win rate your ratio requires to break even. Breakeven win rate = 1 ÷ (1 + reward-to-risk). At 1:1 that is 1/2, or 50%. At 1:2 it is 1/3, about 33%. At 1:3 it is 1/4, or 25%. Anything above that line is a positive expectancy, and knowing the line stops you abandoning a working strategy because it “loses too often”.
8. Subtract your costs before deciding anything. Spread, commission and slippage come out of every trade, and on short-term strategies they can consume most of the edge. If your average winner is 15 pips and your round-trip cost is 2 pips, more than 13% of the gross has gone before you start. Recalculate expectancy on net results, not on the theoretical target.
9. Recalculate quarterly and act on the components, not the total. If expectancy falls, look at which of the four inputs moved. A dropping average win usually means you are taking profit early; a rising average loss means stops are being moved or ignored; a falling win rate may mean conditions have changed or you are taking lower-quality setups. Each has a different fix.

Size every one of those entries with the position size calculator and check the trade is worth taking with the risk/reward calculator before you commit.

The conditions it needs

Enough trades for the average to mean anything

Expectancy is a statement about a long run, not about the next trade. With a 40% win rate you will see four losses in a row regularly and eight occasionally, and none of that contradicts a positive expectancy. Fifty trades is the minimum for a usable estimate and a few hundred is far better, which is why a journal is a prerequisite rather than an optional extra.

Consistent risk per trade

The R framework assumes each trade risks the same fraction of the account. If you risk 1% on some trades and 4% on others, your R multiples are no longer comparable and a single oversized loss can outweigh a long run of correctly sized winners. Fixed fractional sizing is what makes the arithmetic valid: see position sizing.

Stops that are actually honoured

Every calculation here assumes a loss costs 1R. Moving a stop turns a −1R into a −2R or worse, and because the loss side of the formula is multiplied by the loss rate, a small increase in average loss destroys expectancy quickly. One trade allowed to run to −5R can wipe out eight correct ones.

Targets the market can realistically reach

You cannot raise the ratio for free. Moving the target further out lowers the win rate, because price has to travel further before you are paid. The right target comes from the instrument’s actual range and the structure on the chart, not from a rule that every trade must be 1:3. A 1:3 target that is only hit 10% of the time has a negative expectancy.

When it fails

• Judging a strategy by win rate. Win rate is one of four inputs and on its own it says nothing. A method that wins 30% of its trades at 4R is strong; one that wins 80% at 0.2R loses money. Traders chase win rate because being right feels good, and that preference is the most expensive one in retail trading.
• Cutting winners and letting losers run. This is the same mistake seen through the formula. Closing a winner at +0.7R because you are nervous lowers the average win; giving a loser room because it might come back raises the average loss. Both terms move the wrong way at once, which is how a positive-expectancy strategy is turned negative by the person trading it.
• Assuming a ratio can be chosen without affecting the win rate. Setting a 1:5 target does not create a 1:5 strategy. It creates a strategy with a much lower win rate, and whether that is an improvement depends entirely on how often the market actually travels that far. The two numbers move together and must be measured together.
• Calculating on too small a sample. Twenty trades can show a strongly positive expectancy purely by chance, and the same twenty trades from a different fortnight can show the opposite. Acting confidently on a small sample, particularly by increasing size, is one of the most common ways traders convert a mildly good period into a serious drawdown.
• Leaving costs out of the calculation. Spread, commission and swap are subtracted from every trade regardless of outcome. Strategies with small targets often have a positive gross expectancy and a negative net one, which is why they look fine in a spreadsheet and lose money in an account.
• Expecting expectancy to apply to the next trade. A +0.6R expectancy does not mean the next trade returns 0.6R; it means the average does, over many trades. Traders who expect the edge to show up immediately abandon working methods during ordinary losing runs, which is the most common way a real edge gets thrown away.

