Bollinger Bands: How to Use Them (And When Not To)

Bollinger Bands are one of the most widely used technical indicators in the world, yet they are also one of the most frequently misapplied. Understanding what Bollinger Bands actually measure, what they signal, and — critically — what they do not signal is essential for any trader who uses them.

1. Origin and Construction

Bollinger Bands were developed by John Bollinger in the early 1980s. Bollinger, a Chartered Financial Analyst (CFA) and Chartered Market Technician (CMT), was looking for a way to create adaptive trading bands that responded to changing market volatility, unlike fixed-percentage envelopes that were common at the time. He received a U.S. trademark for “Bollinger Bands” in 2011.

The construction is straightforward. A Bollinger Band consists of three lines:

• Middle Band: A 20-period simple moving average (SMA) of closing prices.
• Upper Band: The middle band plus 2 standard deviations of closing prices over the same 20-period window.
• Lower Band: The middle band minus 2 standard deviations of closing prices over the same 20-period window.

Mathematically:

Middle Band = SMA(Close, 20)
Upper Band  = SMA(Close, 20) + 2 * StdDev(Close, 20)
Lower Band  = SMA(Close, 20) - 2 * StdDev(Close, 20)

Why 20 and 2?

Bollinger chose the 20-period lookback as a compromise between responsiveness and smoothness. A shorter window (such as 10 periods) would make the bands more responsive to recent price changes but also more noisy. A longer window (such as 50 periods) would produce smoother bands but would be slow to adapt to changing conditions. Twenty periods represents approximately one month of trading days, which Bollinger found to be a practical balance.

The choice of 2 standard deviations is grounded in basic statistics. If price returns were normally distributed, approximately 95% of observations would fall within 2 standard deviations of the mean. In practice, stock price returns are not normally distributed — they exhibit fat tails (excess kurtosis) and sometimes skewness. This means that prices will fall outside the bands more often than the 5% that a normal distribution would predict. Bollinger himself acknowledged this: the bands are a practical approximation, not a precise statistical boundary.

2. Bandwidth: Measuring the Squeeze

One of the most useful derivatives of Bollinger Bands is Bandwidth, which measures how wide or narrow the bands are relative to the middle band:

Bandwidth = (Upper Band - Lower Band) / Middle Band

Bandwidth is a direct measure of realized volatility. When volatility is high, the bands widen and Bandwidth increases. When volatility is low, the bands narrow and Bandwidth decreases. This is the key insight that makes Bollinger Bands adaptive: unlike fixed-percentage envelopes, the bands automatically expand and contract in response to market conditions.

The Squeeze

A Bollinger Band Squeeze occurs when Bandwidth contracts to a low level relative to its recent history. This signals a period of unusually low volatility, which Bollinger argued is often a precursor to a significant move in either direction. The logic is rooted in the empirical observation that volatility is mean-reverting: periods of low volatility tend to be followed by periods of higher volatility, and vice versa.

The Squeeze does not predict the direction of the subsequent move — only that a move is likely. This is a critical distinction that many traders miss. A Squeeze followed by a breakout above the upper band suggests upside momentum. A Squeeze followed by a breakdown below the lower band suggests downside momentum. The Squeeze itself is direction-neutral.

Identifying a Squeeze typically involves looking for Bandwidth at a multi-month or multi-year low. Some traders define the Squeeze as Bandwidth falling below a specific percentile of its rolling distribution (e.g., below the 5th percentile of the last 120 periods). Others use the simpler approach of visually identifying the narrowest bands on the chart.

3. %B: Where Is Price Within the Bands?

Bollinger %B quantifies the position of the current price relative to the upper and lower bands:

%B = (Price - Lower Band) / (Upper Band - Lower Band)

The interpretation is intuitive:

%B is useful for systematizing Bollinger Band signals. Instead of eyeballing whether price is “near” the upper or lower band, you can define precise thresholds: buy when %B drops below 0.0, sell when %B rises above 1.0, and so on. It also facilitates backtesting, since %B can be computed programmatically for any price series.

