Reading Volatility with Bollinger Bands
This article explains how to use price channels created with standard deviation to interpret the contraction and expansion of volatility, and to identify overbought and oversold conditions.
What are Bollinger Bands?
Bollinger Bands are an indicator devised by John Bollinger in the 1980s. They consist of a centerline and two bands that envelop the price above and below it. The centerline typically uses a 20-day simple moving average, while the upper and lower bands are calculated by adding or subtracting a certain multiple of the standard deviation for the same period. The commonly used default settings are a 20-day period and a standard deviation multiplier of 2.
The key is that the band width is not fixed. When prices fluctuate, the standard deviation increases, causing the bands to widen. When prices are calm, the standard deviation decreases, and the bands narrow. In essence, Bollinger Bands are less of a trend indicator and more of a channel that visualizes volatility.
Calculation Principles and Numerical Examples
The centerline is the average of the closing prices over the most recent 20 days. To this, the standard deviation of the closing prices for the same 20 days is calculated. The upper band is set as the centerline plus 2 times the standard deviation, and the lower band is the centerline minus 2 times the standard deviation. For example, if the 20-day average is $100 and the standard deviation is $5, the upper band would be $110 and the lower band would be $90.
Even with the same average of $100, if volatility is low and the standard deviation decreases to $2, the bands narrow to a width of $4, with the upper band at $102 and the lower band at $98. Conversely, if sharp price swings are frequent and the standard deviation increases to $8, the bands widen to a width of $32, with the upper band at $116 and the lower band at $84. Assuming a normal distribution, approximately 95 percent of data falls within ±2 standard deviations of the mean, meaning that prices moving outside the bands are statistically rare events.
Interpretation Guidelines
The basic intuition is to interpret prices touching or exceeding the upper band as short-term overheating, and prices touching or falling below the lower band as short-term weakness. However, a band touch itself is not a trading signal. In a strong uptrend, prices can 'walk the band,' continuously staying near the upper band. If you interpret an upper band touch as a sell signal in such a scenario, you might miss out on further trend continuation.
It is useful to observe changes in band width. A 'squeeze' where the bands narrow significantly indicates a state of compressed volatility, suggesting the potential for a large price movement thereafter. However, a squeeze does not indicate direction, so whether the breakout will be upward or downward needs to be corroborated with other evidence, such as volume or trend.
Practical Application Methods
The %B indicator and the Bandwidth indicator, which quantify the band width, are frequently used. %B indicates where the price is within the bands, ranging from 0 to 1. If the price is at the lower band, %B is 0; at the centerline, it's 0.5; and at the upper band, it's 1. For example, if %B is 0.05, it means the price is near the lower band, and if it's 0.95, it's near the upper band.
From a counter-trend perspective, in a sideways market, it can be used to buy when the price hits the lower band and rebounds, and to sell when it hits the upper band and pulls back. From a trend-following perspective, the centerline is viewed as a support or resistance level. In an uptrend, a price pullback to the centerline followed by a rebound can be taken as a clue for trend continuation. Combining Bollinger Bands with other indicators like RSI or volume analysis, and acting only when two or three pieces of evidence align, is a way to increase the reliability of signals.
Pitfalls and Limitations
The most common misconception is viewing a band touch as an automatic trading signal. When a trend is strong, prices can cling to one of the bands for an extended period, making mechanical counter-trend trading prone to losses. Since Bollinger Bands are based on moving averages, they are inherently a lagging indicator and cannot predict sudden gaps or event-driven volatility.
Standard deviation assumes a normal distribution, but actual stock price returns often exhibit 'fat tails,' meaning extreme movements outside the bands occur more frequently than statistical predictions. Furthermore, the distribution of volatility differs between markets with price limits, like the Korean market (referring to the KOSPI and KOSDAQ markets, which have daily price limits), and markets without such limits, like the US market. Therefore, the same settings may not always be equally effective. It is necessary to adjust the period and multiplier based on the stock's characteristics and the timeframe being analyzed.
Key Takeaways
Bollinger Bands are not a tool for predicting direction but for interpreting the state of volatility. Understand that a narrowing band width indicates volatility compression, and a widening band width indicates volatility expansion. Interpret band touches within the broader context of trend and volume, rather than in isolation.
This article is an educational overview explaining the principles of the indicator and does not recommend trading any specific stock. Any indicator works more reliably when used in conjunction with cross-verification from multiple sources and applied alongside risk management principles such as stop-loss criteria and position sizing.
Related reading
Get the app
Your watchlist and portfolio, one tap from your home screen — the Margin Call app