The first principle: dollar risk drives position size
Position size is calculated from account risk percentage and stop distance, not chosen first. The formula:
Position size = risk amount ÷ (stop distance × pip value per lot)
Worked example on a $10,000 account at 1% risk per trade ($100): with a 50-pip stop on EUR/USD at $10/pip per standard lot, position size = $100 ÷ (50 × $10) = 0.20 standard lots. With a 100-pip stop on the same parameters: position size = $100 ÷ (100 × $10) = 0.10 standard lots.
Same dollar risk, different lot size — because the stop distance changes. A trader who fixes lot size first and varies stop distance produces inconsistent dollar risk: a fixed 1.0 lot on a 25-pip stop risks $250 (2.5% of the account); the same 1.0 lot on a 100-pip stop risks $1,000 (10%). Consistent per-trade risk requires the stop to drive the size, not the other way around.
Stop-loss placement: respecting volatility
A Stop-loss placed inside normal price noise can be triggered by that noise before the trade thesis plays out. Average true range (ATR) quantifies this noise as the instrument's typical distance per bar, gaps included, and moves with Volatility. It is shown as a price difference in the instrument's own quote, which converts to pips on EUR/USD (0.0070 = 70 pips) and USD/JPY (0.70 = 70 pips) and reads directly in dollars per ounce on gold, because gold has no universal pip.
On an illustrative EUR/USD daily ATR of 70 pips, an assumed input, a 20-pip stop on a position held through the day sits inside one third of that range — a price excursion that occurs routinely without the underlying trend changing. A stop that close is within reach of noise alone.
An ATR-based alternative sets the stop at a multiple of the relevant ATR that the trader chooses, such as 1× or 2×. With that 70-pip ATR, a 1× ATR stop is 70 pips and a 2× ATR stop is 140 pips. The position size adjusts to honour the same dollar risk: at $100 risk and the 70-pip stop, position size = $100 ÷ (70 × $10) = 0.14 standard lots. The trade-off: wider stops reduce the trigger rate from noise, but with dollar risk fixed the profit target moves out in proportion, so price has to travel further to reach the same R-multiple, the realised expression of the Risk-reward ratio. The dollar reward at that R-multiple does not change.
The asymmetric cost of drawdown
A 10% Drawdown requires an 11.1% return to recover. A 25% drawdown requires 33.3%. A 50% drawdown requires 100%. The asymmetry compounds — every percentage point lost requires more than a percentage point of return to replace, because the return is calculated on a smaller equity base.
Quantifying the impact of risk per trade: a $10,000 account at 5% risk per trade has 95% of equity remaining after one loss-making trade, assuming the stop fills at its level. After three consecutive losses, equity sits at $10,000 × 0.95³ = $8,574 — a 14.3% drawdown requiring a 16.6% return to recover. After ten consecutive losses, equity is $10,000 × 0.95¹⁰ = $5,987 — a 40.1% drawdown requiring a 67.0% return.
The same account at 1% risk per trade, after ten consecutive losses: $10,000 × 0.99¹⁰ = $9,044 — a 9.6% drawdown requiring a 10.6% return to recover. The lower per-trade risk transforms the same losing streak from career-threatening to routine.
At a 55% win rate, with trades independent of each other, the probability that a given ten-trade run is all losses is about 1 in 2,937 (0.45¹⁰). Over many trades that small chance adds up: a trader running 50 trades a week, roughly 2,600 a year, has about a 38.5% chance of at least one streak of ten or more losses in a year, and about 62% over two years. At that frequency the streak is not unusual; the risk-per-trade percentage is what determines whether it ends the account, and Leverage leaves the loss per pip unchanged on the same position size.
Risk per trade as the structural choice
Risk per trade is a choice about drawdown before it is a choice about return: the same losing streak costs a very different share of the account at 1% than at 5%.
Whatever the Win rate, losing streaks arrive, and the risk per trade sets what they cost. At 2% risk on the current balance, ten consecutive losses take 18.3% of the account (1 − 0.98¹⁰), and recovering needs a 22.4% gain. At 5% risk the same streak takes 40.1% (1 − 0.95¹⁰), and recovering needs a 67.0% gain.
Lowering risk per trade shrinks the gain from each profitable trade and the loss from each loss-making trade in the same proportion. The trade-off is mainly about variance — accepting smaller absolute moves in exchange for a higher probability of surviving any specific drawdown sequence. When the risk per trade is already too high for a strategy's edge, compounded losses outweigh the larger gains, and lowering it raises long-run growth as well.
What risk management is not
Risk management is not predicting which trades will lose. The trader does not know in advance which trades will trigger the stop-loss. The plan is structurally agnostic about which specific trades fail.
Risk management is not avoiding loss-making trades. The trader expects loss-making trades. The plan accepts them as the cost of the win-rate distribution and sizes positions so the account survives the losing sequences the win-rate distribution makes likely.
Risk management is not a guarantee. The plan reduces the probability of catastrophic drawdown, its Risk of ruin, to a lower level that can be estimated under stated assumptions; it does not eliminate the possibility. Black-swan events, Price gap risk on weekends, and broker counterparty risk all sit outside the standard position-sizing framework. The framework is necessary, not sufficient — it solves a specific problem (variance from normal trade-sequence outcomes) and leaves others to broader account-management decisions.