Risk & Backtesting Lesson 2 10 min read

Advanced Position Sizing

The 1% rule is a starting point — not the full picture. As your strategy gets more defined and your portfolio grows, a more nuanced approach to position sizing can significantly improve both returns and drawdown control. This lesson covers the framework used by professional traders.

Risk Per Trade vs. Portfolio Exposure

Two separate risk dimensions every trader needs to manage:

Example: You have 5 open positions, each risking 2% of account. Your total exposure is 10%. If there's a market crash affecting all positions at once (correlation risk), you can lose 10% in a single event. This happens frequently in crypto where all coins move together.

Correlation Risk in Crypto

Crypto assets are highly correlated — especially during sell-offs. When BTC drops 10%, almost everything else drops with it, often more. This means:

The Correlation Trap A trader with five 2% risk positions thinks they have a diversified portfolio. But during a crypto-wide sell-off, all five positions hit their stops simultaneously — total loss: ~10%. In high-correlation markets, treat your total exposure as your real risk, not the per-trade risk.

The Volatility-Adjusted Position Size

Instead of always risking the same percentage per trade, volatility-adjusted sizing accounts for how erratic an asset is moving. The formula uses ATR (Average True Range):

Volatility-Adjusted Position Sizing
InputExample A (Low Vol)Example B (High Vol)
Account€10,000€10,000
Risk per trade (1%)€100€100
Entry price€60,000€60,000
ATR (daily)€600 (1%)€2,400 (4%)
Stop = 2× ATR€1,200€4,800
Position size€100 / €1,200 = 0.083 BTC€100 / €4,800 = 0.021 BTC

High-volatility assets automatically get smaller positions — protecting you from outsized losses

Kelly Criterion — Optimal Bet Size Curve
Full Kelly (theoretical only) Half Kelly Use this in practice Over-betting Growth reverses Ruin possible Bet size → 100% of bankroll Real models are imperfect → use Half Kelly max. Never full Kelly in live trading.

Kelly Criterion — A Framework, Not a Formula

The Kelly Criterion is a mathematical formula for optimal bet sizing: f = (bp - q) / b, where b = odds, p = win probability, q = loss probability.

In trading terms, Kelly tells you the optimal fraction of your bankroll to risk given your edge. The problem: estimating your true edge precisely enough for Kelly to be useful is very difficult in practice.

Most professional traders use "half-Kelly" — betting half the Kelly-optimal amount. This dramatically reduces variance while giving up only a fraction of expected return. As a rule of thumb: if full Kelly says 8%, you trade 4%.

Kelly Requires an Honest Edge Estimate Kelly only works if your win rate and average win/loss estimates are accurate. Using your last 10 trades to estimate gives a very noisy signal. You need 50+ trades across different market conditions before the inputs are reliable enough to use. Before that: stick to 1-2% fixed risk.
Portfolio Heat — Total Risk Exposure Across All Open Positions
POSITION RISK % HEAT BAR BTC Long 1.0% ETH Long 0.8% SOL Short 0.5% TOTAL HEAT 2.3% MAX 6% Rule: Sum of all position risks should never exceed 5-6% of portfolio at once. 2.3% heat = still room to add trades or ride existing ones through volatility.

Scaling In vs. Full Position Entry

Rather than entering your full position at once, some traders scale in across 2–3 entries:

Risk management: each partial entry has its own stop, but the combined maximum risk across all entries should still not exceed your per-trade risk limit.

For most beginners, scaling in adds complexity without much benefit. Master full-position entries first.

Key Takeaways

Ready to practice?

Work through all lessons in the Risk & Backtesting track, then test your knowledge with the quiz.

Take the Track Quiz →
← Previous Backtesting Your Strategy Next → Walk-Forward Testing