4. Mean-Reversion Strategy

The most common way to trade Bollinger Bands is as a mean-reversion indicator: buy when price touches or falls below the lower band, sell (or short) when price touches or rises above the upper band. The logic is that price has moved an unusually large distance from its average and is likely to revert.

This strategy works well in range-bound, sideways markets where price oscillates around a relatively stable mean. In these conditions, touches of the upper and lower bands represent temporary extremes that tend to reverse. The trader buys the dip, sells the rip, and collects the spread.

However, mean-reversion Bollinger Band strategies fail badly in trending markets. In a strong uptrend, the price can remain at or above the upper band for extended periods. A trader who shorts every touch of the upper band in a bull market will be run over by the trend. Similarly, in a strong downtrend, the price can remain at or below the lower band indefinitely, and buying every touch of the lower band will generate a series of losing trades.

5. Breakout Strategy

The opposite approach uses Bollinger Bands as a breakout indicator: buy when price breaks above the upper band (especially with expanding volume), because this signals the start of a strong uptrend. Short when price breaks below the lower band for the opposite reason.

This strategy works well in trending markets, particularly when the breakout follows a Squeeze. The logic is that the period of low volatility (Squeeze) has created a coiled spring, and the breakout above or below the bands signals that the spring has been released. The trader rides the trend for as long as it persists.

Breakout strategies fail in range-bound markets, where the price briefly pokes above or below the bands and then reverses. These false breakouts (or “head fakes,” as Bollinger calls them) generate whipsaw losses for the breakout trader.

6. Walking the Band

In strong trends, the price can “walk” along the upper or lower Bollinger Band for extended periods. During a powerful bull move, the price repeatedly touches or exceeds the upper band, pulls back briefly to the middle band or just above it, and then pushes back to the upper band. The lower band is never touched. This pattern — walking the upper band — is a signature of strong upward momentum.

The reverse occurs in bear markets: the price walks along the lower band, with brief rallies to the middle band followed by renewed selling.

Walking the band is critical to understand because it represents the scenario where mean-reversion Bollinger Band strategies are most dangerous. A trader who sells every time the price touches the upper band during a band-walk will accumulate a series of losses as the trend continues. Bollinger himself has explicitly warned against treating band touches as automatic sell signals for exactly this reason.

7. Combining with Other Indicators

Because Bollinger Bands alone cannot distinguish between trending and ranging conditions, they are most effective when combined with other indicators that provide this context:

Volume Confirmation

A breakout above the upper band accompanied by a surge in volume is more likely to be a genuine trend initiation than a breakout on thin volume. Volume confirmation helps filter out false breakouts and increases the probability that a breakout trade will follow through.

RSI Divergence

When price touches the lower band but RSI (Relative Strength Index) shows a higher low (positive divergence), the mean-reversion signal is strengthened: price is at an extreme but selling momentum is waning. Conversely, when price touches the upper band but RSI shows a lower high (negative divergence), the sell signal is strengthened.

ADX for Trend Strength

The Average Directional Index (ADX) measures trend strength regardless of direction. When ADX is below 20, the market is generally range-bound and mean-reversion Bollinger strategies are more appropriate. When ADX is above 25, the market is trending and breakout strategies are more appropriate. Using ADX as a filter can help resolve the mean-reversion vs. breakout contradiction.

Keltner Channel Squeeze

Some traders define the Bollinger Band Squeeze more precisely by comparing Bollinger Bands to Keltner Channels (which are based on ATR rather than standard deviation). When the Bollinger Bands contract inside the Keltner Channels, a Squeeze is in effect. When the Bollinger Bands expand outside the Keltner Channels, the Squeeze has “fired” and a directional move is underway. This approach, popularized by John Carter, provides a more precise Squeeze definition than Bandwidth alone.

8. Double Bottoms and M-Tops

Bollinger described specific price patterns that are most significant when they occur at the bands:

W-Bottoms

A W-bottom (double bottom) that forms at or below the lower band is a classic Bollinger Band reversal pattern. The first low touches or penetrates the lower band. The price rallies to or above the middle band. The second low forms near the lower band but does not penetrate it as deeply as the first low. This pattern, combined with a %B reading that is higher on the second low than the first, suggests that selling pressure is exhausted and a reversal is likely.

