Risk Per Trade & Position Sizing
Trading Mastery
11 min read
Calculate optimal position sizes based on account risk, stop distance, and volatility to ensure no single trade can cause catastrophic loss.
Loading
11 min read
Calculate optimal position sizes based on account risk, stop distance, and volatility to ensure no single trade can cause catastrophic loss.
Risk per trade is the fixed dollar amount you accept losing if a single trade hits its stop, expressed as a percentage of account equity (typically 0.5%–2%). Position size is then derived: position size = (account × risk%) / (entry − stop). This lesson covers the formula, the math behind "risk small", R-multiples, and when to use Kelly or volatility-based sizing.
Even with a strong trading edge, poor risk management will blow up your account.
Great traders don’t just focus on how much they can make. They focus relentlessly on how much they could lose.
This post will teach you:
Most beginner traders:
The result? Inconsistent outcomes, emotional swings, and blown accounts.
A trader without risk control is just a gambler in disguise.
Risk per trade = the amount of capital you're willing to lose on a single trade.
Rule of thumb: Risk 0.5% to 2% of your account per trade.
| Account Size | 1% Risk | 2% Risk |
|---|---|---|
| $1,000 | $10 | $20 |
| $10,000 | $100 | $200 |
| $50,000 | $500 | $1,000 |
This amount stays fixed, no matter where your stop-loss is placed.
Once you know your risk per trade, you can determine your position size using this formula:
position size = risk$ / stop size
Account $10,000. Risk per trade 1% = $100. Stop distance $600. Position size = $100 / $600 = 0.1667 BTC (notional approx $10,000, i.e. 1x leverage). Sized from stop, not from leverage.
This keeps your risk at $100, even if you’re wrong.
On a leveraged perp at 5× leverage, the same $10,000 account / 1% / $600 stop still gives 0.1667 BTC notional. Margin used = notional/5 ≈ $2,000. Liquidation price ≠ stop — it sits wherever margin is exhausted, often far beyond your $600 stop. Always size from the stop, never from leverage.
Don’t just think in dollars. Think in R.
R-multiples standardize performance and let you compare across trades, strategies, and timeframes. The framework comes from Van K. Tharp, Trade Your Way to Financial Freedom (2nd ed., 2007), which is the canonical source for R-multiples and expectancy.
Consistent traders think in R. Inconsistent traders think in $.
| Trade | Entry | Stop | Risk | Position Size | Exit | P/L | Result |
|---|---|---|---|---|---|---|---|
| 1 | $100 | $95 | $100 | 20 shares | $115 | +$300 | +3R |
| 2 | $102 | $98 | $100 | 25 shares | $98 | -$100 | -1R |
| 3 | $95 | $90 | $100 | 20 shares | $100 | +$100 | +1R |
Result: +3R – 1R + 1R = +3R total profit, no over-risking, calm execution.
Risk of ruin is the probability that a sequence of losses takes your account below the point where you can't recover. It's not metaphor — it's the math of bounded random walks. At a 50% win rate / 1:1 R/R, risking 1% per trade gives essentially zero ruin probability over 1,000 trades; risking 10% per trade gives ~60%.
Let's say your max drawdown tolerance is 20%. If you risk 5–10% per trade, you could reach that in 2–3 losing trades.
If you risk 1% of starting equity per trade (fixed-fractional of initial), 20 consecutive losses = 20% drawdown. If you risk 1% of CURRENT equity each trade, the drawdown compounds: 20 losses ≈ 18.2%, and you need ~22 losses to cross 20%.
| Risk per trade | Losses to 10% DD (fixed-init) | Losses to 20% DD (fixed-init) | P(ruin) at 50% win rate, 1:1 R/R |
|---|---|---|---|
| 0.5% | 20 | 40 | ≈ 0 |
| 1% | 10 | 20 | ≈ 0 |
| 2% | 5 | 10 | ~1% |
| 5% | 2 | 4 | ~30% |
| 10% | 1 | 2 | ~60% |
Ruin probability over 1,000 trades vs risk-per-trade. Note the convex blow-up between 2% and 10%.
Independent of edge, drawdowns are guaranteed. With a 50% win rate, P(at least 7 consecutive losses in 100 trades) ≈ 55%. With a 60% win rate, P(at least 5 in a row) ≈ 50%. Size so that this normal sequence is uncomfortable, not fatal.
Position size determines whether your edge can survive its own variance. Underrisk and the edge compounds; overrisk and the math of ruin makes the edge irrelevant.
