Direct answer
Expectancy in forex is a way to express what the average outcome per trade would be, given a specific set of assumptions about trade results. It does not predict future trades by itself; instead, it turns a payoff model into an “expected value” using probabilities and the sizes of wins and losses. If your inputs (win rate, win/loss magnitude, and costs) are wrong or change over time, the expectancy estimate can change and stop matching reality.
Mechanism or definition: the core idea
A useful plain-language definition is: expectancy is the mean result you expect from repeating the same decision process many times, assuming the same probability of each outcome and the same payoff for each outcome.
In forex, “result” usually means profit or loss measured in your chosen unit (for example, account currency) for one completed trade. To compute expectancy, you typically model trades as having at least two outcome groups:
- Winning trades (profit)
- Losing trades (loss)
You then combine:
- The probability of winning (often called win rate)
- The probability of losing (often 1 minus win rate)
- The average profit when a win happens
- The average loss when a loss happens
- Any trading costs included in the profit/loss definition
A common structure is:
- Expectancy = (Probability of win × Average win) + (Probability of loss × Average loss)
Important: the “average win” and “average loss” must be defined in the same way you measure outcomes. If you measure profit after commissions and spread, then expectancy must use after-cost figures. If you measure profit before costs, then costs must be excluded consistently or added later. Otherwise, you model a different quantity than the one you observe.
Inputs and outputs: what you put in, what you get out
Inputs
Expectancy calculations require several inputs that can be stable within your model, but variable in practice:
- Outcome probabilities: The chance a trade ends as a win versus a loss. This depends on the strategy’s behavior and market conditions. Even if your strategy rules are constant, probabilities can shift.
- Payoff sizes: The average profit for winning trades and the average loss for losing trades. These depend on position sizing, stop/target assumptions, and execution.
- Cost model: Spread, commission, financing (if applicable), and slippage. In real trading, execution quality affects how closely realized results match your assumed payoff.
- Trade definition: What counts as one trade matters (entry/exit rules, partial closes, break-even handling, and whether you include weekends/rollover impacts).
- Measurement basis: Whether you compute results in pips, in account currency, or as returns. Expectancy is only meaningful relative to the measurement basis you choose.
Output
The output is an expected value per trade under your assumptions—often expressed as:
- Expected profit (or loss) per trade in money terms, or
- Expected return per trade (if you normalize by position size or account value)
A key limitation is that expectancy is only as good as the model inputs. Even if your formula is correct, using outdated or mismatched probabilities and payoffs can produce an estimate that does not reflect the next set of trades.
Evidence or example: a worked, checkable calculation
Below is a simplified example to show the sequence. This is an illustrative model, not a claim about any real forex system.
Assumptions for the example
Assume you measure each closed trade’s net profit after estimated costs, and classify outcomes as:
- Win: profit of +100 units on average
- Loss: loss of −60 units on average
- Win probability: 40%
- Loss probability: 60%
These numbers are the inputs. If your real data shows different averages or probabilities, the expectancy changes.
Calculation sequence
- Probability-weight the win payoff:
- 0.40 × 100 = 40
- Probability-weight the loss payoff:
- 0.60 × (−60) = −36
- Add them:
- Expectancy = 40 + (−36) = 4 units per trade (on average under these assumptions)
What the example means (and what it does not)
- It means: if the same win rate, average win, and average loss continued to apply, the average result per trade would be about 4 units.
- It does not mean: the next trade will profit, or that future results will match the same probabilities.
Because forex conditions change, your probabilities and payoff distribution can shift. Also, execution and costs can be misestimated, which changes the “net” numbers you used.
Limitations and failure modes: where expectancy breaks
1) Probability mismatch
If the true probability of winning differs from your estimated win rate, the expectancy estimate can be biased. Probabilities are not fixed constants; they can change with volatility regimes, liquidity, and how strictly trades follow the strategy rules.
2) Payoff distribution changes
Even if the win rate stays similar, the average win and average loss can change. For example, execution slippage can reduce profits faster than it increases losses, or spreads can widen during certain sessions.
3) Costs and execution not modeled consistently
Expectancy is sensitive to how you define net outcomes. A model that ignores spread and slippage will generally overstate performance compared to a model that includes them. This is a common failure mode because costs are often small per trade but matter when embedded in averages.
4) Measurement and trade counting errors
Expectancy depends on trade definitions. If you include different treatment for partial closes, withdrawals, fees, or rollovers, then your measured “win” and “loss” averages may not match your calculation method.
5) Hidden assumptions about “repeatability”
Expectancy is a statement about repeating the same process under the same conditions and assumptions. In practice, traders adapt, markets evolve, and strategies may behave differently as conditions change.
Material limitation to remember
Historical averages do not guarantee future expectancy. They only inform the inputs you might use for a model, and even then the informational value depends on whether conditions and implementation remain comparable.
Verification and the next question: how to check expectancy yourself
To independently verify expectancy, you need a consistent method for extracting inputs from your own recorded results:
- Use the same trade definition used in your expectancy model.
- Use outcomes that match your expectancy measurement basis (for example, net of costs).
- Compute empirical win rate and empirical average win/loss from the same dataset.
- Recalculate expectancy from those measured inputs.