Trading Expectancy: The Number That Says If a Strategy Works
Trading expectancy is your average result per trade. Learn the formula in R, how costs and sample size change it, and how to measure it from your journal.
You have taken twenty trades with a new method. You might be up, you might be down, and either way you want to know the same thing: does this strategy actually work? Trading expectancy is the number that answers that question, as long as you measure it honestly and over enough trades.
Quick answer: Trading expectancy is the average amount a strategy makes or loses per trade over many trades. Expectancy = (win rate × average win) − (loss rate × average loss). Measured in R, a method that wins 40% of the time at 2R has an expectancy of +0.2R. It must stay positive after costs, over a large sample.
What is trading expectancy?
Expectancy is the average result of one trade, taken across a long run of trades that all follow the same rules. It is the single number that combines how often you win with how much you win and lose. A strategy with positive expectancy makes money over time; one with negative expectancy loses it, however good individual weeks look.
The formula has two halves, one for the winners and one for the losers:
Expectancy = (win rate × average win) − (loss rate × average loss)
Take a method that wins 40% of its trades, makes 2 units on each winner and loses 1 unit on each loser. Its expectancy is (0.40 × 2) − (0.60 × 1) = 0.8 − 0.6 = +0.2 units per trade. Over ten typical trades, that is four wins worth +8 units and six losses worth −6, a net +2, or +0.2 per trade.
Two things about this number trip people up. First, expectancy is an average over many trades, not a forecast for the next one. No single trade in the figure made +0.2R; each one made +2R or lost 1R. Second, the order of wins and losses is random. The four winners could just as easily have come last, after six losses in a row, and the expectancy would be exactly the same.
Expectancy is what traders mean when they talk about an edge. A method has an edge if its expectancy is reliably above zero after all costs. Everything else, including a high win rate, a beautiful equity curve over a month, or a confident feeling about your entries, is secondary. Our guide to the risk reward ratio shows how win rate and reward-to-risk combine; this article goes further, into what expectancy looks like in real accounts, how costs and frequency change it, and why small samples are so misleading.
How to calculate expectancy in R multiples
You can measure expectancy in money, but it is far more useful in R. One R is the amount you risk on a trade: the distance from entry to stop loss, turned into money by your position size. A trade that hits its stop is −1R. A trade that makes twice what you risked is +2R. Every result becomes an R multiple: profit or loss divided by the amount risked.
R strips out account size, pair and position size, so a 25-pip trade on GBP/USD and a 60-pip trade on USD/JPY can be compared fairly. It also separates two different questions: whether your method works (expectancy in R) and how much you bet on it (risk per trade in money). Step through one trade to see an R multiple being measured.
Measuring one trade in R on GBP/USD. The stop distance defines 1R; the planned target is +2R; the trade was closed early at +1.5R, and that real figure is what goes into the expectancy calculation. Illustrative prices.
Once results are in R, the expectancy formula has a simpler twin that gives exactly the same answer:
Expectancy (in R) = sum of all R multiples ÷ number of trades
That is just the average trade. The longer formula is still worth knowing because it shows why the average is what it is. The table compares methods with very different personalities.
| Win rate | Average win | Average loss | Expectancy per trade |
|---|---|---|---|
| 40% | 2R | 1R | (0.40 × 2) − (0.60 × 1) = +0.20R |
| 30% | 3R | 1R | (0.30 × 3) − (0.70 × 1) = +0.20R |
| 60% | 1R | 1R | (0.60 × 1) − (0.40 × 1) = +0.20R |
| 70% | 0.5R | 1R | (0.70 × 0.5) − (0.30 × 1) = +0.05R |
| 50% | 1R | 1.2R | (0.50 × 1) − (0.50 × 1.2) = −0.10R |
The first three rows have identical expectancy and very different lives. The 30% method spends long stretches losing; the 60% method feels comfortable most days. The fourth row wins seven times in ten yet barely clears zero, and the last row shows how a few losses that run past the stop, a common habit, turn an even method into a losing one.
To turn expectancy into money, multiply by what 1R is worth to you. At +0.2R and $50 risked per trade, the average trade is worth $10. Doubling your risk doubles that, and doubles every loss along the way, which is why the 1% rule and careful position sizing come before any thought about growth.
Expectancy after costs: the number that matters
Most expectancy examples quietly assume trading is free. It is not. You pay the spread on every trade, often a commission as well, sometimes swap for holding overnight, and stop orders can suffer slippage. These costs are paid on winners and losers alike, so they shrink every win and enlarge every loss.
