Trade Expectancy Calculator

Trade expectancy is the average dollar amount a strategy makes or loses per trade after accounting for win rate, win size, loss size and per-trade costs. The formula is Expectancy = (Win % × Avg Win) - (Loss % × Avg Loss) - Avg Cost. A worked example: a strategy with a 55% win rate, average win of $200, average loss of $100 and average cost of $5 has expectancy of (0.55 × $200) – (0.45 × $100) – $5 = +$60 per trade. Over 100 trades, that is +$6,000 in projected P&L. The calculator above runs this maths instantly and adds profit factor, R-multiple, breakeven win rate, Kelly Criterion sizing and a 1,000-path Monte Carlo on top.

The honest answer depends on trade frequency and account size, but the Strategy Health verdict in the calculator above uses these bands. Excellent: expectancy above 15% of average loss, profit factor above 2.0, R-multiple at or above 1.5, costs under 10% of gross profit. Good: expectancy 5% to 15% of average loss, profit factor 1.5 to 2.0, costs under 20%. Marginal: expectancy 0 to 5%, profit factor 1.2 to 1.5. Risky: profit factor 1.0 to 1.2, no real edge. Avoid: negative expectancy or profit factor below 1.0. For a $100 average loss, that means anything above +$15 per trade is excellent, +$5 to +$15 is good, 0 to +$5 is marginal, anything negative is uninvestable.

The Kelly Criterion calculates the fraction of capital to risk on each trade that maximises long-run geometric growth, given a known edge. The formula in trading terms is Kelly % = Win % - ((1 - Win %) / R-multiple). For a 55% win rate strategy with a 1.5R reward-to-risk, Kelly returns 55% – (45% / 1.5) = 25% of capital at risk per trade, which is Full Kelly. The calculator above also surfaces Half Kelly = 12.5% and Quarter Kelly = 6.25%. Full Kelly is the mathematical optimum but produces drawdowns large enough that almost no human can stomach them in live trading. Half Kelly is the professional standard and captures roughly 75% of the growth at 25% of the variance.

Full Kelly assumes you know your true win rate and reward-to-risk exactly. In reality, every backtest sample is finite and the true edge could be 5 to 10 percentage points lower than what you measured. If you size at Full Kelly and the true edge is lower, you over-bet and lose a much larger fraction of capital in every losing run. Half Kelly is the standard hedge: it cuts the bet size in half but only loses about 25% of the long-run growth rate. The variance and the maximum drawdown both drop sharply. Concrete example: at Full Kelly 25% per trade, a 6-trade losing streak (perfectly normal at 55% win rate) compounds a 78% drawdown. The same streak at Half Kelly 12.5% produces a 55% drawdown. At Quarter Kelly 6.25%, only 32%. Most live traders quit at 30% drawdown, so the smaller Kelly fraction is the only one that survives reality.

The Quick Simulation panel runs 1,000 Monte Carlo paths against your strategy parameters using a seeded mulberry32 pseudo-random generator (so the same inputs always produce the same output, which is essential for reproducible analysis). You can choose 10 to 500 trades per simulation, set starting balance from $100 to $10 million, and select a risk percent from 0.1% to 50% per trade. The output includes probability of profit, probability of ruin, p5 worst-case balance, p95 best-case balance, median final balance and a sample-paths visualisation. The separate Full Monte Carlo panel runs the same engine over 10 to 5,000 trades with a fixed-dollar or fixed-percent sizing model, plus a histogram of final balances and 20 sample equity curves rendered on screen.

Expectancy is a dollar amount per trade. Profit factor is a ratio of gross profit to gross loss. They measure the same edge from two angles. A strategy with +$50 expectancy at $100 average loss has a profit factor around 1.9 (gross winnings 1.9x gross losses). Expectancy answers “how much money does this make on average per trade”. Profit factor answers “how durable is the edge”. The two move together but profit factor is more stable across small samples, which is why prop firms grade traders on profit factor and not raw expectancy. Both numbers are shown side by side on the Expectancy Dashboard above. Use expectancy to project P&L, use profit factor to grade the quality of the edge.

Four reasons, in order of frequency. One: you forgot to include real per-trade costs. The Costs % of profit metric on the dashboard shows what percentage of your gross profit goes to commission, slippage and spread. Above 30% means costs are eating most of the edge. Two: your sample is too small. With 20 trades, even a true 60% strategy can show 35% by chance. Wait for 100 closed trades before declaring an edge dead. Three: you are over-sized. A positive-expectancy strategy at Full Kelly can drawdown 60% on a normal losing streak and you quit before the math plays out. Drop risk to Half or Quarter Kelly. Four: the backtest was curve-fit. Run the sensitivity heatmap. If a 5-percentage-point drop in win rate flips the strategy from green to red, the edge is fragile and probably did not survive out-of-sample.

Yes. Expectancy is asset-agnostic. The four inputs (win rate, average win in dollars, average loss in dollars, average cost in dollars) work identically for equities, futures, forex, crypto spot, crypto perpetuals, options or any market where you can quantify a trade as a P&L number. The only caveat: enter your average cost in the same currency as your wins and losses. For forex, that means the pip value times the pip distance in your account currency, plus the spread in dollars. For crypto on a high-fee venue, costs can easily run $20 to $40 a round trip (0.20% to 0.40% on a $10,000 position) which a backtest without costs will completely miss. The Strategy Lab dropdown lets you save separate named strategies per market (e.g. “BTC Swing”, “EURUSD London Open”, “SPY 0DTE”) and switch with one click.