Volatility and Standard Deviation: Measuring Investment Risk

beginnerPublished: 2025-12-29

The S&P 500 averaged 10.38% annual returns (1926-2025) with 18% standard deviation (volatility, Source: historical market data). That means in two-thirds of years, returns fell between -8% and +28% (average ± 1 std dev), and in the other one-third, you got wildly different results (-37% to +54%). The practical antidote: standard deviation tells you the typical range of outcomes (not just the average), and matching your portfolio's volatility to your risk tolerance prevents panic-selling when inevitable swings occur.

What Standard Deviation Measures (Dispersion Around Average)

Standard deviation measures how spread out returns are around the average (mean). Low standard deviation = returns cluster tightly around average (predictable). High standard deviation = returns vary wildly year-to-year (unpredictable, risky).

Example (low volatility bond): a Treasury bond fund averages 5% annual return with 6% standard deviation. Returns typically range from -1% to +11% (5% ± 6%). Most years cluster near 5% (predictable income, small variation).

Example (high volatility stock): a tech stock averages 15% annual return with 35% standard deviation. Returns typically range from -20% to +50% (15% ± 35%). Annual outcomes vary massively (one year +60%, next year -30%, highly unpredictable).

The point is: average return alone is misleading (two investments can have the same 10% average, but one swings ±5% yearly while the other swings ±30%). Standard deviation captures the uncertainty in achieving that average (how rough the ride will be).

The 68-95-99.7 Rule (Interpreting Standard Deviation)

Standard deviation follows a normal distribution (bell curve) for most financial assets. This gives you probability ranges for outcomes:

~68% of returns fall within 1 standard deviation of the mean (average ± 1 std dev). For S&P 500 (10% average, 18% std dev): 68% of years delivered returns between -8% and +28% (10% - 18% = -8%; 10% + 18% = +28%).

~95% of returns fall within 2 standard deviations. For S&P 500: 95% of years delivered returns between -26% and +46% (10% ± 36%).

~99.7% of returns fall within 3 standard deviations. For S&P 500: 99.7% of years delivered returns between -44% and +64%. Anything outside this range is a 3-sigma event (extremely rare—2008's -37% was inside 2 standard deviations, not a statistical anomaly).

Example (using bond fund with 5% return, 6% std dev):

  • 68% of years: returns between -1% and +11% (5% ± 6%)
  • 95% of years: returns between -7% and +17% (5% ± 12%)
  • 99.7% of years: returns between -13% and +23% (5% ± 18%)

The durable lesson: when you see a stock fund with 20% average return and 40% standard deviation, don't anchor on the 20% (you'll rarely get exactly 20%). Expect returns between -20% and +60% in most years (20% ± 40%). If you can't stomach a -20% year without panic-selling, that fund is too volatile for you (find lower std dev options).

Benchmark Standard Deviations by Asset Class (Stocks, Bonds, Cash)

Different asset classes have characteristic volatility ranges. Stocks are 2-3x more volatile than bonds. Cash has almost no volatility (stable value but low return).

Stocks (S&P 500):

  • Standard deviation: 15-20% annualized (long-term average ~18%)
  • Return: 10.38% annualized (1926-2025)
  • Typical range: -8% to +28% annually (in 68% of years)
  • Interpretation: expect large swings (-30% to +40% in any given year)

Bonds (investment-grade, Bloomberg US Aggregate):

  • Standard deviation: 5-8% annualized (long-term average ~6%)
  • Return: 5-6% annualized (1926-2024)
  • Typical range: -1% to +11% annually (in 68% of years)
  • Interpretation: much more stable than stocks (2-3x lower volatility)

Cash (T-bills, money market):

  • Standard deviation: 0-1% annualized (essentially zero)
  • Return: 3-4% annualized (1926-2024)
  • Typical range: 3% to 4% annually (very predictable)
  • Interpretation: no volatility but low return (purchasing power risk from inflation)

Portfolio blends (based on stock/bond mix):

