Position Sizing - How Much to Put in Each Stock
Position sizing determines how much capital to allocate to each investment. It is the least discussed and most impactful skill in individual stock investing. Two investors can hold the same 15 stocks and produce wildly different returns based solely on how much they put in each one. If the best-performing stock is a 2% position and the worst-performing stock is a 10% position, the portfolio underperforms despite containing a winner. If those allocations are reversed, the same stocks produce a completely different outcome.
Professional portfolio managers obsess over position sizing. Joel Greenblatt, who compounded at 40% annually over 20 years at Gotham Capital, has said that concentration in the best ideas was a larger contributor to his returns than stock selection itself. The key was not finding good stocks; hundreds of investors found the same ones. The key was sizing them aggressively enough to matter.
The Equal-Weight Starting Point
The simplest approach is equal weighting: divide the portfolio equally among all positions. With 20 stocks, each receives 5%. With 15 stocks, each receives approximately 6.7%. This method requires no judgment about relative conviction, no complex calculations, and no ongoing adjustment beyond periodic rebalancing.
Equal weighting has a meaningful advantage: it prevents the catastrophic outcome of overweighting a single disastrous position. If every stock is 5% of the portfolio, the maximum damage from any single stock going to zero is 5%. That is painful but survivable. A 20% position going to zero takes years to recover from.
Equal weighting also introduces a systematic small-cap tilt. In a portfolio of 20 stocks ranging from Apple ($3 trillion market cap) to a $5 billion mid-cap, equal weighting gives the same allocation to each. Market-cap weighting, by contrast, would put vastly more in Apple. Since small and mid-cap stocks have historically outperformed large-caps over long periods, the equal-weight approach captures this premium by default.
The limitation of equal weighting is that it treats every idea as equally compelling. In practice, an investor's 3rd-best idea is usually significantly better than their 15th-best idea. Allocating equal capital to both wastes the informational advantage of deeper analysis on the highest-conviction positions.
Conviction-Based Sizing
Conviction-based sizing assigns larger allocations to ideas the investor understands best and finds most compelling. The framework is straightforward: rank positions by conviction level, assign position sizes proportionally, and set maximum and minimum thresholds.
A practical structure:
- Tier 1 (highest conviction): 6% to 10% each, maximum 3 to 4 positions
- Tier 2 (high conviction): 4% to 6% each, 5 to 7 positions
- Tier 3 (moderate conviction): 2% to 4% each, 5 to 8 positions
This creates a portfolio where 30% to 40% of capital is concentrated in the 3 to 4 best ideas, while the remaining 60% to 70% provides diversification through smaller positions. If the Tier 1 positions perform well, they drive meaningful portfolio returns. If they underperform, the diversified Tier 2 and 3 positions limit the damage.
Conviction should be grounded in analytical depth, not emotional attachment or confirmation bias. High conviction means the investor has thoroughly analyzed the business model, competitive position, financial statements, management quality, and valuation. It does not mean reading one bullish article and feeling excited. An investor who has spent 40 hours analyzing a company's 10-K filings, listening to earnings calls, and reading industry reports has earned a larger position than one arrived at through a 10-minute review.
The Kelly Criterion
The Kelly criterion, developed by physicist John Kelly in 1956, provides a mathematical formula for optimal bet sizing:
Kelly % = W - (1 - W) / R
Where W is the probability of winning and R is the ratio of average win to average loss. If an investor believes a stock has a 60% probability of appreciating 50% and a 40% probability of declining 25%, the Kelly percentage is:
0.60 - (0.40 / 2.0) = 0.60 - 0.20 = 0.40 or 40%
The formula suggests a 40% allocation, which is reckless for a single stock. This illustrates a well-known property of the Kelly criterion: full Kelly sizing produces optimal long-term growth but with enormous volatility. Most professional investors who use the Kelly framework apply "half Kelly" or "quarter Kelly," reducing the suggested position size by 50% to 75%.
The practical challenge of the Kelly criterion for stocks is estimating the inputs. Unlike blackjack, where probabilities can be calculated precisely, stock investing involves highly uncertain probability distributions. An investor who estimates a 60% chance of a 50% gain may be off by 20 percentage points in either direction. Small estimation errors produce large sizing errors.
The Kelly criterion is more useful as a conceptual tool than a precise calculator. Its central insight, that position size should increase with both the probability of being right and the magnitude of the expected payoff, is directionally correct even when the exact numbers are unreliable.
