What it means
Position sizing in FX is the calculation that turns a chart-driven stop-loss decision into a specific lot size. The standard formula: lot size = (account equity × risk per trade %) / (stop distance in pips × pip value). Most retail traders skip this and trade fixed lot sizes — which produces fixed dollar risk on tight stops and oversized risk on wide stops, the opposite of what's needed.
Why it matters
Position sizing is the second-biggest determinant of long-term FX results, after strategy edge. Two traders with the same entry and exit rules but different sizing methods produce wildly different results over 12 months. The Kelly criterion + risk-of-ruin math both say: size each trade as a fixed fraction of equity, not a fixed lot count.
How to use it
Pick a fixed risk percentage per trade (0.5-1% for new traders, 1-2% for experienced). Before each trade, compute lot size = (equity × risk%) / (stop pips × pip value per lot type). Adjust lot type (micro vs mini vs standard) to land on the closest integer multiple. Never override the calculation because the trade 'feels right'.
$10,000 account, 1% risk per trade = $100 risk. EUR/USD trade with a 30-pip stop. Lot size = $100 / (30 × $10 standard-lot pip value) = 0.33 standard lots = 3 mini lots + 3 micro lots. The trader inputs 0.33 lots into the broker platform.
The fixed-fractional sizing formula
Standard formula used by virtually every professional FX desk: position size (in lots) = (account equity × risk % per trade) / (stop distance in pips × pip value in account currency). For a $10,000 USD account, 1% risk, 30-pip stop on EUR/USD: position = ($10,000 × 0.01) / (30 × $10/standard-lot-pip) = 0.33 standard lots. The broker accepts decimal lot sizes; this becomes 33,333 EUR notional exposure.
- 0.5% risk per trade: very conservative, suits new traders or unproven strategies
- 1% risk per trade: standard for established traders with edge
- 2% risk per trade: aggressive, reserved for high-conviction setups
- 5% risk per trade: rarely justifiable, exceeds Kelly criterion for most edges
Why fixed-lot sizing destroys accounts
Common mistake: 'I trade 1 mini lot per trade.' Result: a 10-pip stop = $10 risk, a 100-pip stop = $100 risk. The trader has implicitly accepted a 10x larger risk on the wider-stop trade — usually the lower-probability setup. Fixed-fractional sizing reverses this: tight stops get larger lot sizes (more chances to be right), wide stops get smaller lot sizes (less risk if wrong).
Adjusting for pair volatility
Different pairs require different stop distances for the same probability of getting stopped out. Using the 30-day ATR (Average True Range): GBP/JPY 30-day ATR ≈ 180 pips, EUR/USD 30-day ATR ≈ 75 pips. A 1-ATR stop on GBP/JPY (180 pips) vs EUR/USD (75 pips) requires position size scaled inversely. The advanced formula: position = (equity × risk%) / (ATR multiple × pip value). This naturally normalises across pairs of different volatility.
Kelly criterion and optimal sizing
The Kelly criterion gives the theoretically optimal bet size for a given edge: f = (bp - q) / b, where b is odds, p is win probability, q is loss probability. For a typical FX strategy with 50% win rate and 1.5:1 reward/risk, Kelly suggests 16% per trade — far too aggressive in practice because edge estimates are noisy. The practical rule: size at 0.1× to 0.25× of theoretical Kelly. So a 16% Kelly becomes 1.6% to 4% per trade in practice.
Position sizing during drawdowns
Two schools. (1) Fixed-fractional: risk % stays constant regardless of equity. As account shrinks during drawdown, position sizes shrink proportionally — automatic risk control. (2) Anti-martingale: increase risk % after losses to 'recover'. Mathematically guarantees ruin in a long-enough sequence. (3) Optimal-F (a Kelly variant): adjust risk % based on rolling win rate. Used by some quant desks. For most traders, fixed-fractional at 0.5-1% is the practical sweet spot.
Common position-sizing mistakes
Three errors. (1) Using leverage as a position-size proxy ('I'll trade at 30:1') — leverage is a constraint, not a sizing input. (2) Trading correlated pairs at full size each — three long EUR/USD, GBP/USD, AUD/USD positions at 1% each is effectively 2.5-3% risk on the same trade thesis (USD weakness). (3) Adjusting lot size mid-trade to 'average down' — converts a 1% loser into a 2-3% loser with worse expectancy.
Frequently asked
How much should I risk per trade?
Most experienced FX traders risk 0.5-1% per trade. New traders should start at 0.25-0.5% until they verify their strategy has positive expectancy. 2%+ per trade is reserved for high-conviction setups or already-proven strategies.
What is the difference between risk and position size?
Risk is the dollar amount you can lose if your stop-loss is hit. Position size is the lot size that produces that risk given the stop distance. Position size is calculated from risk, not chosen independently.
How do I size positions in correlated pairs?
Treat correlated pairs (EUR/USD, GBP/USD, AUD/USD all expressing 'short USD') as a single trade thesis. If your max risk per thesis is 2%, distribute that across the correlated positions. Three positions at 1% each in correlated pairs is effectively 2.5-3% risk, not 3%.
Should I use the Kelly criterion?
Use a fractional Kelly (0.1× to 0.25× of full Kelly) as an upper bound on your risk-per-trade. Full Kelly is theoretically optimal but assumes accurate edge estimates, which retail traders rarely have. The practical translation: most strategies suggest 0.5-2% per trade by fractional Kelly.
What happens to position sizing during a losing streak?
Under fixed-fractional sizing, your account shrinks during a losing streak, so 1%-of-equity translates to smaller dollar risk per trade — automatic risk control. Under fixed-lot sizing, losses compound because position size stays constant while equity shrinks.
How do I size positions when scaling in or out?
Decide the maximum position size in advance based on the total risk you'll accept. Scaling in (adding to a winner) shouldn't exceed that max. Scaling out (taking partial profit) reduces both position and remaining risk proportionally.
Want a worked example or a deeper dive? Ask Rocky how this concept applies to your specific watchlist or trade idea.
Ask Rocky