This is how pros survive bad weeks, news spikes, emotional mistakes, and losing streaks. For explicit ruin probability tables across edge configurations, see Perry Kaufman, Trading Systems and Methods (5th ed., 2013), Ch. 23 "Risk Control".
| Method | Inputs needed | Stop type | Drawdown profile | Best for |
|---|---|---|---|---|
| Fixed-dollar ($X per trade) | $ amount | any | gets worse as account grows | nobody, long-term |
| Fixed-fractional (1–2% of equity) | account × % | any | bounded, predictable | retail discretionary |
| Volatility-based (ATR multiplier) | account × % + ATR | k·ATR | smoother across regimes | systematic |
| Kelly Criterion (full f*) | account × f* | any | growth-optimal but ~50% expected DD | edges measured with high confidence |
| ¼-Kelly / Fractional Kelly | account × f*/4 | any | far lower variance, ~75% of growth | prop desks, real-world ops |
Fixed-fractional is the default for retail traders building their first system. Move to ATR-based or fractional Kelly only after you've measured your edge across hundreds of trades.
Once you're consistent, you may explore more advanced ways to scale your risk responsibly.
Equity-based scaling E.g., reduce position size by 10% if your account drops by 10% (to protect from compounding drawdowns)
Volatility-based stops
Use ATR to set the stop distance, then size from it: stop = entry − k·ATR (k typically 1.5–3); position size = risk$ / (k·ATR). Example: $10,000 account, 1% risk = $100, BTC at $60,000, ATR(14) = $800, k=2 → stop $1,600 away → size = $100 / $1,600 = 0.0625 BTC. This keeps $ risk fixed while adapting position size to current volatility.
Kelly Criterion (Advanced) The Kelly formula calculates the optimal bet size based on your edge and win/loss profile:
Worked example:
Why ½/¼-Kelly is the practical choice — not just convention: full Kelly assumes you know p and b exactly. In trading you estimate them from a small, non-stationary sample, so your true edge is almost certainly smaller than your measured edge. ½-Kelly accepts ~75% of the growth rate for far less variance; ¼-Kelly is what most prop desks actually run.
Kelly optimizes growth, not utility. It increases volatility and drawdowns if used at full size. Use it carefully and only with stable, proven metrics. For the academic foundations see Edward O. Thorp, "The Kelly Criterion in Blackjack, Sports Betting, and the Stock Market" (2006), and Aaron Brown, Red-Blooded Risk (2011); for fixed-fractional theory see Ralph Vince, The Mathematics of Money Management (1992).
❗ Regardless of method: Never adjust risk based on emotion. Size up based on data, not desperation.
Risk 0.5% to 2% of your account equity per trade. This is the fixed-fractional rule and it's the default for retail discretionary trading because it bounds drawdown predictably regardless of stop distance. Smaller percentages (0.5–1%) are appropriate when you're learning or your edge is unproven; 2% is the upper limit before consecutive-loss math gets dangerous.
Use the formula: position size = risk dollars / stop size in dollars, where risk dollars = account equity × risk %. Example: a $10,000 account at 1% risk has $100 of risk; if the stop is $600 away from entry, position size = $100 / $600 = 0.1667 BTC. The formula keeps your dollar loss fixed regardless of where the stop is.
1R is your risk per trade — the dollar amount you'd lose if a single trade hit its stop. A trade that earns $300 when you risked $100 is +3R; a loss of $50 on the same trade is −0.5R. R-multiples normalize performance across trades, strategies, and timeframes, which is why consistent traders track results in R rather than dollars.
Not at full size. Full Kelly assumes you know your win rate and risk/reward exactly, but trading edges are estimated from small, non-stationary samples — your true edge is almost certainly smaller than your measured edge. Most prop desks run ¼-Kelly; retail traders are usually safer with fixed-fractional 1–2% per trade than with any Kelly fraction.
Set the stop a multiple k (typically 1.5 to 3) of ATR away from entry, then derive position size from that stop: position size = risk dollars / (k × ATR). This keeps dollar risk fixed while adapting position size to current volatility — bigger positions in calm regimes, smaller positions in volatile ones.
If you want consistency in results, start with consistency in risk.
Your strategy’s edge plays out over time—but only if you're still alive to trade it.
Manage risk like a surgeon, not a gambler. Keep your losses small, your capital safe, and your focus sharp.
Next: Drawdowns and Variance makes the math of consecutive losses concrete; The 17 Most Important Trading Metrics shows how R-multiples roll up into expectancy and Sharpe.