Trading expectancy is one of the most useful numbers available for judging whether a strategy rests on solid foundations. It does not measure the outcome of any single trade; it measures the average value you can expect per trade over time.
This article covers what expectancy actually is, why it matters more than the metric most traders reach for first, the three ways to calculate it, and the specific adjustments that improve it without destabilising the rest of the system.
What Trading Expectancy Is and Why It Matters
Expectancy tells you how much a strategy generates, on average, per trade executed. It is a statistical value, so it says nothing about what will happen on the next trade. What it does tell you is whether the system has a genuine edge over a large sample — and how large that edge is.
Many traders evaluate a strategy by looking only at the percentage of winning trades, the win rate. This is a common error, because that figure tells you a fraction of the story. A strategy can win frequently and still lose money overall, if the average loss is larger than the average win. Equally, a strategy can have an unimpressive success rate and be reliably profitable, provided the winners are considerably larger than the losers.
Expectancy exists to combine those elements into a single number: probability of winning, probability of losing, average win and average loss. That allows you to see whether, across a broad sequence of trades, the strategy produces value or consumes capital.
The underlying point is that trading cannot be judged on a handful of trades. A good strategy can go through a bad run. A weak strategy can produce a temporary good patch. Expectancy reduces the weight of random impressions and forces the discussion back onto data.
Calculating it matters for one more reason: it helps you avoid impulsive decisions. If a strategy has positive expectancy, a losing period can be assessed with far more composure, because it has a context to sit in. If expectancy is negative, the calculation tells you that persevering with the same approach is simply not worth doing — which is information most traders acquire much more expensively.
Method 1: The Standard Expectancy Formula
The most widely used calculation is straightforward:
Expectancy = (Win rate × Average win) − (Loss rate × Average loss)
Take an example. A strategy wins 50% of its trades and loses 50%. The average win on the profitable trades is $120, and the average loss on the losing trades is $80. The calculation is:
(0.50 × 120) − (0.50 × 80) = 60 − 40 = 20
Expectancy here is positive at $20 per trade. That means each trade has, on average, produced $20 of expected value. It does not mean every trade will make $20. It means that across a large sample, and assuming statistical conditions stay broadly similar, $20 is the average you should expect.
Notice what this reveals about win rate. A 50% win rate looks unremarkable on its own, but combined with a 1.5:1 ratio of average win to average loss it produces a clearly positive number. Change the average loss to $130 and the same 50% win rate turns the system negative. The win rate alone was never the variable that mattered.
Method 2: Expectancy in R Multiples
A second approach is to calculate expectancy in multiples of risk, expressed as R. If the initial risk on every trade is 1R, then trade outcomes can be written as +2R, −1R, +1.5R and so on. This is useful because it makes the calculation independent of the monetary size of your positions.
Suppose a strategy wins 40% of its trades with an average win of 2R, and loses 60% of its trades with an average loss of 1R:
(0.40 × 2) − (0.60 × 1) = 0.80 − 0.60 = 0.20R
Expectancy is 0.20R per trade. If your risk per trade is $100, the expected value is $20 per trade. If your risk is $200, it is $40. The advantage of the R framework is that it lets you assess the quality of the strategy itself, without confusing it with leverage or account size — which is exactly what you want when comparing two systems, or the same system before and after a change. Our guide to position sizing covers how to define 1R consistently in the first place.
Method 3: Working Backwards From Your Trade History
The third method uses your actual results. Here you simply sum the outcome of every trade and divide by the number of trades. If 100 trades produced a combined result of $2,500, average expectancy is $25 per trade.
It is the fastest calculation available, and it needs to be interpreted carefully: a sample that is too small can be badly misleading. Twenty trades will tell you almost nothing, because a single outlier dominates the average. To get a figure worth acting on, analyse a meaningful number of trades and separate them by market condition — a strategy with an expectancy of +0.3R in trending conditions and −0.2R in ranging conditions has an aggregate number that hides the only insight that matters.
It is also worth remembering that a historical expectancy calculated on an over-optimised backtest inherits all of that optimisation. If the numbers came from a system whose parameters were tuned until the equity curve looked good, expectancy will look good too. Our article on how to spot a fragile trading strategy covers the warning signs, and Monte Carlo testing is one way to check whether the figure survives resampling.
How to Improve Expectancy
Improving expectancy means working on the components that make up expected value: entry quality, exit management, the relationship between wins and losses, and the selection of operating conditions. Some adjustments make a substantial difference without changing the strategy at all.
- Reduce the average loss. One of the most direct routes to better expectancy is controlling the average losing trade. This does not mean placing arbitrarily tight stops — a stop too close to entry will eject you from the market repeatedly for no technical reason, lowering the win rate faster than it lowers the average loss. It means defining in advance the point at which the trade thesis is no longer valid, and placing the stop there. Smaller losses, when they are consistent with the strategy, improve the overall risk-to-reward relationship.
- Improve the quality of the trades you take. Not every signal carries the same value. A strategy may produce a large number of entries while only a subset of them occur under genuinely favourable conditions. Filtering trades by context, volatility, liquidity, trend or technical level can reduce the number of trades while improving the average result. This is not manipulating the calculation; it is narrowing the sample to the conditions where the edge actually exists.
- Keep a precise trading journal. Without reliable data, expectancy is a guess. A journal lets you record entry, exit, initial risk, result in currency, result in R, market context and any execution errors. Over time that record shows you where the strategy works and where it loses efficiency — and it distinguishes a strategy in a drawdown from a strategy being undermined by poor execution, which is a distinction discussed further in our piece on why a trading system stops working.
One Number, Honestly Calculated
Expectancy is not a forecast and it is not a guarantee. It is a summary of whether the mathematics of your trading is working in your favour, and by how much. Its real value is that it is difficult to argue with: a negative expectancy calculated over a decent sample is not a run of bad luck to be traded through, and a small positive expectancy is not a reason to increase size aggressively.
Calculate it in R rather than currency, insist on a sample large enough to mean something, and separate the figure by market condition. What comes out is the closest thing to an objective answer to the only question that matters: is this method worth continuing to trade?
