Standard Deviation

What it measures

Standard deviation is the foundational statistical measure of how spread out price (or returns) has been around its own average over the lookback period — it's the same calculation that builds the width of Bollinger Bands. A higher standard deviation means price has been swinging more widely; a lower one means it's been tightly clustered near its average.

How readings are interpreted

Rising standard deviation means volatility is expanding — larger price swings are becoming the norm. Falling standard deviation means volatility is contracting — price is settling into a tighter range, often a precursor to an eventual breakout once the compression resolves.

Conventional levels

• Rising standard deviation: volatility expanding, larger expected price swings
• Falling standard deviation: volatility contracting, tighter expected range
• Extended periods of unusually low standard deviation have historically preceded sharp volatility expansions (the same logic underlying the Squeeze setup)
• Because it's expressed in raw price units, direct comparison across differently-priced securities requires normalizing (similar to how NATR normalizes ATR)

Where it works, and where it does not

A pure volatility measure with no directional component — useful in virtually any market condition as a read on how much price is moving, regardless of which way. Foundational to many other tools (Bollinger Bands, Z-Score) rather than typically used entirely on its own.

Commonly read alongside

• Bollinger Bands: Directly built from this exact calculation — Bollinger Band Width is essentially a normalized view of the same information
• Z-Score: Also built from standard deviation, measuring how many standard deviations price sits from its own mean
• NATR: A comparable volatility read from a different calculation base (true range rather than closing price dispersion)

Known limitations

• Expressed in raw price units, so it isn't directly comparable across different-priced securities without normalization
• Assumes a roughly normal distribution of price changes, which real markets don't always follow (see Kurtosis and Skew for exactly this limitation)
• Reacts to both up and down moves identically — unlike the Ulcer Index, it doesn't distinguish "good" volatility from "bad" (downside) volatility

In practice

• Best used as a building block for other decisions (position sizing, band width) rather than as a standalone signal
• Compare a security's current standard deviation to its own historical range to judge whether volatility is currently elevated or subdued
• Pair with Kurtosis and Skew for a fuller picture, since standard deviation alone assumes a symmetric, normal-ish distribution that isn't always accurate
• A useful screening input for finding either "calm, low-volatility" or "currently expanding volatility" candidates depending on your strategy