  • 100% stocks: ~18% std dev
  • 80/20 (stocks/bonds): ~14% std dev
  • 60/40: ~11% std dev (40% reduction in volatility vs 100% stocks)
  • 40/60: ~8% std dev
  • 100% bonds: ~6% std dev

Why this matters: moving from 100% stocks (18% std dev) to 60/40 (11% std dev) cuts volatility by 40% while reducing expected return by only 15% (from 10% to 8.5%). That's an asymmetric trade-off (much less risk for modest return reduction). Adding bonds beyond 60/40 provides diminishing marginal benefit (40/60 has 8% std dev vs 11% for 60/40—only 3% improvement for sacrificing more return).

Standard Deviation vs Actual Drawdowns (Maximum Loss Matters More)

Standard deviation estimates typical volatility but doesn't capture tail risk (maximum drawdown in crashes). The S&P 500's 18% std dev suggests 95% of years should fall within -26% to +46% (2 standard deviations). But actual worst years exceeded this: -43% (1931), -37% (2008), -26% (1974).

Why the mismatch: financial returns don't follow a perfect normal distribution (they have fat tails, meaning extreme events happen more often than bell curve predicts). Crashes like 2008 are 2-sigma events (within statistical expectations) but feel catastrophic (and are psychologically harder than std dev implies).

Maximum drawdowns (peak-to-trough declines) by asset class:

  • S&P 500: -50% (2007-2009), -49% (2000-2002), -34% (2020 COVID, recovered in 5 months)
  • Bonds (Agg): -13% (2022 rate hikes, worst bond market in 40 years), -5% (1994)
  • 60/40 portfolio: -32% (2008), -24% (2000-2002), -20% (2022)

The point is: standard deviation gives you the typical range (most years). Maximum drawdown tells you the worst-case scenario (what you must survive without panic-selling). When evaluating risk, ask: can I tolerate a -30% to -50% drawdown (for stocks) or -20% to -30% (for 60/40) without selling at the bottom? If not, reduce volatility (add bonds, shift to lower std dev assets).

Using Standard Deviation to Compare Investments (Risk-Adjusted Returns)

Sharpe ratio measures return per unit of risk (risk-adjusted return). Formula: (Return - Risk-Free Rate) / Standard Deviation. Higher Sharpe = better risk-adjusted performance (you're getting more return for each unit of volatility you endure).

Example (comparing two funds):

  • Fund A: 12% return, 20% std dev → Sharpe = (12% - 4% risk-free) / 20% = 0.40
  • Fund B: 9% return, 10% std dev → Sharpe = (9% - 4%) / 10% = 0.50

Fund B has lower absolute return (9% vs 12%) but better risk-adjusted return (0.50 vs 0.40 Sharpe). You're getting 0.50% return per 1% of volatility with Fund B vs 0.40% with Fund A. If you can tolerate the volatility, Fund A wins (higher absolute return). If you want smoother ride, Fund B wins (better return per unit of risk).

Example (stocks vs bonds over 30 years):

  • Stocks: 10% return, 18% std dev → Sharpe = (10% - 4%) / 18% = 0.33
  • Bonds: 6% return, 6% std dev → Sharpe = (6% - 4%) / 6% = 0.33

Historically, stocks and bonds had similar Sharpe ratios (both ~0.33) over very long periods. Stocks delivered higher absolute returns (10% vs 6%) but required enduring much higher volatility (18% vs 6%). Bonds gave lower returns with much less risk (same risk-adjusted outcome). This is why 60/40 portfolios worked well historically (combining assets with similar Sharpe ratios but different return/risk profiles created efficient portfolio).

The test: if two investments have similar returns but different standard deviations, choose the lower volatility option (you're getting same outcome with less stress). If two have similar standard deviations but different returns, choose the higher return (you're being paid more for same risk). Use Sharpe ratio to compare across different return/risk profiles.

