You are watching a stock trade in a tight range for three weeks. The bands contract, the channel narrows, and every volatility tool on your chart is quietly agreeing: something is about to move. But what exactly are those tools measuring? Strip away the overlays and the color coding, and nearly all of them rest on the same statistical building block. Standard deviation.
Most charting platforms let you add Bollinger Bands or Keltner Channels in two clicks. Fewer traders stop to ask what the envelope width actually represents. Understanding standard deviation on its own terms gives you a sharper read on every volatility-derived indicator you will ever use.
What Standard Deviation Measures
Standard deviation quantifies dispersion. It tells you how far individual closing prices have strayed from their average over a set lookback period. A low reading means prices are clustered near the mean. A high reading means they are spread out.
That single idea drives most of the volatility toolkit. Bollinger Band Width is literally the distance between bands expressed as a percentage, and that distance is set by a standard deviation multiplier. Historical Volatility scales standard deviation to an annualized figure. Even Keltner Channel Width, which uses ATR instead of standard deviation, is measuring the same underlying concept: how much price is moving relative to its recent average.
I find the raw standard deviation line more useful than people expect. Before I check whether Bollinger Bands are squeezing, I look at the standard deviation value itself. It removes one layer of interpretation.
The Formula
The standard deviation of closing prices over an n-period lookback is:
\sigma = \sqrt{\frac{1}{n} \sum_{i=1}^{n} (C_i - \bar{C})^2}Where C_i is each closing price, \bar{C} is the simple moving average of those closes, and n is the lookback period.
Step by step:
1. Calculate the simple moving average of the last n closing prices.
2. For each close, subtract the mean and square the result.
3. Average those squared differences.
4. Take the square root.
Most charting platforms use the population standard deviation (dividing by n) rather than the sample version (dividing by n-1). For a 20-bar lookback, the difference is negligible. I use 20 periods on daily charts, which matches the default Bollinger Band setting and keeps the readings directly comparable.
Reading the Indicator
Standard deviation plots as a single line in a sub-window beneath price. The line rises when price swings widen and falls when they contract. There are no fixed overbought or oversold levels. The reading is always relative to recent history.
Three states matter:
Low and falling. Volatility is compressing. Price is coiling. This is the pre-breakout environment that squeeze traders watch for. It does not predict direction, only that the current calm is unlikely to last.
Rising sharply. A breakout or breakdown is in progress. The standard deviation spike confirms that price is moving with genuine dispersion, not just a single bar anomaly. I pay attention to how fast the line rises. A gradual increase often means a trending move. A vertical spike can signal a climax.
High and rolling over. The expansion phase is maturing. When standard deviation peaks and begins to decline, the trend is losing its rate of change. This does not mean the trend is over, but the easy part likely is.
Practical Workflow
Here is how I use standard deviation as a standalone overlay, separate from Bollinger Bands:
I add a 20-period standard deviation to the sub-window. Then I visually mark the lowest readings over the past 100 bars. Those troughs correspond to the tightest consolidation phases on the price chart. I look for a standard deviation reading that drops to at least 30% below its own 100-bar average. That flags a squeeze worth watching.
When standard deviation breaks above its recent range, I check direction on price. Is price closing above or below the 20-period moving average? The standard deviation tells me volatility is expanding. Price tells me which side won.
I also use it as a filter. If I am considering a mean-reversion trade, I want standard deviation to be elevated and rolling over. If it is still rising, the move has momentum and fading it is riskier.
Where Standard Deviation Fits With Other Tools
Standard deviation is the input, not the output. Think of it as the engine beneath three categories of indicators:
Envelope indicators. Bollinger Bands plot the moving average plus and minus a standard deviation multiplier (typically 2). The bands widen and contract in direct response to the standard deviation value. If you understand the raw number, you understand why the bands behave the way they do.
Volatility gauges. Bollinger Band Width and percent-B are both derived from standard deviation. Historical Volatility annualizes it. Each adds a layer of normalization, but the core measurement is the same.
Regime filters. Some systematic traders use a standard deviation threshold to classify the market as trending or ranging. When standard deviation exceeds its 100-bar moving average, the market is in an expansion regime. Below it, contraction. This is a blunt filter, but it keeps you from applying mean-reversion entries in a trending environment.
Common Mistakes
Treating low volatility as a signal. A compressed standard deviation tells you volatility is low. It does not tell you to buy or sell. You still need a directional trigger. I have seen traders enter positions purely because “the squeeze is tight,” then watch price drift sideways for another two weeks.
Ignoring the lookback period. A 10-period standard deviation reacts quickly but produces more noise. A 50-period version smooths out the reading but lags meaningful compression phases. The 20-period default works for daily charts. On intraday timeframes, I shorten it to 14.
Comparing absolute values across instruments. A standard deviation of 2.50 on a $300 stock is not comparable to 2.50 on a $30 stock. If you want to compare volatility across tickers, normalize by dividing standard deviation by the closing price. That gives you a coefficient of variation, which is what Historical Volatility effectively does.
Quick Reference
Default lookback: 20 periods (matches Bollinger Bands default).
Input: closing prices.
Output: single line showing dispersion from the mean.
Low readings: volatility compression, potential squeeze setup.
High readings: volatility expansion, active trend or climax.
Falling from a peak: expansion phase fading, trend rate of change declining.
Best paired with: a directional indicator (moving average, ADX) to confirm which way price is breaking once volatility expands.
When Dispersion Earns a Spot on Your Chart
Standard deviation is not glamorous. It does not paint arrows or flash buy signals. But it answers the one question every other volatility tool is trying to answer: how spread out are prices right now compared to their recent average?
If you use Bollinger Bands, you are already using standard deviation. You just might not realize it. Adding the raw indicator to your chart strips away the envelope and gives you a cleaner view of the volatility cycle. Compression, expansion, exhaustion. That cycle repeats on every instrument and every timeframe. Standard deviation is the simplest way to see it.
Educational content only. Not investment advice. Trading involves risk. You are responsible for your decisions.