Trade Expectancy Calculator: What Is One Trade Actually Worth?

You win 45% of your trades. Your winners average 2R, your losers average 1R. Good? Bad? You genuinely cannot tell by looking at it, which is annoying, because those are the three numbers everybody quotes at each other.

So let's just run it.

The winners bring in 45% × 2R, which is 0.90R. The losers take out 55% × 1R, which is 0.55R. Subtract, and you get +0.35R per trade.

That's expectancy. It's what one trade has been worth on average, once you fold in how often you win and what the wins and losses are actually sized at.

What the number actually is

The formula is:

(Win Rate × Average Win) − (Loss Rate × Average Loss)

Loss rate is just 100% minus your win rate. Win 45%, lose 55%. There's no third bucket, a trade is a winner or it isn't.

And R, if you haven't run into it, is just the amount you risked on the trade. If you risk $100 and make $200, that's a 2R winner. If you risk $100 and lose it, that's -1R. Using R instead of dollars means you can compare your results from when you had a $500 account against your results now, without the account size distorting everything. Same strategy, same R, different dollar amounts.

Try the Trade Expectancy Calculator

Punch in your win rate, your average winner and your average loser and the trade expectancy calculator does the arithmetic. Dollars or R, either is fine, just don't mix them. Don't put a win rate in as 45 in one box and 0.55 in another, and don't put winners in R and losers in dollars. It'll give you a number and the number will be garbage.

Positive, negative, and the awkward middle

Above zero means the combination of your wins and losses has been making money. +0.25R means every trade has been worth a quarter of what you risked on it, on average. If 1R is $100 for you, that's about $25 a trade.

Below zero means it's been losing money. Say you win 40%, your winners are 1R and your losers are 1R. That's 0.40R coming in and 0.60R going out. Negative 0.20R. Taking more of those trades does not fix it, it just gives the math more chances to do what it's already doing.

And zero is 50% win rate with 1R winners and 1R losers. Break even, mathematically, before you've paid a single spread. After spread and commission and the slippage you didn't notice, it's negative. Break even isn't neutral, it's a slow leak.

Why "what's a good win rate" is a question nobody can answer

Someone will tell you that you need a 60% win rate. Ignore them. A 60% win rate can be brilliant or catastrophic and the number itself tells you nothing.

Look at these two:

Strategy A wins 70% of the time. Winners average 0.5R, losers average 1.5R. Expectancy: -0.10R.

Strategy B wins 40% of the time. Winners average 2.5R, losers average 1R. Expectancy: +0.40R.

Strategy A wins way more often and loses money. Strategy B is wrong most of the time and makes money. So the useful question isn't "is my win rate high enough," it's "does the relationship between how often I win and how big my wins are come out positive." That's a question with an actual answer.

You can also get to positive expectancy from completely different directions. With a 1R average loser, a 60% win rate and 1R winners gives you +0.20R. A 50% win rate with 1.5R winners gives you +0.25R. A 40% win rate with 2R winners gives you +0.20R. Three strategies that would feel wildly different to trade, all fine.

Use what you actually did, not what you meant to do

This is the part that stings, so let's get it over with.

Your plan says 2R targets, 1R stops. You win 45%. Plug those in and you get +0.35R, and you feel pretty good about yourself.

Then you open your journal and your average winner isn't 2R. It's 1.2R, because you keep closing early when the trade goes green and your stomach starts doing the thing. Run it again: 0.45 × 1.2 is 0.54, minus 0.55, equals -0.01R.

Same setups. Same stops. Same market. The strategy you designed has a real edge and the strategy you're actually trading is break-even-to-slightly-worse. Those are two different strategies and only one of them is real.

It's worth calculating both, honestly. Planned expectancy tells you whether the model you built is any good. Realized expectancy tells you what you did with it. When they don't match, the gap tells you where to look. Win rate dropped? You're taking setups that weren't in the tested method. Average winner shrank? Early exits, moving targets, taking partials you didn't plan on. Average loser grew? You're moving stops or holding past the point where the trade was clearly wrong. Each one points at a different fix, and only one of them is "find a new strategy," which is the one everybody reaches for first.

Partials and exits change the math, sometimes in your favour

Say your target is 3R but you scale out along the way, so your average winner ends up at 1.6R. Your win rate is 55% because you're banking something on more trades. Losers average 1R.

(0.55 × 1.6) − (0.45 × 1) = +0.43R

Now hold the whole position to 3R instead. Win rate drops to 30% because price doesn't get all the way there very often.

(0.30 × 3) − (0.70 × 1) = +0.20R

The version with the smaller winners made more per trade. Not always, but in this case, yes. Bigger targets are not automatically better, because the target only pays if price reaches it. Move from a 2R target to 5R and if your win rate falls from 50% to 15%, you've gone from +0.50R to -0.10R. There's no prize for having the biggest number written on your trade plan.

Cutting losers early cuts both ways too. If your planned loss is 1R and your journal says your average loser is 0.7R because you get out when the setup breaks, that's expectancy improving. If your average loser is 1.4R because you keep giving it room, well. The calculator just found your problem.