For different levels of experience

If you are brand new

Here is the shortest useful version. Before every trade, ask how much you lose if you are wrong and how much you make if you are right. If the second is not clearly larger than the first, do not take it. That one habit puts you ahead of most beginners.

Then start writing trades down in R. Risk the same small amount each time (1% of the account, say) and record each result as a multiple of it: +2R, −1R, +0.5R. After fifty trades you can calculate your own expectancy with the formula on this page, and you will know something about your trading that no amount of watching videos would have told you.

The part that surprises most new traders is that losing more often than you win is completely normal and completely fine. At a 1:3 ratio you only need to be right one time in four to break even. Being wrong is not failure, being wrong expensively is. Work targets and stops out beforehand on the risk-reward calculator.

If your results are inconsistent

If you have been trading a while and are stuck around breakeven, the diagnosis is almost always in these four numbers, and you can find it in an afternoon. Pull your last hundred trades, calculate win rate, average win and average loss, and put them in the formula. The result tells you which of three different problems you have.

If expectancy is slightly negative and your average loss is bigger than 1R, you are not respecting stops, that is a discipline problem with a mechanical fix. If your average win is below your intended target, you are taking profit early, usually out of fear of giving back gains. If both are as planned but the win rate is below the breakeven line for your ratio, the setup itself has no edge and no amount of discipline will rescue it.

Those three problems look identical from the inside, all of them feel like “I keep almost making it work” and they have completely different solutions. That is what expectancy is for: it turns a vague sense of underperformance into a specific number that moved.

If you are experienced

Per-trade expectancy is the first moment and is necessary but not sufficient. Two systems with identical expectancy and very different variance are not equivalent, since the path determines drawdown, position-sizing capacity and, in practice, whether the method is tradeable at all. Standard deviation of R multiples, and the resulting distribution of maximum drawdown, matter as much as the mean for anything you intend to size properly.

Estimation error also deserves more attention than it usually gets. A hundred-trade sample gives a confidence interval on win rate wide enough to accommodate materially different expectancies, and the fatter the R distribution the worse this becomes. Where results depend on a small number of large winners, as trend-following results do, the mean is dominated by the tail and ordinary summary statistics are unstable. Bootstrapping the trade sequence gives a far more honest picture of the drawdown distribution than any single realised curve.

For sizing, expectancy feeds directly into Kelly and its fractional variants, but the inputs are estimates with error, so the optimal fraction is a point estimate on a curve that punishes overestimation severely and underestimation mildly. That asymmetry is the whole argument for fractional Kelly. Finally, expectancy assumes trade independence: if your results cluster by regime, and most do, sequence risk is understated by the per-trade figure and should be modelled separately.

Risk management for this strategy

Expectancy and position sizing are two halves of one system, and the arithmetic on this page only holds if the second half is in place. Every R multiple assumes the same amount was at risk on every trade. Vary the risk and the averages become meaningless: one oversized loss recorded as −1R when it actually cost four times a normal trade corrupts the whole dataset.

The practical rule is to fix the percentage first and let it determine the lot size, rather than fixing the lot size and discovering the risk. Decide on 0.5% or 1% per trade, place the stop where the idea is invalidated, and use the position size calculator to convert the distance into a lot size. That keeps every trade worth exactly 1R.

Remember too that a positive expectancy does not protect you from a losing streak, and streaks are longer than intuition suggests. At a 40% win rate, five consecutive losses is an ordinary event and should not require a change of plan. Size so that a run of eight is uncomfortable rather than catastrophic, and read risk management alongside this page.

Where Market Structure Pro fits

Expectancy is arithmetic, and the arithmetic is the easy part. The hard part is the input: how do you take the higher-quality setups more often, so that the win rate side of the equation is worth having? Filtering out the trades that were never good enough is where most of the improvement in a real trader’s expectancy comes from.

