Position Sizing Methods Compared: Fixed, Kelly, Risk-Based

Why Position Sizing Matters More Than Entries

Here's a counterintuitive fact: a strategy with random entries and good position sizing can outperform a strategy with perfect entries and poor sizing. Van Tharp's research demonstrated this decades ago, yet most traders spend 90% of their time optimizing entry signals and 10% on position sizing.

The math is simple. A 50% drawdown requires a 100% return to recover. A 25% drawdown only needs 33%. Position sizing is the primary tool that controls drawdown depth. Everything else — indicators, patterns, timing — is secondary to capital preservation.

Let's compare the four most common methods using actual backtest data to see how each performs under real market conditions.

Method 1: Fixed Dollar Amount

The simplest approach — risk the same dollar amount on every trade regardless of account size. If you start with $10,000 and risk $200 per trade, you always risk $200 whether your account grows to $15,000 or shrinks to $7,000.

Advantages: Dead simple to implement. Consistent psychological experience — every trade "feels" the same. Easy to track P&L per trade.

Disadvantages: Doesn't compound. As your account grows, $200 becomes a smaller percentage, slowing growth. As it shrinks, $200 becomes a larger percentage, accelerating losses. This creates an asymmetric risk profile that works against you in both directions.

In StratBase backtests, fixed dollar sizing typically produces the smoothest equity curves but the lowest total returns. It's a training-wheels approach — useful for learning, suboptimal for growth.

Method 2: Fixed Fractional (% Risk Per Trade)

Risk a constant percentage of your current equity on each trade. The standard recommendation is 1-2% per trade. With a $10,000 account at 2% risk, you risk $200. If the account grows to $12,000, you risk $240. If it drops to $8,000, you risk $160.

This is the industry standard for good reason:

• Natural compounding — position sizes grow with your account
• Automatic de-risking — smaller positions during drawdowns slow the bleeding
• Mathematically impossible to go to zero — you always risk a fraction, never the whole
• Simple to backtest — one parameter to optimize (the percentage)

The main debate is what percentage to use. Conservative: 0.5-1%. Standard: 1-2%. Aggressive: 2-5%. Above 5% per trade, you're essentially gambling — a 10-trade losing streak at 5% risk produces a 40% drawdown.

Method 3: Kelly Criterion

The Kelly Criterion calculates the mathematically optimal bet size that maximizes long-term growth rate. The formula: f* = (b × p - q) / b, where b is the reward-to-risk ratio, p is the probability of winning, and q = 1 - p.

Example: your strategy has a 55% win rate and 2:1 reward/risk. Kelly says: f* = (2 × 0.55 - 0.45) / 2 = 32.5% per trade. That's enormous — and therein lies the problem.

Full Kelly produces maximum geometric growth but with massive volatility. A 50-70% drawdown is normal under full Kelly. Most traders use fractional Kelly:

• Half Kelly (f*/2): 75% of optimal growth, dramatically lower drawdowns
• Quarter Kelly (f*/4): Still solid compounding, very manageable risk
• Tenth Kelly (f*/10): Conservative, approximates 1-2% fixed fractional

The catch: Kelly requires accurate estimates of win rate and R:R. If your estimates are wrong (and they always are somewhat), Kelly can over-bet catastrophically. This is why practitioners always use fractional Kelly and recalculate parameters regularly from rolling backtest windows.

Method 4: Volatility-Based (ATR Sizing)

Adjust position size based on current market volatility, typically measured by Average True Range (ATR). The formula: Position Size = (Account × Risk%) / (ATR × Multiplier).

If a crypto asset has ATR of $500 and you risk 2% of $10,000 ($200) with a 2× ATR stop: Position = $200 / ($500 × 2) = 0.2 units. When volatility doubles (ATR = $1,000), position size halves automatically.

This method shines in markets with regime changes. During calm periods, you take larger positions (more opportunity per unit of risk). During chaos, you automatically shrink exposure. The result: more consistent risk across different market environments.

In StratBase, you can configure ATR-based stops and the platform automatically adjusts position sizing to maintain your target risk percentage. This is particularly valuable for crypto strategies where volatility can 5x overnight.

Backtesting Position Sizing: What the Data Shows

We ran the same trend-following strategy on BTC/USDT (2020-2025) with all four methods at comparable risk levels:

ATR-based produced the best risk-adjusted returns (highest Sharpe). Half Kelly had the highest absolute return but with the deepest drawdown. Fixed fractional offered a solid middle ground. Fixed dollar lagged in total return but had a smooth ride.

The right choice depends on your risk tolerance and strategy type. For most traders, fixed fractional (1-2%) is the sweet spot of simplicity and performance.

Further Reading

• Backtesting on Investopedia
• Sharpe Ratio on Investopedia
• Drawdown on Investopedia