Back to Concepts Index

Concept Guide

Position Sizing

Position sizing determines how much capital to allocate to each investment — arguably the most important and most underappreciated skill in portfolio management. A strategy with a mediocre win rate but excellent position sizing can produce outstanding results; a strategy with a high win rate but poor position sizing can produce ruin. The Kelly Criterion, percent-of-equity sizing, and volatility-based sizing all provide systematic frameworks for making this decision rigorously.

Level: IntermediatePart V - Risk ManagementPublished Deep Guide

The Kelly Criterion: Optimal Sizing for Long-Run Wealth

The Kelly Criterion, derived by physicist John Kelly at Bell Labs in 1956, provides the mathematically optimal fraction of capital to bet on each opportunity to maximize the long-run geometric rate of capital growth. For a simple bet: Kelly fraction = (p × b - q) / b, where p = probability of winning, q = 1 - p, and b = net odds (amount won per dollar risked). For stock investing: Kelly fraction ≈ (Expected Return) / (Variance), or more practically, (Edge / Odds).

Full Kelly sizing maximizes expected log wealth (geometric growth) but produces aggressive positions and severe drawdowns in bad streaks — drawdowns that are psychologically intolerable and practically damaging even if mathematically optimal over infinite time. The standard practical solution is Half Kelly or Quarter Kelly: sizing at 50% or 25% of the pure Kelly fraction reduces median drawdowns dramatically while sacrificing only modestly from the maximum long-run growth rate. At Half Kelly, the expected drawdown is roughly half that of Full Kelly with roughly 75% of the long-run compound growth rate.

Kelly Criterion:

Win probability p = 0.55, Loss probability q = 0.45
Net odds b = 1.5 (win $1.50 for every $1 risked)

Full Kelly = (p × b - q) / b
           = (0.55 × 1.5 - 0.45) / 1.5
           = (0.825 - 0.45) / 1.5 = 0.375 / 1.5
           = 25% of capital

Half Kelly = 12.5% of capital

Percent-of-Equity and Volatility-Based Sizing

The most common practical approach for equity portfolios is percent-of-equity position sizing: allocate a fixed percentage of the portfolio to each position — often 2-5% for concentrated portfolios, 1-2% for more diversified ones. Equal weighting (1/N for N positions) is simple and removes the question of how much each position should receive relative to others. Research shows equal-weight portfolios frequently outperform value-weight portfolios in small-cap and international markets due to the rebalancing premium — selling relative winners and buying relative losers within the portfolio.

Volatility-based position sizing (discussed in the ATR section) adjusts position size to equalize each position's dollar volatility contribution: Position Size = (Account Risk per Trade) / (Stop Distance in dollars). If you risk 1% of a $100,000 account per trade ($1,000) and the stock has a stop distance of $5.00 (2× ATR), the position size is 200 shares ($1,000 / $5.00). This ensures that a volatile stock that hits its stop causes the same dollar loss as a less volatile stock — the position in the volatile stock is smaller to compensate. This prevents a few volatile positions from dominating the portfolio's risk profile.

Concentration, Diversification, and the Sizing Decision

The tension between diversification (spreading risk) and concentration (expressing conviction) is at the heart of position sizing. Buffett and Munger advocate concentrated portfolios (10-15 positions) for investors with genuine research edge, arguing that diversifying beyond your best ideas dilutes returns without proportionally reducing risk once adequate diversification is achieved. Index-oriented quantitative investors run 50-500 positions, recognizing that their edge is statistical (diversified factor exposure) rather than stock-specific.

The practical guideline: match concentration level to your edge source. If your edge is deep company-specific research, 10-20 positions concentrates that edge appropriately. If your edge is systematic factor exposure, 40-100 positions captures more of the statistical benefit. If your edge is unclear, start diversified (15-25 positions, equal weight) until conviction and edge are validated by track record. Never size positions so large that a single position loss causes catastrophic, psychologically disruptive damage to the overall portfolio — behavioral discipline requires staying in the game.

Key Takeaways

  • - The Kelly Criterion provides mathematically optimal position sizing for maximum long-run wealth growth; Half Kelly reduces drawdowns substantially while preserving most of the growth rate.
  • - Volatility-based sizing equalizes dollar risk across positions — smaller size in volatile securities, larger in stable ones — maintaining consistent risk contribution per position.
  • - Equal weighting (1/N) provides a simple, well-studied baseline; it often outperforms value weighting in less efficient markets through the rebalancing premium.
  • - Concentration should match edge source: stock-specific research → 10-20 positions; systematic factor exposure → 40-100 positions; unclear edge → start diversified.
  • - Never size a single position so large that losing it creates psychologically disruptive portfolio damage — behavioral capacity to stay invested is a constraint on optimal sizing.

Concept FAQs

How should I size positions differently in volatile markets?

In high-volatility markets, ATR-based sizing automatically reduces position sizes because stop distances expand with volatility. Many systematic traders also apply a portfolio-level volatility target: when market volatility (VIX) spikes, reduce total portfolio exposure to maintain a constant dollar volatility contribution from the equity allocation. This approach — sometimes called volatility targeting — naturally reduces equity exposure in turbulent markets and increases it in calm markets, providing systematic risk management without requiring macro forecasting.

Is it ever appropriate to hold a position larger than 5% of the portfolio?

For investors with very high conviction and deep research edge, positions of 10-20% are defensible — Buffett has held 30-40% of Berkshire in a single name at various points. The critical question is whether the position size exceeds your maximum tolerable loss if the thesis is wrong. A 20% position that loses 50% costs 10% of total portfolio wealth. If you can handle that loss without panic selling and without it materially impairing your lifestyle, the concentration is within your tolerance. If the thought of that scenario causes visceral distress, the position is too large regardless of conviction.

Put It Into Practice

Apply this concept using live stock signals, AIQ rankings, screeners, and side-by-side comparisons.

Market Regime DashboardSet risk context before position sizingPortfolio OptimizerImprove risk-adjusted portfolio mixRisk vs Return ExplainedTranslate risk metrics into decisionsSharpe Improvement GuideRaise risk-adjusted outcomesAI Stock SignalsBuy, Hold, or Sell setup for any stockAIQ RankingsTop-ranked stocks by composite signalStock ScreenerFilter by momentum, valuation, quality, and riskCompare Two StocksSide-by-side factor analysis
Related Concepts
Understanding Investment Risk Standard Deviation & Volatility Beta - Market Sensitivity Alpha - Excess Return Sharpe Ratio Sortino Ratio

Explore More

Use This Concept

Top Momentum StocksCompare SPY vs QQQPortfolio OptimizerStock Screener

Continue Learning

Concept Deep DiveMaximum DrawdownVolatility & VaRRisk vs Return Explained
Educational content only. Nothing on this page constitutes investment advice.