Bollinger Bands

Bollinger Bands are a technical analysis indicator that places a band above and below a moving average of an asset's price. The bands are set a chosen number of standard deviations away from the average, usually two, so they widen when prices are volatile and narrow when prices are calm. Traders use them to judge whether a price is relatively high or low compared with its recent range and to spot changes in volatility, usually as part of wider trading strategies.

The indicator was developed by financial analyst John Bollinger in the early 1980s and is one of the most widely used tools on charting platforms for shares, currencies, commodities and cryptocurrencies. It is usually drawn on a bar chart or candlestick chart alongside other indicators such as the relative strength index. Bollinger registered the name as a trademark and has written extensively on how the bands should and should not be used.

Imagine walking a dog on an extending lead. Most of the time the dog stays fairly close to you, but sometimes it pulls out to the end of the lead before coming back. On a quiet street the lead stays short; in a busy park the dog roams further and the lead stretches out.

Bollinger Bands are like that lead. The middle line is where you walk, the average price. The bands show how far the price usually wanders from the average. When the market is calm, the bands come in close; when it is wild, they spread out. A price touching the outer band is like the dog at the end of the lead: unusual, but not necessarily about to turn back.

With the common settings of a 20-period average and two standard deviations, the three lines are:

• $\text{SMA}_{20}$ is the simple average of the last 20 closing prices.
• $\sigma_{20}$ is the standard deviation of those 20 closing prices.

Suppose a share's last 20 closes average $50.00 and their standard deviation is $1.50. The middle band is $50.00, the upper band is $50 + 2 \times 1.5 = 53$ and the lower band is $50 - 2 \times 1.5 = 47$. If today's close is $52.40, it sits near the upper band. A related measure, %b, shows where the price is within the bands: $(52.4 - 47) \div (53 - 47) = 0.9$, meaning 90% of the way from the lower to the upper band. Bandwidth, $(53 - 47) \div 50 = 0.12$, measures how wide the bands are relative to the average.

Common interpretations include:

• Relative highs and lows. Prices near the upper band are high relative to the recent range, and near the lower band relatively low; Bollinger stresses that this alone is not a buy or sell signal.
• The squeeze. When bandwidth falls to unusually low levels, volatility has been compressed, and traders watch for a breakout, though the direction is not indicated.
• Walking the band. In strong trends, prices can move along the upper or lower band for extended periods, so touching a band is not a reversal signal on its own.
• Patterns. Analysts look for double bottoms in which the second low stays inside the lower band, or tops that fail to reach the upper band, as signs of weakening momentum.
• Mean reversion. In range-bound markets, some traders buy near the lower band and sell near the upper, accepting the risk that a range may break.

Bollinger recommends combining the bands with indicators based on different information, such as volume or momentum, rather than with tools that measure the same thing.

Crypto traders use Bollinger Bands on everything from one-minute to weekly charts. Because crypto prices can move far more than share prices, the bands are often very wide, and prices frequently move outside them during sharp rallies or crashes. Some traders adjust the settings, for example using different periods or standard deviation multiples, to suit round-the-clock trading. Squeezes in bitcoin's daily chart are widely discussed on social media before large moves, though many squeezes resolve without a dramatic breakout. As with any indicator, results depend on the time frame and market.

Bollinger Bands describe past prices and are not predictive on their own. The standard deviation assumes that price changes are roughly normally distributed, but markets show fat tails and sudden jumps, so moves beyond the bands happen more often than the statistics imply. The choice of settings is arbitrary, and tuning them to past data risks overfitting, a problem discussed under backtesting. In strongly trending markets, treating band touches as reversal signals can lead to repeated losses, as in a bull trap or bear trap. The bands are best seen as a way to put prices and volatility in context rather than as a trading system.