There is a neat way to see their effect. Convert your average cost per trade into R by dividing it by your stop distance:
Cost in R = cost per trade in pips ÷ stop distance in pips
Because you pay that cost on every trade, it comes straight off the expectancy: net expectancy = gross expectancy − cost in R. A method with +0.2R before costs, paying 0.06R per trade, has +0.14R after them. The catch is in the division. The same cost in pips becomes a far bigger slice of R when the stop is small.
This is the main reason short-term methods are harder than they look. A scalping idea that tests well on a chart can lose most of its edge to costs, while the same idea on a wider stop barely notices them. Before you trust any expectancy figure, ask whether it was measured before or after costs.
Work out your own cost per trade below. The defaults are the trade from the chart above: 0.20 lots of GBP/USD, a 0.8-pip spread and a $7-per-lot round-turn commission, held intraday. The total is $3.00, which on $50 of risk is 0.06R per trade.
Spread = pips × $10 × lots. Commission and swap are quoted per lot. A trade must earn more than the total before it makes anything.
Two practical points follow. If you measure expectancy from your real account results, costs are already included, which is exactly what you want. If you measure it from a backtest or from chart study, subtract your cost in R from every trade, and add a little for slippage on stops. Our guide to forex trading costs and the Academy lesson on trading costs show how to find each component for your own account.
Why 20 trades prove nothing: sample size and variance
A method’s true expectancy is hidden. All you ever see is a sample of trades, and small samples are dominated by luck.
Take the +0.2R method from earlier: 40% winners at 2R. Over 20 trades it is expected to make +4R. But the number of winners in any 20-trade run varies a lot. The figure below shows how likely each outcome is, worked out from the probabilities of winning and losing.
A method with a real, positive edge finishes 20 trades in the red one time in four. In about 96% of runs, its 20-trade win rate lands somewhere between 20% and 60%. Now flip it round. A method with zero edge, 33.3% winners at 2R, shows a profit after 20 trades just over half the time, and matches or beats the +4R the genuine method is expected to make about one time in three. Twenty trades cannot tell those two methods apart.
More trades narrow the uncertainty, but slowly. The spread of possible results shrinks with the square root of the number of trades, so to halve the uncertainty you need four times as many trades.
Exact probabilities tell the same story. That +0.2R method is still behind after 100 trades about 9% of the time, and after 200 trades about 2.5% of the time. Fifty trades, the figure our trading journal guide uses for a first review, is enough to spot broken rules and obvious problems. It is not enough to prove an edge of this size, and a smaller edge needs even more trades.
You can feel this for yourself. The simulator below plays out 100 trades at 1% risk with a 40% win rate at 2R. The expectancy line under the chart never changes, but run it several times and watch how different the paths are: some climb steadily, others spend weeks below the starting balance.
Each run uses the same settings with different luck. Run it several times: the shape changes, the drawdowns don't go away.
Two lessons come out of this. First, do not abandon a method after a bad 20 trades, and do not double your risk after a good 20 trades; both are reactions to noise. Second, the bad runs are not a sign that something is broken. A losing streak of six, seven or eight trades is a normal part of a 40% method, which is why the Academy lesson on drawdowns sizes risk for the worst runs rather than the average one. Our guide to drawdown in trading goes deeper into what those runs feel like.
What would you do?
You have taken 20 trades with a new EUR/USD pullback method at 1% risk. You followed your rules every time and the journal shows +9R, an expectancy of +0.45R per trade. You feel you have found your edge. What do you do next?
Expectancy × frequency: what a method earns over time
Expectancy per trade is only half the picture. What you actually earn over a month is expectancy multiplied by the number of trades you take:
Average result per month = expectancy per trade × trades per month
A method at +0.4R taking 4 trades a month averages +1.6R. One at +0.15R taking 15 trades a month averages +2.25R. At 1% risk per trade, those are roughly 1.6% and 2.25% of the account per month on average, before compounding, and only if the edge is real and stays real. That last condition is where frequency becomes a trap.
The thinner the edge per trade, the more fragile it is. A +0.06R method has almost no room for a slightly wider spread, a little slippage or a few rushed entries; a +0.4R method absorbs the same shortfall comfortably. And in practice the shortfall is not the same for everyone: high-frequency methods use smaller stops, so the same costs are a larger fraction of R, as the cost figure showed.
There is also a behavioural version of the trap. The easiest way to take more trades is to lower your standards, and lower-quality trades usually have lower expectancy. Adding ten marginal trades a month at +0.02R each adds very little, while adding ten impulsive ones at −0.3R each wipes out a good month. More trades only help if every extra trade carries the same edge. Our guide to overtrading covers how that slide happens, and the Academy lesson on overtrading has practical limits that stop it.
A month is a tiny sample. Four trades a month at +0.4R means many months will be negative even if the method is sound. Judge the method on hundreds of trades; judge yourself each month on whether you followed the rules.
