Risk-reward ratio and expectancy in crypto trading

How to calculate a risk-to-reward ratio

For a planned long trade, price risk per unit is entry minus stop; potential price reward per unit is target minus entry. For a short trade, reverse the directions. Divide potential reward by planned risk to get a reward multiple. Some traders write the same relationship as risk:reward, such as 1:2; others call the reward 2R. State the convention because the ratio alone can be ambiguous.

Consider a hypothetical long entry at $100, stop at $96 and target at $108. Before costs, the stop is $4 away and the target is $8 away. That is $4 planned price risk for $8 potential price reward, or risk:reward 1:2 (a +2R target against a −1R stop). It is a comparison of distances between chosen prices. It gives no probability that price will reach either one.

A stop should correspond to the condition that would invalidate the trade idea. Moving the stop closer merely to improve the displayed ratio may contradict the setup’s invalidation rule. Moving it farther away increases the planned loss for the same position size and reduces reward relative to risk. Moving the target farther away increases the stated potential reward but does not establish that the target is realistic.

A risk-reward ratio is not expectancy

Expectancy is the average result per trade implied by the size of wins and losses and how often each occurs. If p is the win fraction, W is average net win and L is the average net loss written as a positive amount, then:

Expected result per trade = p × W − (1 − p) × L

If all results are divided by the same initial risk amount, the result is expectancy in R. For a simple two-outcome model with no costs, a payoff of +2R on wins and −1R on losses breaks even when p = 1 / (2 + 1) = 33.33%. At a 1:3 risk:reward ratio, that arithmetic threshold is 25%. Those are break-even calculations under the stated payoffs, not estimates of a strategy's real win rate.

Real trades can have partial exits, variable winners and losses, funding, fees and gaps. In that case, use the observed average net win and loss from a consistently defined sample rather than substituting the advertised target multiple for average results. Include losing trades, skipped signals under the written rules, and the order in which a stop and target were reached.

A planned 1:2 ratio is not a realized 2R result

A displayed ratio uses planned prices; realized R comes from the fills and costs that actually occurred. Suppose a hypothetical long entry is $100, the stop is $96, the target is $108 and the position is 10 units. The plan shows $40 of gross risk and $80 of gross reward, or +2R if all 10 units reach the target. If 5 units instead exit at $104 and the remaining 5 hit the $96 stop, the first part earns $20 and the second loses $20: realized gross P&L is $0, or 0R against the original $40 risk, before fees and funding. Record each partial fill and calculate the final result using one stated R denominator and actual costs; do not substitute the target multiple for the trade result.

Explore position size and expectancy assumptions

Use the calculator to see how entry, stop, target, account size, risk budget, fees, leverage and a user-supplied win-rate assumption affect the arithmetic. The result depends on those inputs.

In this calculator, 1R is the modeled net loss at the stop, including the entered fees and flat funding assumption. The separate figure below uses a different definition: R is the planned price loss before its assumed 0.10R cost. Compare results expressed in R only when they use the same definition.

A calculator cannot validate the win rate, predict a fill, model every contract type or decide whether a trade is suitable. Treat expectancy as a scenario until a defined method has been evaluated with relevant trade data and costs.

Set position size from the stop and a chosen loss budget

First define the price level or event that invalidates the idea. Then choose a loss budget for that position and calculate quantity from the distance to the stop. A wider stop means smaller quantity for the same budget; a tighter stop means larger quantity. The stop location comes from the setup's rule, while size adjusts to the budget.

A simplified spot or linear-contract estimate is:

quantity ≈ cash loss budget ÷ modeled loss per unit at the stop

Modeled loss per unit should include the assumed entry and exit fills plus applicable trading costs. Check contract multiplier, quantity increment, minimum order size, quote currency, margin and product-specific P&L rules before using a result. Inverse contracts, options and products with settlement in another asset need their own payoff calculation.

For example, an account of $5,000 and an explicitly chosen 1% loss budget give $50. The 1% is an example input, not a universal recommendation. Three separate positions each planned to risk $50 would represent $150, or 3% of that account, if all three losses reached their modeled exits. Correlated exposure can make simultaneous positions matter even when each trade is sized separately.

Use ATR as a volatility reference, not a stop rule

Average True Range summarizes recent true ranges on a selected market and timeframe. True Range includes the current high-low span and gaps from the previous close. ATR has no directional sign: it describes recent movement, not where price should go next or how much it can move in the future. Its period and smoothing method affect the value.

