Third-Party Research & Methodology Only

This section shares summaries of third-party academic research and descriptions of quantitative models. The content represents the findings of the original researchers, not the opinions or recommendations of Foxholm Financial. Foxholm Financial does not publish hypothetical or backtested performance metrics on its quantitative research pages. All content is restricted to methodology, signal construction, factor logic, and risk architecture. SEC rules require that investment advisers not present misleading performance data, and our methodology-only approach reflects that standard and the firm's fiduciary obligations.

Standard Deviation

Risk Metric Statistical Measure

Standard deviation measures how spread out a set of values is around their average. In investing, it is the most common way to quantify risk, because it captures how much an asset's returns tend to bounce above and below their mean over time.

A low standard deviation means returns cluster tightly around the average, suggesting more predictable performance. A high standard deviation means returns swing widely, indicating greater uncertainty. Most risk-adjusted performance metrics, including the Sharpe ratio, use standard deviation as their measure of total risk.

Definition

Standard deviation is calculated in three steps. First, find the average (mean) of all the return values. Second, measure how far each individual return falls from that average and square those distances. Third, average those squared distances and take the square root of the result.

Formula Concept

Standard Deviation = Square Root of the Average of Squared Deviations from the Mean

Each return is compared to the average return. The differences are squared so that positive and negative deviations do not cancel each other out. Taking the square root at the end brings the result back to the same units as the original returns (percentages, in the case of investment returns).

When applied to investment returns, standard deviation is typically annualized. If monthly returns have a standard deviation of 3%, the annualized figure is approximately 3% × the square root of 12, which equals roughly 10.4%. Annualizing makes it easier to compare assets measured over different time frames.

Interpretation: The 68-95-99.7 Rule

If returns follow a normal distribution (a bell-shaped curve), standard deviation tells you how likely different outcomes are. This relationship is known as the empirical rule, or the 68-95-99.7 rule.

Range	Probability	Example (10% mean, 15% std dev)
Within 1 standard deviation	About 68% of the time	Returns between −5% and +25%
Within 2 standard deviations	About 95% of the time	Returns between −20% and +40%
Within 3 standard deviations	About 99.7% of the time	Returns between −35% and +55%

In the example above, an investment with a 10% average annual return and 15% standard deviation would be expected to land between −5% and +25% in roughly two out of every three years. A return worse than −20% should happen less than 3% of the time, assuming returns are normally distributed. In practice, financial returns are not perfectly normal, which is one of the key limitations discussed below.

Practical Example: Comparing Two Stocks

Consider two stocks with the same average annual return of 8% over the past ten years.

Metric	Stock A	Stock B
Average annual return	8%	8%
Standard deviation	12%	25%
Best year	+22%	+45%
Worst year	−8%	−30%

Both stocks delivered the same average return, but the experience of holding them was very different. Stock A's returns stayed relatively close to the 8% average. Stock B swung from a 45% gain in its best year to a 30% loss in its worst. An investor in Stock B needed to tolerate much larger short-term losses to earn the same long-run return. Standard deviation captures this difference in a single number.

This is why standard deviation matters for portfolio construction. Two investments with identical average returns are not interchangeable if one is twice as volatile as the other. The more volatile investment requires a longer time horizon and a greater tolerance for temporary losses.

Why Standard Deviation Matters for Portfolios

Standard deviation is foundational to modern portfolio theory. Harry Markowitz's 1952 framework for portfolio construction treats standard deviation as the definition of risk. The entire concept of diversification (spreading investments across assets that do not move in lockstep) works because combining assets with imperfect correlation reduces the portfolio's overall standard deviation.

Consider a simple example. Two assets each have a standard deviation of 20%. If their returns are perfectly correlated (they always move together), a 50/50 portfolio also has a 20% standard deviation. But if their correlation is 0.3, the portfolio's standard deviation drops to roughly 15%. The individual assets are equally risky, but the combination is less risky than either one alone. This is the mathematical basis of diversification.

Standard deviation also plays a central role in risk-adjusted return metrics. The Sharpe ratio divides excess return by standard deviation. The efficient frontier plots portfolios by their expected return against their standard deviation. Mean-variance optimization finds the portfolio weights that minimize standard deviation for a given level of expected return.

Known Limitations

Limitations to Keep in Mind

Assumes returns are normally distributed. Standard deviation is most meaningful when returns follow a bell-shaped curve. In reality, financial returns often have "fat tails," meaning extreme events (both gains and losses) happen more frequently than a normal distribution predicts. The 2008 financial crisis produced losses that were multiple standard deviations beyond what the model would expect.
Treats upside and downside equally. A stock that occasionally shoots up 30% and one that occasionally drops 30% could have the same standard deviation, but most investors view those situations very differently. The Sortino ratio addresses this by measuring only downside deviation, ignoring positive surprises entirely.
Backward-looking. Standard deviation is calculated from historical data. A stock with a low standard deviation over the past five years is not guaranteed to remain stable going forward. Regime changes (shifts in market conditions, interest rate environments, or company fundamentals) can make historical volatility a poor predictor of future volatility.
Sensitive to time period and frequency. Daily returns produce different standard deviation estimates than monthly or annual returns. A stock measured over a calm three-year period will look less risky than one measured over a period that includes a market crash. Always check the measurement period and data frequency when comparing standard deviations from different sources.
Does not capture sequence risk. Two investments can have the same standard deviation but very different patterns of returns. For someone drawing income from a portfolio (such as a retiree), the order in which gains and losses occur matters as much as their magnitude. Standard deviation does not distinguish between these sequences.