It cannot tell you what happens on the next trade

Expectancy of +0.35R does not mean the next trade makes 0.35R. Your next trade can be -1R. So can the one after that, and the one after that.

Think of it like a casino. The house doesn't know whether the woman at the blackjack table is about to clean them out tonight. They know that across a hundred thousand hands the math tips their way, and they're happy to lose the individual hand. You're running the same arrangement, except you're the house and every trade is a hand.

Which also means positive expectancy comes bundled with losing streaks. Win rate 30%, winners 4R, losers 1R gives you +0.50R, which is strong. It also means you're losing seven trades out of ten. If you bail after four losses in a row you'll never be in the market for the winner that makes the whole thing work, and the +0.50R stays theoretical forever.

That's why two strategies with identical expectancy can be completely different to live with. One wins often with small winners and occasional bigger losses, the other loses constantly and gets rescued by rare big winners. Same number on the page, wildly different Tuesday afternoons. A strategy can be mathematically good and psychologically terrible for you specifically, and it's fine to trade the one you can actually stick to.

Small samples will lie to your face

Five trades, four winners, +1.2R expectancy. Congratulations, you know nothing.

One of those was probably a +8R fluke and it's dragging the entire average around by itself. Take the outlier out and the method might be negative. That doesn't automatically mean you should delete the outlier, if occasional monsters are a genuine part of how the strategy works then they belong in the data. But you should know how dependent the whole thing is on them.

Ten trades is interesting. Twenty or thirty and you're starting to see shape. Fifty and up is worth taking seriously, a hundred plus and you've got something to study. There's no magic threshold where the numbers become true, more data just means fewer flukes per average.

Same warning for filters. Add a filter to a 300-trade sample and go from +0.20R to +0.35R on 180 trades, that's probably real. Add another filter that leaves you 25 trades at +0.90R and you haven't found an edge, you've found 25 trades. That's curve-fitting, which is when you tune a strategy so tightly to past data that it only works on past data.

Slice it up

This is where the calculator turns into something genuinely useful instead of a party trick.

Once you've got a decent sample, run expectancy separately by setup. Setup A over 200 trades might be +0.42R, Setup B over 175 might be +0.18R, and Setup C over 160 might be -0.07R. You didn't have to guess which one was bleeding, it's right there. Now go figure out whether Setup C needs a filter, needs different management, or needs to be dropped.

Then do it by pair. EUR/USD +0.40R, GBP/USD +0.25R, USD/JPY -0.05R. Then by session. London +0.45R, New York +0.20R, Asia -0.10R. Then by day of the week if you've got the trades for it. None of this is a verdict, it's a place to look. "I think I trade London better" becomes "my London trades have produced more per trade over 140 trades," which you can actually act on.

The one that really gets people is splitting trades into the ones where you followed your rules and the ones where you didn't. Rule-following trades: +0.40R. Rule-breaking trades: -0.35R. Awkward. But then you don't need a better strategy, you need to trade the one you already have. That's a much cheaper fix than another six months of searching for a system.

Costs, and why a thin edge isn't really an edge

If your journal numbers already have spread and commission baked in, fine. If they don't, your real expectancy is lower than what you calculated.

Gross expectancy of +0.10R can land at +0.02R after spread, commission, slippage and swap. That is technically positive and practically nothing. Any small change in your execution wipes it out. This matters more the more trades you take and the smaller your targets are, because you're paying the toll more often on smaller distances.

On the flip side, a small edge that survives costs is worth having. +0.15R doesn't sound like much, but over 200 trades the simplified total is 30R. If 1R is $100, that's $3,000. It won't arrive in a straight line and some of those months will be ugly, but it adds up in a way that a single fantastic trade doesn't.

Risking more doesn't improve your edge

Worth saying plainly because I see this one constantly. Your method has +0.30R expectancy and you decide to double your risk. Your expectancy is still +0.30R. It didn't move. Every R is just worth twice as much money now, wins and losses both.

Expectancy in R measures the strategy. Your risk percentage measures how hard the strategy's results hit your account. If you risk 1% per trade and 1R is 1% of the account, +0.40R expectancy is roughly +0.40% per trade. Risk half a percent and it's +0.20%. The edge didn't change, the leverage on it did. Keep those two decisions separate in your head, because one is about whether the method works and the other is about whether you survive the drawdown while it does.

What to do with it

Go to your journal and get three numbers: how often you won, what your winners averaged, what your losers averaged. Not your targets, not your plan, what actually happened. Put them in the calculator.

If the number's positive, your job is to keep doing the same thing enough times for it to show up. If it's negative, at least you know now, and you know which of the three inputs to go after. And if you don't have enough trades to calculate it, that's the real answer, and the fix is to go log some trades.

Then run it again in a few months on your last hundred. If lifetime is +0.35R and the last hundred is +0.12R and the last fifty is -0.08R, something's changed. Might be the market, might be you. But you'll see it in the numbers a lot sooner than you'll feel it in your account.

Educational purposes only. Forex trading involves risk. Trade expectancy is a historical statistical measure based on the data entered and does not predict or guarantee future performance. Results may be affected by sample size, spreads, commissions, slippage, execution, and changing market conditions.