Market Structure Pro is built around that filter. It fuses 27 tools into one verdict (TRADE, TRANSITION or NO TRADE) with a confidence percentage, an A/B/C grade and a plain-English explanation of what is driving it. The grade is directly useful here, because it lets you tag every journal entry with the read at the time and then calculate expectancy separately for A-grade and C-grade trades. If your C-grade trades have a negative expectancy, you have found real money without changing anything about your entries.

The ranging and chop filter matters for the same reason. Its whole job is to return NO TRADE in dead or choppy conditions: exactly the conditions that produce the small, scrappy losses which drag the average win down and the loss rate up. Because the state locks on the closed bar and does not repaint, the record you review afterwards is the record you saw at the time. MSP is decision support: it places no trades, is not a signal service and guarantees nothing. What it can do is make the quality of each trade a recorded variable rather than a feeling.

Frequently asked questions

How do you calculate expectancy in trading?

Expectancy = (win rate × average win) − (loss rate × average loss). Work in R multiples, where one R is the amount risked on the trade. For example, a 40% win rate with an average win of 3R and an average loss of 1R gives (0.40 × 3) − (0.60 × 1) = 1.20 − 0.60 = +0.60R per trade.

What is a good risk-reward ratio?

There is no universally good ratio, because raising the target lowers the win rate; the two have to be judged together. Most traders find that something around 1:2 or 1:3 is where realistic targets and achievable win rates meet, and at 1:3 you only need to win more than 25% of trades to break even. A ratio is only good if the market on that instrument actually travels that far often enough.

Is a high win rate important in trading?

Much less than most traders assume, because it is only one of four inputs into expectancy. A strategy winning 40% of trades at 3R produces +0.60R per trade, while one winning 70% at 0.5R produces just +0.05R: twelve times less, despite feeling far more successful. Chasing win rate usually means cutting winners short, which lowers the average win and hurts overall results.

What win rate do I need to be profitable?

It depends entirely on your reward-to-risk ratio. The breakeven win rate is 1 divided by (1 + reward-to-risk): 50% at 1:1, about 33% at 1:2, and 25% at 1:3. Anything above that line is profitable before costs, and costs mean you need a little more than the theoretical figure.

What is an R multiple?

An R multiple expresses a trade’s result as a proportion of the amount risked on it. Risking £100 and making £300 is +3R; hitting the stop is −1R. Using R rather than currency makes trades on different instruments and at different account sizes directly comparable, which is what allows expectancy to be calculated across a whole trading record.

How many trades do I need before expectancy means anything?

At least 50 for a rough figure and several hundred for a reliable one. Small samples move around dramatically, and twenty trades can show a strongly positive expectancy purely by chance. The risk of acting on too few trades is that traders increase position size on the strength of a number that was never stable.

Does expectancy include spread and commission?

It should. Calculate it from net results after spread, commission and any swap, because those costs come out of every trade whatever the outcome. Strategies with small targets frequently show a positive gross expectancy and a negative net one, which is precisely why they look workable on paper and lose money in a live account.

What is the difference between expectancy and profit factor?

Profit factor is gross profit divided by gross loss, so a value above 1 means the strategy made money overall. Expectancy expresses the same information as an average per trade, which makes it more useful for planning because it tells you what one trade is worth and therefore what a hundred trades might produce. Expectancy also combines directly with position sizing, whereas profit factor does not.

Why is my expectancy falling?

Look at which of the four inputs moved rather than at the total. A falling average win usually means winners are being closed early; a rising average loss means stops are being widened or ignored; a falling win rate points either to changed market conditions or to lower-quality setups being taken. Each cause has a different remedy, which is why the components are more informative than the headline number.

Related reading

• Risk-Reward Calculator: Turn a stop and target into a ratio and a breakeven win rate.
• Position Sizing: The other half of the system: keeping every trade worth exactly 1R.
• Trading Journal and Review: You cannot calculate expectancy without recorded trades.
• Risk Management: Surviving the losing streaks a positive expectancy still produces.
• Backtesting: Estimating expectancy before risking money, and the traps in doing so.