M-Tops

The reverse is an M-top (double top) at or above the upper band. The first high touches or exceeds the upper band. The price pulls back to or below the middle band. The second high approaches the upper band but does not exceed it as far as the first high. This suggests that buying pressure is waning and a reversal is likely.

These patterns are more reliable when confirmed by volume (declining volume on the second peak/trough) and momentum (RSI divergence).

9. Academic Evidence: Do They Work?

The academic evidence on Bollinger Band trading strategies is mixed. This is not surprising given the inherent contradiction between mean-reversion and breakout applications, and the dependence on market regime for either to work.

Lento and Gradojevic (2007) tested Bollinger Band trading strategies in a paper published in Applied Financial Economics Letters. They examined simple mean-reversion rules (buy when price crosses below the lower band, sell when price crosses above the upper band) across multiple markets. Their results were mixed: the strategies were profitable in some markets and time periods but not others, and profitability was sensitive to the parameter choices (lookback period, standard deviation multiplier) and the market conditions during the test period.

This finding is consistent with the broader academic literature on technical analysis. Technical indicators that depend on market regime (trending vs. ranging) tend to perform well in some periods and poorly in others. Without an independent method of identifying the current regime, the indicator alone is insufficient for generating consistent profits.

Bollinger himself has been clear that the bands are a tool for analysis, not a mechanical trading system. Tags of the upper and lower bands are not automatic buy or sell signals. The bands provide context about volatility and relative price levels, which must be combined with other analysis (fundamental, technical, or both) to generate actionable trading decisions.

10. Common Mistakes

Several common errors arise from misunderstanding what Bollinger Bands do and do not signal:

• Treating band touches as signals. A touch of the upper band does not mean “sell.” A touch of the lower band does not mean “buy.” Bollinger has explicitly stated this repeatedly. Band touches indicate that price is at a relative extreme, but whether that extreme represents a reversal or a trend continuation depends on the broader context.
• Ignoring the regime. Applying a mean-reversion strategy in a trending market (or a breakout strategy in a ranging market) will generate losses. The regime determination is more important than the Bollinger Band signal itself.
• Over-optimizing parameters. Fitting the lookback period and standard deviation multiplier to historical data is a recipe for overfitting. The default parameters (20, 2) are not optimal for every market and timeframe, but chasing the “best” parameters on historical data will not produce robust future results.
• Assuming normal distribution. The 2-sigma bands do not capture 95% of price action because returns are not normally distributed. Fat tails mean that extreme moves (beyond the bands) occur more frequently than a Gaussian model would predict. In volatile or crisis markets, this difference can be dramatic.
• Using Bollinger Bands in isolation. As a standalone indicator, Bollinger Bands provide volatility context but not direction. They are most effective when combined with volume analysis, momentum oscillators, or trend-strength indicators that provide the additional information needed to generate a directional view.

11. Practical Implementation

For traders who want to incorporate Bollinger Bands into their workflow, here is a practical framework:

1. Identify the regime. Use ADX, visual trend analysis, or a moving-average slope to determine whether the market is trending or ranging. This is the most important step.
2. In ranging markets, look for mean-reversion setups: buy near the lower band with RSI divergence, sell near the upper band with RSI divergence. Use tight stops because the range can break at any time.
3. In trending markets, look for pullbacks to the middle band as buying opportunities (in an uptrend) or shorting opportunities (in a downtrend). Do not fade the trend by selling at the upper band or buying at the lower band.
4. Watch for Squeezes. When Bandwidth contracts to an extreme low, prepare for a directional move. Wait for the breakout to confirm direction before entering, and use volume to validate the breakout.
5. Never use Bollinger Bands as your sole indicator. Combine them with volume, momentum, and trend-strength analysis to build a more complete picture.

Bollinger Bands are a powerful tool for understanding volatility and relative price levels. They are not a trading system. Used correctly, they provide valuable context that improves decision-making. Used mechanically, without understanding their limitations, they will generate as many losses as profits.