A planned risk-reward ratio compares the amount a trade could gain at its target with the amount it could lose at its stop, for a stated position size. Expectancy combines the frequency and size of wins and losses with costs. One describes a plan; the other describes a set of assumed or observed outcomes. Neither tells you what the next trade will do.

Start with the arithmetic

For a simplified set of trades with only wins and losses, net expectancy per trade is:

Expectancy = (win probability × average win) − (loss probability × average loss) − average cost.

Consider a fictional sample with a 40% win rate, an $80 average win, a $50 average loss and $5 average round-trip cost. Its weighted gains are 0.40 × $80 = $32; weighted losses are 0.60 × $50 = $30. After cost, the sample expectancy is $32 − $30 − $5 = −$3 per trade. This is an arithmetic illustration, not a measured crypto strategy.

The same assumptions imply a break-even win rate before any account-level compounding. Let p be the win rate. Then p × $80 − (1 − p) × $50 − $5 = 0, so p = $55 ÷ $130 ≈ 42.3%. At a 40% win rate, expectancy is −$3; at 60%, it is +$23, provided the average win, average loss and cost really stay fixed. The graph shows only that one-variable sensitivity.

fixed-assumption expectancy and break-even win rateUnder fictional fixed averages of an 80 dollar win, 50 dollar loss and 5 dollar cost per trade, net expectancy rises from minus 29 dollars at a 20 percent win rate to plus 23 dollars at 60 percent; the arithmetic break-even is about 42.3 percent.SAME AVERAGE WIN, LOSS AND COST · ONLY WIN RATE CHANGESExpectancy = p × $80 − (1 − p) × $50 − $5 · fictional dollars per trade+$20$0−$2020% · −$2940% · −$3break-even ≈ 42.3%60% · +$2320%40%60% win rateSensitivity of one assumed payoff model · not strategy or market performance
Under fictional fixed averages of an 80 dollar win, 50 dollar loss and 5 dollar cost per trade, net expectancy rises from minus 29 dollars at a 20 percent win rate to plus 23 dollars at 60 percent; the arithmetic break-even is about 42.3 percent. Illustrative teaching graphic · not live market data

Change the payout profile and the threshold moves. A second fictional sample wins 60% of the time, averages $20 per win, loses $50 per loss and pays $2 average cost. Its result is 0.60 × $20 − 0.40 × $50 − $2 = −$10 per trade. Its break-even rate is ($50 + $2) ÷ ($20 + $50) ≈ 74.3%. A high win percentage can coexist with a negative average when losses are larger than wins.

two fictional samples where weighted losses and costs exceed weighted gainsSample A has a 40 percent win rate, an 80 dollar average win, a 50 dollar average loss and 5 dollars cost: weighted gains are 32 dollars, losses plus cost are 35 dollars, net expectancy is minus 3 dollars. Sample B has a 60 percent win rate, a 20 dollar average win, a 50 dollar average loss and 2 dollars cost: weighted gains are 12 dollars, losses plus cost are 22 dollars, net expectancy is minus 10 dollars. Fictional arithmetic only; it does not estimate strategy performance.A HIGHER WIN RATE DOES NOT GUARANTEE POSITIVE EXPECTANCYSAMPLE A · 40% WINSSAMPLE B · 60% WINSweighted gains$32losses + cost$35net expectancy−$3 / tradeweighted gains$12losses + cost$22net expectancy−$10 / tradeWeighted amounts include each fictional sample's stated win, loss and cost assumptionsIllustrative arithmetic only · no market result or performance estimate
Sample A has a 40 percent win rate, an 80 dollar average win, a 50 dollar average loss and 5 dollars cost: weighted gains are 32 dollars, losses plus cost are 35 dollars, net expectancy is minus 3 dollars. Sample B has a 60 percent win rate, a 20 dollar average win, a 50 dollar average loss and 2 dollars cost: weighted gains are 12 dollars, losses plus cost are 22 dollars, net expectancy is minus 10 dollars. Fictional arithmetic only; it does not estimate strategy performance. Illustrative teaching graphic · not live market data

Planned risk differs from realized loss

A planned 2:1 reward-to-risk ratio might compare a $50 target with a $25 stop loss for a chosen quantity. It is a comparison of planned amounts, not proof that the target will fill twice as often or that every stopped trade loses exactly $25. A gap, partial fill, spread, slippage, fee or funding payment can change realized outcomes. For a perpetual, use the venue’s actual contract and settlement rules rather than treating a chart distance as a complete cash result.