In the $100 example, suppose the chart's ATR is $2 and a hypothetical strategy rule places a long stop two ATR below entry: $100 − (2 × $2) = $96. This simply illustrates one chosen rule. It does not make two ATR a standard or optimal stop. A structure-based invalidation may sit elsewhere, and changing the stop changes quantity and the potential ratio. The crypto stop-loss guide covers those placement and order-type choices.

Include fees and execution assumptions

The displayed stop and target are not guaranteed fill prices. A stop-market may execute beyond its trigger; a stop-limit can remain unfilled. A venue may trigger from last, index or mark price, which can differ from the candle series you are viewing. For example, Bybit documents a market-order slippage protection range and conditional triggers with configurable price references for supported derivatives; Kraken documents stop and stop-limit order types with trigger choices. Check the current rules for the actual venue, product and order.

Here is a hypothetical long sizing calculation using the $100 entry and $96 stop. It assumes an ATR of $2 on the selected chart interval, a 2×ATR stop rule, a $5,000 account, a 1% ($50) loss budget, 0.1% fees on each side, 0.1% adverse slippage on each fill and a 0.001 quantity step. These rates and fills are example inputs only, not typical costs or a promise of execution.

At that size, the modeled position value at entry (notional) is $1,139.5384 and estimated round-trip fees are $2.231310. Actual orders may have different fees, fills, funding and contract rules. The example excludes funding, gaps, partial execution, liquidation and minimum-notional constraints; it is not a maximum-loss guarantee. The stop-loss guide discusses these limitations in more detail.

See how costs change a hypothetical expectancy

This teaching model assumes every win is +2R gross, every loss is −1R gross, 40% of trades win, and every trade costs 0.10R. Here, R means the gross planned price-loss unit before that assumed cost. The figures are chosen to show the calculation; they are not observations from ChartScout, a crypto market sample, or a chart-pattern study.

Before costs, expectancy is 0.40 × 2R − 0.60 × 1R = +0.20R. Subtracting 0.10R from every result makes the net outcomes +1.90R and −1.10R. Net expectancy is 0.40 × 1.90R − 0.60 × 1.10R = +0.10R. Gross break-even is 33.33%; after this flat cost, it is 1.10 ÷ (1.90 + 1.10) = 36.67%.

A fixed cost in R is a simplifying assumption. Real costs vary with notional, venue, maker/taker status, spread, fill, funding and holding time. For a real system, calculate net trade results using its actual rule set and a clearly stated R denominator.

Check trigger prices, margin and contract details

A chart can display last-trade candles while a stop uses a mark or index trigger. A price difference does not by itself show that an order malfunctioned. Read the venue's trigger definition and set the chart series to match where possible. Conditional-order behavior also differs: an order can become active only after its trigger, and the venue's margin checks may still cancel it.

Leverage changes the collateral and liquidation context; it does not turn a planned stop into a guaranteed exit or make a larger price move safer. On a linear product, changing leverage does not change the price P&L for the same contract quantity, though the initial margin requirement changes. Inverse contracts and other products have different P&L mechanics. A liquidation engine can close a position before a stop fills depending on margin, trigger reference, fees and market conditions. The examples here do not calculate exchange liquidation prices.

Evaluate a defined method instead of borrowing pattern odds

To estimate expectancy, define the market, venue, timeframe, setup qualification, entry, invalidation, target or management rule, position size and observation horizon before reviewing outcomes. Record all qualifying trades, including failures and skipped entries, with the prices and fees that the method could actually use. Account for funding where relevant, partial fills, gaps and whether stop or target was reached first. If candle data cannot establish the sequence, mark that trade as ambiguous or use data granular enough to resolve it.

Develop the rule on one sample, then evaluate it on later data or a separate sample that was not used to design it, and inspect performance across different conditions. Report the number of observations and uncertainty rather than turning a small sample into a precise probability. A measured-move target is a possible price objective under a rule; it is not a hit rate. Historical stock-pattern statistics do not supply a win probability for a crypto instrument or a different entry and exit method.

Compare net average win, net average loss, win fraction and expectancy together. A higher target multiple can coexist with fewer wins; a tighter stop can increase stop-outs; a lower-cost venue assumption can be wrong for the order type actually used. No single risk/reward ratio is “good” without the rest of the method and its measured outcomes.

Related guides

Continue with crypto stop-loss placement and sizing, how to spot fake breakouts, chart patterns and volume analysis, and multi-timeframe analysis. Each guide separates a chart observation from an execution or probability claim.

Frequently asked questions