Trading performance metrics: win rate, expectancy, R-multiples and drawdown

win rate = winning trades ÷ total closed trades

The fraction of closed trades that made money. 12 winners out of 30 trades is a 40% win rate. On its own it says nothing about profitability - it must always be read next to the size of the average win and loss. Chasing a high win rate is one of the classic ways to build a strategy with small wins and rare, ruinous losses.

R = amount risked at entry (entry price − invalidation, × size)

“R” is the amount a trade was risking: the distance from entry to the planned invalidation, times position size. Expressing each outcome as a multiple of R - a trade that risked $100 and made $250 is +2.5R; one that lost its full risk is −1R - normalises trades of different sizes onto one scale, so a small experiment and a full-size position can be compared fairly.

average risk-reward = average win (in R) ÷ average loss (in R)

The realised ratio between the average winner and the average loser. Note this is measured from outcomes, not from hopes: the ratio your targets promised at entry and the ratio your exits deliver are usually different numbers, and the gap between them is itself worth journaling.

expectancy = (win rate × average win) − (loss rate × average loss)

The average result per trade, combining win rate and win/loss size into a single number. Worked example: a 40% win rate, average win +2R, average loss −1R:

(0.40 × 2R) − (0.60 × 1R) = 0.8R − 0.6R = +0.2R per trade

A positive expectancy means the process makes money on average; this trader earns a fifth of their risked amount per trade, despite losing more often than winning. If risk per trade is 1% of the account, that is roughly +0.2% of the account per trade before fees - which is why fees and funding must be included in the per-trade figures, not remembered separately.

drawdown = (equity peak − equity trough) ÷ equity peak

The largest peak-to-trough fall in account equity: an account that grew to $10,000 and fell to $8,500 before recovering had a 15% maximum drawdown. It measures the pain a strategy inflicts on the way to its returns, and it is the number most likely to make a trader abandon a profitable process. Position sizing is its main lever - risking 1% of the account per trade means even ten consecutive losses draw down roughly 10%, a survivable stretch that a 5%-per-trade sizing turns into a catastrophic 40%.

All of these are estimates from a sample, and small samples lie. Ten trades tell you almost nothing - a 40%-win-rate process will quite often show 6 or 7 winners, or 1 or 2, in any given run of ten. Treat metrics over fewer than 30–50 trades per setup as provisional, resist redesigning your process after every losing week, and never compute them from a curated record: a journal that only contains the trades you felt like logging produces numbers that flatter and mislead. This is the practical argument for importing fills automatically rather than logging by hand.

One more honesty rule: where a figure cannot be established - a position whose cost basis is unknown, a fill that could not be priced - exclude it and say so, rather than estimating. A statistic quietly padded with guesses is worse than a smaller, true one.