Portfolio Volatility Is Not Weighted Average (Diversification Benefit)

When you combine assets with different volatilities, portfolio volatility is lower than the weighted average of individual volatilities (if assets are not perfectly correlated). This is the mathematical benefit of diversification.

Example (naive calculation, ignoring correlation):

  • You hold 60% stocks (18% std dev) and 40% bonds (6% std dev)
  • Weighted average volatility: (0.60 × 18%) + (0.40 × 6%) = 10.8% + 2.4% = 13.2%

Actual portfolio volatility (accounting for correlation):

  • If stock-bond correlation is -0.3 (negative, typical in low-inflation regime), actual portfolio std dev is ~11% (lower than weighted average 13.2% because negative correlation reduces volatility)
  • If correlation is +0.5 (positive, high-inflation regime like 2022), portfolio std dev is ~12% (closer to weighted average because positive correlation provides less diversification benefit)

The point is: negative correlation between assets amplifies diversification benefit (portfolio volatility falls below weighted average). Positive correlation reduces benefit (portfolio volatility approaches weighted average). This is why stock-bond correlation matters—when it's negative (2008, 2020), your 60/40 portfolio has much lower volatility than expected. When it's positive (2022), volatility is higher than expected (both assets moving together).

Formula (for two assets): Portfolio variance = (w1² × σ1²) + (w2² × σ2²) + (2 × w1 × w2 × σ1 × σ2 × ρ), where w = weight, σ = std dev, ρ = correlation. Take square root to get portfolio std dev. (You don't need to calculate this—just understand that correlation affects portfolio volatility.)

High Volatility Assets Require Longer Time Horizons (Volatility Decay Over Time)

Volatility decreases as time horizon lengthens (annualized volatility shrinks when measured over 5, 10, 20 years vs 1 year). This is why stocks are appropriate for long horizons despite high short-term volatility.

Example (S&P 500 rolling returns):

  • 1-year periods (1926-2025): 18% annualized std dev (huge variation: -43% worst, +54% best)
  • 5-year periods: 10-12% annualized std dev (range narrows: worst ~-3% annualized, best ~28% annualized)
  • 10-year periods: 6-8% annualized std dev (range further narrows: worst ~-1% annualized, best ~20% annualized)
  • 20-year periods: 3-5% annualized std dev (almost always positive: worst ~+3% annualized, best ~18% annualized)

Why volatility declines: over short periods (1 year), luck dominates (market can go anywhere). Over long periods (20 years), fundamentals dominate (companies grow earnings, dividends compound, valuations mean-revert). The law of large numbers smooths outcomes (you experience many up years and down years, averaging toward long-term mean).

The durable lesson: if you need money in 1-2 years, stocks' 18% volatility is unacceptable (you could hit a -30% year right when you need to withdraw). If your horizon is 20+ years, that same 18% annual volatility becomes 3-5% annualized (very tolerable). Match asset volatility to time horizon: stocks for 10+ years, bonds for 3-7 years, cash for <3 years.

Detection Signals (Your Portfolio Volatility Doesn't Match Risk Tolerance)

You're likely holding too much volatility if:

  • You check your portfolio daily and feel stressed when it's down 5-10% (sign you can't handle the volatility). Reduce stock allocation or stop checking daily (volatility is noise over days/weeks, signal over years/decades).
  • You sold stocks during March 2020 COVID crash (down -34%) or 2022 rate hikes (down -18%) because you couldn't tolerate the drawdown. You locked in losses by panic-selling (instead of rebalancing or holding). Next time, add bonds to reduce volatility before the crash.
  • Your portfolio has 25%+ standard deviation and you're 5 years from retirement (you can't afford a -40% year at age 64). Shift toward lower volatility (60/40 or 40/60) to reduce max drawdown risk.