How to measure your expectancy from a trading journal
All of the above stays theory until you have your own numbers. The only reliable source of them is a trading journal in which every trade is recorded as an R multiple, in the same format, at the time. The method is simple:
- Record every trade in R. Divide the actual profit or loss, after costs, by the amount you planned to risk. A trade closed early for +38 pips on a 25-pip stop is +1.52R; a stop that slipped two pips is about −1.08R.
- Log all of them. Including the impulsive trades, the ones you are embarrassed about and the ones you “would not normally take”. Leaving them out flatters the number.
- Calculate the four statistics. Win rate, average win in R, average loss in R and expectancy. In a spreadsheet, expectancy is just the average of the R column.
- Split by setup and by plan-followed. A single overall figure can hide one setup that earns and another that loses, or a profitable method dragged down by broken rules.
- Note the sample size beside the result. “+0.15R over 50 trades” and “+0.15R over 300 trades” are very different findings.
Try logging a few results below. Enter some trades marked as following the plan with results of +2 or −1, then a couple marked as broken with results such as −1.5, and watch the average change.
Trade journal
Saved in this browser only. Over fifty trades, the last column shows what breaking the plan costs you.
Step 4 is where the journal earns its keep. The “result when plan broken” figure is often the biggest single leak in a retail account: a method with positive expectancy, executed with a few rule-breaking trades a month, can easily end up negative overall. The Academy lesson on the trading journal sets out a two-minute format and a weekly review, and your trading plan is where the rules being measured are written down. If you are starting from scratch, our guide on how to write a trading plan has a template.
The example keeps the same risk and the same rules after a good first 50 trades, and keeps measuring. Our guide to forex risk management covers the risk side.
Common trading expectancy mistakes
Judging a method by win rate
A high win rate feels like skill, but the table earlier showed a 70% method barely above zero and a 30% method comfortably positive. Win rate means nothing without the average win and average loss beside it.
Trusting a small sample
Twenty or thirty trades are mostly luck. A genuine +0.2R method loses over 20 trades one time in four, and a zero-edge method looks profitable half the time. Hold your conclusions until you have well over 100 trades.
Measuring before costs
Backtests, chart reviews and demo results often ignore spread, commission, slippage and swap. On short stops, costs can remove half or more of a small edge. Always quote expectancy after costs.
Leaving trades out of the journal
The trades you do not want to record are usually the ones that matter most. A journal that skips impulsive trades produces an expectancy for a trader who does not exist.
Changing the rules mid-sample
Each change to entries, stops or targets creates a new method with a new expectancy, and resets the sample. Change one thing at a time, and only after enough trades to judge it.
Regulators in the UK and EU require CFD providers to tell retail clients what share of their accounts lose money, and that figure is usually well over half (see the European Securities and Markets Authority for the background to these rules). Our guide to why most retail traders lose money explores the reasons; a negative or unmeasured expectancy, traded at too large a size, sits behind many of them.
Frequently asked questions
What is a good trading expectancy?
Any expectancy that stays reliably above zero after all costs, over a large sample, is a working edge. For most realistic methods it is a modest fraction of R per trade. Be sceptical of very high figures measured over a few dozen trades: they usually reflect luck, and they tend to shrink as the sample grows.
How do you calculate expectancy in trading?
Use (win rate × average win) − (loss rate × average loss). In R multiples, it is simply the sum of all your trade results divided by the number of trades. A method that wins 40% of the time, averages +2R on winners and −1R on losers has an expectancy of 0.8R − 0.6R = +0.2R per trade.
How many trades do you need to know your expectancy?
More than most people think. Twenty trades tell you almost nothing, and fifty give only a rough first read. For a method with an edge of around +0.2R, it takes roughly 200 trades before the plausible range stops reaching below zero. Smaller edges need even larger samples.
Can a strategy with a low win rate have positive expectancy?
Yes. A method that wins 30% of the time with average wins of 3R and losses of 1R has an expectancy of (0.30 × 3) − (0.70 × 1) = +0.2R. The cost is psychological: low win rate methods produce long losing streaks, so risk per trade must be small enough to survive them.
What is the difference between expectancy and profit factor?
Expectancy is the average result per trade, in R or money. Profit factor is total winnings divided by total losses. Both use the same trades: 37.8R of wins against 30.45R of losses gives a profit factor of about 1.24 and, over 50 trades, an expectancy of +0.147R. Expectancy is easier to combine with trade frequency.
Free course
Learn this step by step
“Risk & Money Management” is a free BrokerCut Academy course: 7 short lessons (about 59 minutes), quizzes, interactive tools and a certificate when you finish.
Enrol free