Here is one way the plan can change in a fictional one-unit spot example. Buy at $100, set a target at $150, and set a stop trigger at $75: the price distances suggest $50 of reward for $25 of planned risk, or 2:1 before costs. Suppose a stop-limit order triggers at $75 and fills at its $70 limit, and the round trip costs $3. The realized loss is then $30 price loss plus $3 cost, or $33. If the target fills at $150, the realized gain is $50 less $3, or $47. Under only these fixed outcomes, the effective net reward-to-risk is $47 ÷ $33 ≈ 1.42:1 and the break-even win rate is $33 ÷ ($47 + $33) = 41.25%. All prices, quantity and fees here are made up to show the arithmetic; they are not a quote, recommended order, or estimate of a likely fill.

The order path matters. Coinbase’s current Advanced Trade documentation describes a stop-limit order as placing a limit order after the stop price is reached; it also says downside protection on its bracket and TP/SL orders may fail to execute during extreme volatility. Those are product-specific mechanics and can change. Check the rules for the exact venue, market and order type you use. A trigger is not itself proof of a completed exit.

planned stop, order execution and realized expectancyFictional one-unit spot illustration: a 100 dollar entry, 150 dollar target and 75 dollar stop trigger imply a 2 to 1 planned price-distance ratio. In a separately assumed 70 dollar stop-limit fill with 3 dollars round-trip cost, the net win is 47 dollars, net loss 33 dollars and break-even win rate 41.25 percent. No market data or likely-fill claim is shown.FROM PLANNED LEVELS TO A NET RESULTFICTIONAL ONE-UNIT SPOT EXAMPLE · NOT A MARKET OBSERVATION1 · PlanEntry $100 · target $150Stop trigger $75$50 ÷ $25 = 2:1before costs and fills2 · Suppose it fillsTrigger $75 · limit fill $70Round-trip cost $3Net loss = $33$30 price loss + $3 cost3 · RecalculateNet win $47 · net loss $33$47 ÷ $33 = 1.42:1Break-even 41.25%under these fixed outcomesIllustrative arithmetic only · a triggered order may remain unfilled · venue rules vary
Fictional one-unit spot illustration: a 100 dollar entry, 150 dollar target and 75 dollar stop trigger imply a 2 to 1 planned price-distance ratio. In a separately assumed 70 dollar stop-limit fill with 3 dollars round-trip cost, the net win is 47 dollars, net loss 33 dollars and break-even win rate 41.25 percent. No market data or likely-fill claim is shown. Illustrative teaching graphic · not live market data

When checking a recorded sample, calculate each trade’s realized net result on one consistent basis. Include the same cost categories for wins and losses; do not subtract a cost twice if it is already reflected in the ledger. Keep unfilled and no-trade decisions visible when they are part of the test rule. If the sample uses a fixed dollar risk, convert outcomes to R (the result divided by the amount planned to be lost at the stop) only after stating how gaps and execution costs are handled.

Averages hide important paths

An average can be pulled upward by one unusually large winner even when most trades lose. A small sample can give a very different result from the longer process that produced it, and neighboring trades may share market conditions. Pair the average with trade count, the distribution of net outcomes, drawdown, losing sequences, largest loss, exposure and results across chronological periods. A win-rate estimate is also imprecise when the sample is small; do not present a rounded percentage as a stable property of a strategy.

Do not keep changing the entry, stop, target and cost model until one historical average turns positive, then label that selected result independent evidence. Record the full search, reserve later chronological data for evaluation and name the baseline before comparing results. Repeated parameter choices can fit noise.

Crypto adds market and instrument risks outside this arithmetic. The CFTC warns that virtual-currency price swings can be volatile and that leverage in futures can amplify gains and losses. Its advisory covers the risks of virtual-currency spot and futures/options markets; the exact protections and contract mechanics depend on the product and jurisdiction. A stop distance does not cap every realized loss, and a positive historical expectancy does not remove those risks.

SuperQuant’s expectancy calculator applies the simple formula to values you enter in your browser. It does not estimate uncertainty, verify your trade records or forecast an account result; its example inputs are fictional. Continue with position sizing, the backtest metrics guide and the trading journal to keep planned risk distinct from realized results. This guide is educational, not individualized financial advice or a performance promise.