Standard Deviation & Historical Volatility

Introduction

Volatility — how much price moves — is one of the most important properties of any market, and the two foundational ways to measure it are standard deviation and historical volatility. Standard deviation is the raw statistical measure of how widely price scatters around its average; historical volatility annualises that into the percentage figure traders and risk managers actually quote. Together they underpin a huge amount of technical analysis and risk management: Bollinger Bands are built on standard deviation, position sizing is built on volatility, and option pricing revolves around it. Just as important is how volatility behaves — it clusters and mean-reverts, which is why quiet markets tend to precede storms. This lesson explains both measures, their uses, the historical-versus-implied distinction, and volatility's defining behaviour.

This builds on the Bollinger Bands and ATR lessons — standard deviation is the engine behind Bollinger Bands, and a statistical companion to ATR's range-based volatility.

Quick Definition

Standard deviation measures how widely price is dispersed around its average — a statistical gauge of volatility (high = scattered/volatile, low = clustered/calm). Historical (realised) volatility is the standard deviation of past returns, annualised into a percentage — how much an asset has actually moved over a period. Standard deviation underpins Bollinger Bands; volatility underpins position sizing and option pricing; and crucially, volatility clusters and mean-reverts over time.

The distinction in one line: standard deviation is the statistical tool; historical volatility is that tool applied to returns and annualised to give the standard volatility figure.

Standard Deviation: Measuring Dispersion

Standard deviation answers: how far does price typically stray from its average? Compute the mean over a window, measure how far each value sits from that mean, and standard deviation summarises that spread in a single number:

• High standard deviation → price is widely dispersed around its average — volatile, big swings.
• Low standard deviation → price clusters near its average — calm, small moves.

This is why standard deviation is the engine behind Bollinger Bands: the bands are placed a set number of standard deviations (commonly 2) above and below a moving average, so they widen automatically when standard deviation rises (volatile) and narrow when it falls (calm). When you watch Bollinger Bands expand and contract, you're watching standard deviation change. Standard deviation can also be plotted as its own indicator — a line that rises and falls with volatility — useful for spotting volatility extremes directly.

Historical Volatility: Annualising the Movement

Historical volatility (HV), also called realised volatility, takes standard deviation a step further into the form practitioners actually quote. It is the standard deviation of an asset's returns over a period (say 20 or 30 days), annualised into a percentage. So a stock with "30% historical volatility" has, based on its recent returns, been moving in a way that — scaled to a year — implies a 30% standard deviation of return. HV is backward-looking: it measures what has happened, from real price history. Its uses:

• Gauging the volatility regime: is this asset, right now, unusually calm or unusually turbulent versus its own history?
• Comparing instruments: HV lets you compare how volatile two very different assets are on a like-for-like, annualised basis.
• A baseline for options: HV is the realised yardstick against which the expected (implied) volatility in option prices is judged.

Historical versus Implied Volatility

A crucial distinction, especially if you ever touch options:

• Historical (realised) volatility is backward-looking — how much price has moved, from past data.
• Implied volatility (IV) is forward-looking — the market's expectation of future movement, backed out of option prices. When options are expensive, the market is implying lots of future movement; when cheap, little.

Comparing the two is a classic analysis: if IV is well above recent HV, the market expects more turbulence than has lately occurred (options look richly priced); if IV is below HV, the market expects things to calm down. (Implied volatility is explored in depth in the options material; here the key point is simply that HV is the realised counterpart to options' expected volatility.)

Volatility Clusters and Mean-Reverts

The single most important behavioural fact about volatility is that it is not random — it has two reliable tendencies:

• Clustering: volatility comes in regimes. Calm periods tend to be followed by more calm; turbulent periods by more turbulence. Big moves beget big moves; quiet begets quiet. (This is why a single volatile day often kicks off a volatile stretch.)
• Mean-reversion: over the longer run, volatility reverts toward its average. Extended unusually low volatility tends to precede an expansion (the calm before the storm), and extreme volatility eventually subsides toward normal.

This behaviour is the foundation of the squeeze (from the Keltner/Bollinger lesson): compressed volatility doesn't last, so a squeeze of unusually low volatility warns an expansion is coming. It also explains why traders treat extreme volatility as likely to subside, and persistent calm as a coiled spring.

Volatility and Position Sizing

A practical, vital use: sizing positions and stops to volatility. Because a volatile instrument swings far more than a calm one, putting the same amount of money into each — with the same stop distance — means wildly inconsistent risk. The disciplined fix is to size to volatility:

• In high volatility, use wider stops (to survive the swings) and a smaller position (to keep the money at risk constant).
• In low volatility, use tighter stops and can carry a larger position for the same risk.

Tools like ATR (range-based) or standard deviation give the volatility input for this. The goal is consistent risk per trade regardless of how active the instrument is — one of the cornerstones of sound risk management, covered further in the risk-management material.

Common Misconceptions

• "Standard deviation and historical volatility are different things entirely." HV is standard deviation — applied to returns and annualised. Standard deviation is the tool; HV is its standard, quotable form.
• "Historical volatility predicts the future." It's backward-looking — what has happened. The market's expectation of the future is implied volatility (from options).
• "Volatility is random." It clusters (regimes persist) and mean-reverts (extremes don't last). Low volatility tends to precede high, and vice versa.
• "Position size should be the same regardless of volatility." Sizing to volatility keeps risk consistent; ignoring it means a volatile instrument silently carries far more risk than a calm one.

Real-World Application

A trader notices a market has gone unusually quiet — its historical volatility has dropped to the low end of its multi-year range, and on the chart the Bollinger Bands (built on standard deviation) have contracted to a narrow ribbon. Knowing that volatility mean-reverts — that extended calm precedes expansion — they don't assume the quiet will last; they prepare for a volatility expansion and watch for the breakout direction. They also adjust their risk sizing: because volatility is currently low, their normal stop distance is tight, so they can carry a slightly larger position for the same risk — but they plan to cut size as volatility expands. When the move finally comes, volatility spikes and clusters (turbulence begets turbulence), and they widen stops and reduce size accordingly to keep their risk per trade steady. A second trader, ignoring volatility, used a fixed position size and a fixed stop throughout — taking far too little risk in the calm and far too much when volatility exploded. Reading volatility's level and behaviour shaped both the trade idea and the risk — which is what these measures are for.

Key Takeaways

• Standard deviation measures price dispersion around its average — the statistical gauge of volatility (high = volatile, low = calm) — and is the engine behind Bollinger Bands.
• Historical (realised) volatility is the standard deviation of past returns, annualised — the standard, quotable measure of how much an asset has moved.
• It differs from implied volatility, which is forward-looking (the market's expected future movement, from option prices).
• Volatility clusters (regimes persist) and mean-reverts (extremes don't last) — so low volatility tends to precede high, the basis of the squeeze.
• Size positions and stops to volatility (via ATR or standard deviation) to keep risk per trade consistent — smaller/wider in high volatility, larger/tighter in low.