Bollinger Bands

Contributor Image
Written By
Contributor Image
Written By
Dan Buckley
Dan Buckley is an US-based trader, consultant, and part-time writer with a background in macroeconomics and mathematical finance. He trades and writes about a variety of asset classes, including equities, fixed income, commodities, currencies, and interest rates. As a writer, his goal is to explain trading and finance concepts in levels of detail that could appeal to a range of audiences, from novice traders to those with more experienced backgrounds.

Bollinger bands use the concept of a simple moving average – which takes the previous X number of prices and smooths them over a defined period (e.g., 20 periods/days, 50 periods/days) to see how far away the current price is from the average. This moving average is accompanied by two separate lines typically plotted two standard deviations above and below it.

bollinger bands

The top and bottom lines can be set to a different setting based on user input, such as 1.5 standard deviations or 3 standard deviations from the mean.

Bollinger bands using the standard configuration of a 20-period simple moving average and bands two standard deviations from the mean is known as a (20, 2) setting. Practically all trading software will allow you to adjust this configuration, including a change from a simple moving average to an exponential moving average.

The Best Brokers For Bollinger Band Analysis

CFDs are not available to residents in the United States.

Bollinger bands are commonly used as a “reversion to the mean” indicator. If price is below the Bollinger bands, it might be taken as an indication that price is currently too low. Likewise, if the price is above the bands, price may be interpreted as being too high. Bollinger bands are also commonly used as a volatility indicator. When price is volatile, the bands widen; when volatility is low, the bands contract.

Some traders may interpret the indicator in a different sense. If price is trading outside of the bands, but is trending in the general direction of the indicator – which is fundamentally just three separate but parallel moving averages – Bollinger bands may be considered a trend-following indicator.

The bands could also be viewed purely as a volatility indicator. When the bands contract due to a period of low volatility, commonly called a “squeeze”, it might indicate a forthcoming bout of high volatility. Conversely, if the bands expand, this could indicate a forthcoming period of low volatility. Those who require volatility or a trending market may likely close out trades or reduce positions during periods of band expansion.


Example #1

Here we have an hourly chart of the S&P 500 with Bollinger Bands set to the (20, 2) setting.

bollinger bands

Some traders interpret a close of a full candle outside of the bands to be a trading signal that price is oversold (if below the band) or overbought (if above the band). Thus trade opportunities may be biased in the opposite direction.

Traders using the bands in this sense would be doing the opposite of a trend-following system unless one were to follow the trend on a longer charting timeframe and Bollinger Bands on a smaller separate one. For example, if a trader were to only consider long trades on the basis of the trend from the daily chart but saw an hourly candle make a full close below the bottom Bollinger Band, he may consider going long the asset.

Example #2

If we use the same chart but set to the (20, 3) setting, this would produce more conservative signals for potential “reversion to the mean” opportunities. Naturally the bands are much wider. A three-standard deviation setting would theoretically accommodate 99.7% of the price data rather than just 95% of the price data.

bollinger bands

As one can see, the signals, based on candles leaving the bands, are few and far between.

Example #3

We can also shorten the moving average down to 10 periods. If we keep the standard deviation setting at 2 for a (10, 2) setting, we get the following:

bollinger bands

Given the period is smaller – moving average takes into account most recent 10 periods of price data rather than going back 20 periods in the case of the default – the bands are much more responsive to the current price. This means that price data is more likely to stay within the bands given the to-the-minute price is more heavily baked into the indicator’s formulation.

Example #4

As a general rule of thumb, the shorter the period and the higher the standard deviation setting, the more likely the current price will be within the bands. If we extend out the period to 200 and lower the standard deviation to just 1 (i.e., (200,1) setting as displayed below), the indicator has very little value.

bollinger bands

So much previous price data is integrated into the bands that the current price is just 0.5% of its calculation (compared to 5% for the 20-period default setting) and the standard deviation is so small that a large portion of data will naturally fall outside the bands’ constraints.


Bollinger bands are widely open to interpretation in terms of what they best communicate and vary in terms of how traders deploy them:

  1. In one sense, they represent a “reversion to the mean” indicator designed to show whether price might be overextended in one direction or another.
  2. In another, they could be interpreted as a trend-following indicator where a price outside of the bands indicates a “breakout”.
  3. Bollinger bands can also represent a clear visual of current volatility in the market, with contracted bands denoting low volatility (with the potential for a breakout) and wide bands denoting high volatility (with the potential for a following period of low volatility).

Bollinger bands on their own are not designed to be an all-in-one system. How one interprets them on a chart is very much dependent on the trader.

The indicator’s developer, John Bollinger, recommended using the indicator in conjunction with several other uncorrelated indicators to better interpret trend and volatility, such as the Relative Strength Index (RSI), moving average convergence divergence (MACD), and on-balance volume.

Bollinger Bands %B

Bollinger bands %B translates a portion of the price information in Bollinger bands into one line rather than the multiple bands you see with the standard indicator.

Percent B is intended to show where price is relative to each band.

%B is calculated as: (Price – Lower Band) / (Upper Band – Lower Band)

This can be translated as follows:

The basis behind the Bollinger bands %B indicator is to isolate the trend-following or overbought/oversold aspects of price that the standard Bollinger bands indicator identifies.


Below we have an hourly chart of the S&P 500 with the standard Bollinger bands setting – 20-period moving average and upper and lower bands set to two standard deviations above and below the mean (sometimes graphically depicted as a middle band, but not shown in this case).

bollinger bands %b

As can be observed, the Bollinger bands %B indicator provides the same information as the regular Bollinger bands indicator in terms of the difference between price and the top and bottom bands.

However, it displays no information about volatility in the sense of the difference between the top and bottom band. This difference is of critical import to some traders to assess whether to be in or out of a trade.

A tight distance between the top and bottom band may be indicative of a price breakout in one direction or another (also commonly called a “squeeze”), while a wide distance between the bands may indicate an upcoming period of relatively low volatility.


The Bollinger bands %B indicator quantifies the distance between price and the top and bottom bands, which can help technical analysts determine trends. Some traders also generate trading signals based on whether %B indicator is above 1 or below 0 and could entail either trend following or price reversal. It is nonetheless not recommended to use the %B indicator (or any indicator) in isolation.

To obtain the volatility component – which Bollinger bands %B doesn’t capture – one would either have to rely on the standard Bollinger bands, or use Bollinger BandWidth.