What the bands are
Bollinger Bands are three lines drawn on a price chart. The middle one is a moving average. The other two sit above and below it, set at a distance that depends on how much the price has recently been varying. When prices swing widely the bands spread apart; when prices are calm they pull together.
That makes the indicator a type of volatility envelope: a shape that wraps around price and changes width as conditions change. It was developed by the analyst John Bollinger, and the common default settings are a 20-period average with bands two standard deviations away.
How the bands are built
The calculation has three steps.
- Middle band. Take the simple moving average of the closing price over N periods. The default N is 20.
- Deviation. Calculate the standard deviation of those same N closing prices. Standard deviation measures how far prices typically sit from their average. A larger number means more spread.
- Outer bands. Add a multiple of the deviation to the middle band for the upper line, and subtract it for the lower line. The default multiple is 2.
In short: upper = average + 2 x deviation, lower = average - 2 x deviation.
Because the deviation is recalculated every period, the bands move continuously. The idea of standard deviation as a measure of spread is closely related to realized volatility, the measured size of recent price swings.
A small worked example
Real charts use 20 periods. To keep the arithmetic easy to check, imagine four closing prices of 9, 11, 9 and 11, in some currency unit.
- Average: (9 + 11 + 9 + 11) / 4 = 40 / 4 = 10.
- Differences from the average: -1, +1, -1, +1.
- Squared: 1, 1, 1, 1. Their average is 1.
- Standard deviation: the square root of 1, which is 1.
- Upper band: 10 + 2 x 1 = 12. Lower band: 10 - 2 x 1 = 8.
The bands sit at 8 and 12 around a middle line of 10. If prices were instead 5, 15, 5 and 15, the average would still be 10 but the deviation would be 5, and the bands would be at 0 and 20. Same middle, far wider envelope. (This example divides by the number of values; some software divides by one less, which changes the result slightly. Charting tools usually state which they use.)
What band width shows
Because the distance between the bands depends on deviation, the gap itself is a volatility gauge. One common measure is bandwidth: the upper band minus the lower band, divided by the middle band. In the first example that is (12 - 8) / 10 = 0.4.
Traders often talk about a “squeeze” when bandwidth is unusually low relative to its own recent history. All a squeeze says is that volatility has recently been low. Volatility does tend to come in clusters, with quiet and active stretches, but a squeeze does not say whether the next move will be up or down, when it will come, or whether it will be large.
What the bands do not mean
The most common misunderstanding is to treat the outer bands like walls. Price touching the upper band is not a sell signal, and touching the lower band is not a buy signal. Several points are worth knowing.
- In a strong trend, price can travel along or outside a band for many periods. This is sometimes called “walking the band”.
- Two standard deviations does not enclose a fixed percentage of prices in practice. That percentage figure comes from the normal distribution, and crypto returns often have fatter tails, meaning large moves occur more often than that model suggests.
- The bands are built from past closes, so they lag. They describe recent price variation; they do not forecast it.
- Different settings give different pictures. A shorter average makes the bands twitchier; a larger multiple makes them wider.
Using the bands as a learning tool
Used cautiously, Bollinger Bands are a helpful way to see price volatility on the chart itself, rather than as a separate number. Comparing the bands across periods shows when markets were calm and when they were turbulent. Some people combine them with a momentum reading from an indicator like the relative strength index, though combining indicators built from the same price data can give a false sense of confirmation.
If you want to test any rule based on the bands, do it on past data that you did not use to design the rule, and include fees and price slippage, since both reduce real-world results.
Cryptocurrency prices are volatile, past behaviour does not predict future results, and people can lose all of their money. This article is educational and is not financial advice.