Bollinger Bands Explained

Bollinger Bands are a volatility indicator consisting of a moving average with two bands plotted a set number of standard deviations above and below it. Developed by John Bollinger in the 1980s, they expand when price becomes more volatile and contract when it becomes calmer, so the distance between the bands is a direct visual measure of how much the market has been moving. Their value lies in showing whether price is extended relative to recent behaviour and in flagging periods of unusually low volatility that often precede large moves.

What Bollinger Bands measure

The bands measure volatility relative to a moving average. Standard deviation is a statistical measure of how widely prices are dispersed around their average; a large standard deviation means recent closes have been far from the average, a small one means they have been tightly clustered. Because the bands are set at a multiple of this figure, roughly 95% of closing prices in a normally distributed sample would fall within two standard deviations of the mean.

Price is not normally distributed, so that percentage is an approximation rather than a rule. The practical point is that a close outside the bands is unusual relative to recent activity, and a narrowing of the bands means the market has become quieter than it has been.

How they are calculated

With the standard settings:

1. Middle band = 20-period simple moving average of closing prices
2. Standard deviation = the standard deviation of those same 20 closes
3. Upper band = middle band + (2 × standard deviation)
4. Lower band = middle band − (2 × standard deviation)

To compute the standard deviation, take each of the 20 closes, subtract the 20-period average, square the result, average those squared differences, and take the square root. Charting platforms do this automatically. A related measure, bandwidth, is (upper band − lower band) ÷ middle band and allows the width of the bands to be compared over time.

Another derived value, %B, shows where price sits within the bands: %B = (price − lower band) ÷ (upper band − lower band). A %B of 1.0 means price is at the upper band, 0.5 means it is at the middle, and readings above 1 or below 0 mean price has closed outside the bands.

Common settings

The 20-period, 2-standard-deviation configuration is the default on the JDGlobalFX terminal and almost every other platform. Bollinger himself recommended keeping the period and multiplier in proportion: shorter periods with a smaller multiplier, longer periods with a larger one.

Common interpretations

The squeeze

When the bands contract to their narrowest width in a meaningful lookback, such as the last 100 or 120 candles, volatility has compressed. Markets alternate between quiet and active phases, so a squeeze frequently precedes a strong directional move. The squeeze does not indicate direction. Traders typically wait for a close outside the bands after a squeeze, or for a break of a nearby structural level, and trade in that direction with a stop on the other side of the compression range.

Band walks in trends

In a strong trend, price can close at or beyond one band repeatedly, a behaviour known as walking the band. This is a sign of trend strength, not exhaustion. Pullbacks in a band walk often reach only the middle band before the trend resumes, which makes the 20 SMA a reference for continuation entries.

Mean reversion in ranges

In a sideways market, price tends to oscillate between the bands and revert toward the middle. A tag of the lower band near horizontal support, with a reversal candle, is a common mean-reversion setup, with the middle band or the opposite band as targets. The setup is only sound if the market is genuinely ranging, which must be judged from price structure rather than from the bands themselves.

Closes outside the bands

A close beyond a band means price is more than two standard deviations from its 20-period average. In a range this often marks an extreme; in a trend it often marks a breakout or an acceleration. Context determines which.

A worked example

Suppose AUD/USD has spent two weeks in a narrowing range on the 4-hour chart and the Bollinger bandwidth reaches its lowest level in 120 candles. The bands sit at 0.6620 (lower) and 0.6660 (upper). A candle then closes at 0.6672, above the upper band, and the bands begin to widen. A trader enters long at 0.6675 with a stop at 0.6615, below the lower band and the compression range (60 pips of risk). The target is set at a prior swing high at 0.6795, giving 120 pips of potential reward, a 2:1 ratio. If price closes back inside the bands within the next two candles, the breakout has likely failed and the trader exits early.

Limitations and false signals

Band touches are not signals. The most common misuse is selling every touch of the upper band and buying every touch of the lower band. In trends this loses repeatedly, because band walks are precisely where the strongest moves occur.

The squeeze gives no direction. A breakout from a squeeze can be a false breakout in one direction followed by the real move in the other, a pattern sometimes called a head fake. Waiting for a close beyond the band and beyond a structural level reduces but does not eliminate this.

Lag from the moving average. The middle band is a 20-period SMA and carries the same lag as any moving average. The bands adjust to volatility more quickly than the average adjusts to direction.

The normal-distribution assumption is loose. Price series have fatter tails than a normal distribution, so closes outside the bands occur more often than the theoretical 5% and cannot be treated as rare events.

Parameter dependence. A 10-period band on a 15-minute chart and a 50-period band on a daily chart behave very differently. Signals depend on the settings chosen.

Because Bollinger Bands measure volatility, they pair naturally with a momentum tool such as RSI or a trend tool such as MACD that provides the information they lack. They are most valuable for identifying when a market is compressing or extended, with entries and stops then defined by structure and candlestick behaviour as described in Price Action Trading.

Key takeaways

• Bollinger Bands consist of a 20-period SMA with bands two standard deviations above and below, so their width tracks recent volatility.
• Narrow bands (a squeeze) show compressed volatility that often precedes a strong move; wide bands show an active market.
• In trends price can walk along a band for extended periods, and touching a band is not a reversal signal.
• Mean-reversion trades between the bands only make sense when price structure confirms a range.
• Derived measures such as bandwidth and %B quantify band width and price position for comparison over time.
• Pair the bands with momentum or trend tools and with structural levels; they measure volatility, not direction.