You're likely holding too little volatility (under-earning) if:

  • Your portfolio is 100% bonds or cash (0-6% std dev) and you're 30 years old with 35-year horizon. You're sacrificing 4-5% annual returns vs balanced stock/bond portfolio (costing you $500k+ in lost compounding over 35 years).
  • You can't tolerate any decline and sit in money market earning 3.5% while inflation is 2.7% (real return 0.8%—barely beating inflation). You're taking purchasing power risk instead of market risk (equally bad long-term).
  • You're chasing yield in high-dividend stocks (REITs, utilities, tobacco) because you can't handle growth stock volatility, but those sectors have crashed -40%+ in past cycles (2008: REITs -70%). You haven't actually reduced risk, just shifted it.

The test: calculate your portfolio's standard deviation (weighted average of holdings). If it's >15% and you lose sleep during -10% months, reduce volatility (add bonds, shift to lower-volatility stock funds). If it's <8% and you're decades from retirement, increase volatility (add stocks to capture higher long-term returns).

Next Step: Calculate Your Portfolio's Standard Deviation

Action (15 minutes): estimate your portfolio's volatility and compare to tolerance.

Step 1: List major holdings. For each fund, find the standard deviation (fund fact sheet, Morningstar page, or broker research tools—usually listed under "risk" or "volatility"). If you own individual stocks, assume 25-35% std dev (typical for single stocks). If you own index funds, use benchmarks: S&P 500 = 18%, Total Bond = 6%, International = 20%.

Step 2: Calculate weighted std dev (rough approximation). Multiply each holding's std dev by its portfolio weight, then sum. Example:

  • 60% S&P 500 fund (18% std dev): 0.60 × 18% = 10.8%
  • 40% bond fund (6% std dev): 0.40 × 6% = 2.4%
  • Portfolio approximate std dev: 10.8% + 2.4% = 13.2% (this ignores correlation; actual is ~11% due to negative stock-bond correlation, but 13% is close enough for self-assessment)

Step 3: Apply 68-95 rule to estimate range. With 11% portfolio std dev and 8% expected return (60/40 historical), expect:

  • 68% of years: returns between -3% and +19% (8% ± 11%)
  • 95% of years: returns between -14% and +30% (8% ± 22%)
  • Max drawdown (2008-level crash): approximately -25% (based on 60/40 historical max)

Step 4: Gut-check tolerance. Could you stay invested through a -25% drawdown without panic-selling? If yes, your allocation matches tolerance. If no, reduce volatility (shift toward more bonds, target 8-10% portfolio std dev instead of 11%). If you're confident you could handle -30% to -40%, you can increase volatility (shift toward more stocks, target 14-16% std dev).

Where to find std dev: Morningstar fund pages (free), Vanguard/Fidelity/Schwab fund research (click "risk" tab), Yahoo Finance (Statistics section for ETFs). If not listed, use asset class proxies: large-cap stocks 15-18%, small-cap 20-25%, bonds 5-8%, cash 0-1%.

Standard deviation measures the uncertainty of achieving average returns (not just the average itself). The S&P 500's 18% std dev means you should expect returns between -8% and +28% in most years (not always 10%). Stocks are 2-3x more volatile than bonds (18% vs 6% std dev), which is why adding bonds to portfolios reduces volatility disproportionately (60/40 has 11% std dev, 40% lower than 100% stocks). Use standard deviation to match portfolio volatility to your time horizon and risk tolerance: if you can't stomach a -25% drawdown, don't hold a portfolio with 15%+ std dev (shift toward more bonds, target 8-11% std dev). If you're decades from retirement and sitting in 100% bonds (6% std dev), you're under-earning (add stocks to capture higher long-term returns despite higher volatility). Calculate your portfolio's std dev annually, gut-check whether you could survive a -2 sigma event (2× std dev decline) without panic-selling, and adjust allocation accordingly.

Sources:

Related Articles

Compound Interest: Taxable vs Tax-Advantaged Accounts

beginner

How Economic Cycles Affect Investment Outcomes

intermediate

Inflation-Protected vs Traditional Savings: TIPS and I Bonds Explained

beginner