Maximum Position Size Guidelines
Risk management demands an upper bound on individual position sizes. Even the highest-conviction ideas can be wrong. Enron looked like a high-conviction position for many fund managers in 2000. So did Lehman Brothers in 2007.
For individual investors, a maximum position size of 10% to 15% of the equity portfolio provides a reasonable ceiling. At 10%, a total loss wipes out one-tenth of the portfolio, an amount that aggressive saving can replace within a year or two. Proper diversification across sectors further limits exposure to any single outcome. At 15%, the damage is more severe but still recoverable. Above 15%, a single-stock failure begins to threaten the portfolio's long-term trajectory.
Some exceptional investors have held larger positions. Warren Buffett's Apple stake grew to over 40% of Berkshire Hathaway's equity portfolio. Philip Fisher's Motorola position became a similar proportion of his personal holdings. But these are investors with decades of experience, deep industry knowledge, and a level of conviction that took years to build. For most investors, particularly those in the first decade of their investing career, a 10% ceiling per position provides adequate room for conviction while limiting catastrophic risk.
Sector concentration should also be capped. Holding five stocks at 8% each in the technology sector means 40% of the portfolio is in one sector. When that sector declines, as technology did by 78% from 2000 to 2002, the portfolio suffers accordingly. A maximum of 25% to 30% in any single sector prevents this concentration.
Building Positions Over Time
A position does not need to reach its target size in a single transaction. Building a position gradually has several advantages. It allows the investor to develop conviction as the thesis plays out. It provides multiple entry points at different prices, reducing the risk of buying entirely at a peak. And it imposes discipline by requiring the thesis to remain intact over time.
A practical approach is to start with a half-position and add to it as conviction grows. An investor targeting a 6% position in a stock might buy an initial 3% allocation, study the next earnings report, and add another 3% if the thesis is confirmed. If the thesis deteriorates, the investor has only risked half the intended allocation.
Peter Lynch described a similar approach at Fidelity Magellan. He would buy a small "starter position" in a stock he was researching, which forced him to follow the company more closely. If subsequent analysis supported the thesis, he would add to the position. If not, he would sell the small stake with minimal impact.
Averaging down, buying more of a declining stock, is a position-building strategy that requires extreme caution. If the decline reflects deteriorating fundamentals rather than temporary market sentiment, adding to the position compounds the error. The critical question before averaging down is always: has the investment thesis changed? If the answer is yes, adding to the position is throwing good money after bad. If the answer is genuinely no, and the price decline creates a better entry point for the same thesis, adding can be appropriate.
Position Sizing With New Money
As new capital enters the portfolio through regular contributions, the investor faces a recurring allocation decision: which positions receive the new money? Three approaches work well.
First, direct new money toward underweight positions. If the target for a stock is 5% and it has drifted to 3.5% due to underperformance, new contributions can bring it back toward target without selling anything. This accomplishes rebalancing through contribution rather than trading.
Second, direct new money toward new ideas. If the investor has identified a new stock to add to the portfolio, contributions fund the new position without requiring sales of existing holdings. This is particularly tax-efficient in taxable accounts.
Third, direct new money toward the highest-conviction current holdings. If the investor's top three ideas remain highly attractive at current prices, adding to those positions concentrates the portfolio in its strongest opportunities. This approach trades diversification for concentration but makes sense when conviction is genuinely high and position sizes remain below maximum thresholds.
Monitoring and Adjusting Over Time
Position sizes drift as prices change. A stock that doubles in a portfolio goes from a 5% position to roughly a 9% position (assuming the rest of the portfolio is flat). This drift is not inherently a problem. Letting winners run is one of the most valuable habits in investing. But at some point, the position becomes large enough that its risk dominates the portfolio.
A practical rule: review position sizes quarterly and trim any position that exceeds the maximum threshold. If the ceiling is 10% and a position has grown to 12%, sell enough to bring it back to 8% to 10%. This captures some profit while maintaining a meaningful allocation to a winning investment.
The proceeds from trimming go to underweight positions or new ideas, maintaining overall portfolio balance without requiring a complete restructuring. This incremental approach to position management is less disruptive, more tax-efficient, and more psychologically manageable than large-scale portfolio overhauls.
Sizing matters more than selection for most investors. A portfolio of average stocks with excellent position sizing will outperform a portfolio of excellent stocks with random sizing, because sizing determines how much the portfolio benefits from its best ideas and how much it suffers from its worst.
Put these principles into practice. Track fundamentals, build portfolios, and analyze stocks with AI-powered insights.
Start Free on GridOasis →