How are Bollinger Bands calculated?

Take the closing prices of the last 20 trading days and average them. That average is the middle band. Compute the standard deviation of those same 20 prices, multiply it by 2, then add it to the middle band for the upper band and subtract it for the lower band. Every new day, drop the oldest price, add the newest, and recompute.

What does a Bollinger Band squeeze mean?

A squeeze means volatility has fallen to an unusually low level, with bandwidth hitting roughly a 6-month low. Bollinger’s own observation was that bandwidth is mean-reverting, so quiet periods do not last and compressed volatility tends to expand. The squeeze tells you something is coming, not which way it goes.

Should you sell when price touches the upper Bollinger Band?

No, not on its own. Price can walk the upper band for many sessions during a strong trend, and fading a band walk is a common way to lose money. The upper band is not resistance. A close well outside the band is statistically extreme and some mean-reversion traders use it as a setup, but it needs confirmation from something else.

What settings should I use for Bollinger Bands?

The defaults are a 20-day window with a 2 standard deviation multiplier, which is what John Bollinger originally specified. Traders working shorter horizons sometimes use 10 periods with a 1.5 multiplier, and slower traders sometimes use 50 with 2.5. Drifting far from the defaults is usually a sign you are fitting the indicator to past data.

Bollinger Bands Explained: The Math and the Squeeze

Bollinger Bands are three lines drawn on a chart. The middle is a 20-day simple moving average. The upper band is two standard deviations above it. The lower band is two standard deviations below. That's the whole construction. John Bollinger published it in the 1980s, and it has become one of the three or four most-used indicators in retail technical analysis.

The way I think about Bollinger Bands is that they're a visualization of volatility, not direction. The bands widen when volatility increases and contract when it falls. The price wandering inside or outside the bands tells you something about how stretched the recent move is. They're not a buy signal. They're a context.

Plain English

Standard deviation measures how spread out a set of numbers is around its average. Two standard deviations is wide enough that, in a normal distribution, about 95% of values would fall inside. Stock returns aren't normally distributed, so the real number is closer to 88-90%, but the principle holds: most of the time, price stays inside the bands.

The Math, In Order

Step by step:

Take the closing prices for the last 20 trading days.
Average them. That's the middle band.
Compute the standard deviation of those same 20 prices.
Multiply the standard deviation by 2.
Add it to the middle band: that's the upper band. Subtract it: that's the lower band.

Each new trading day, you drop the oldest price and add the newest, recompute, and the bands shift. The 20-day window and the 2-sigma multiplier are Bollinger's defaults. Some traders use 10/1.5 for shorter horizons or 50/2.5 for slower ones.

What the Bands Are Actually Showing

Three things get talked about most:

The squeeze. When the bands narrow tightly around the moving average, volatility is unusually low. Historically, low-volatility periods tend to precede high-volatility moves. The squeeze doesn't tell you direction, just that something is coming.
The walk. When price “walks the band,” touching or hugging the upper or lower band for many sessions in a row, that's a strong trend. Normal interpretation: you don't fade a band walk. You go with it.
Reversion. When price closes well outside the band, the move is statistically extreme. Mean-reversion traders often look for these as setups for a snap-back to the moving average.

The Squeeze in Detail

John Bollinger's original observation was that bandwidth itself is mean-reverting. Periods of low volatility don't last forever. When bandwidth (the distance between upper and lower bands as a percent of the moving average) hits a 6-month low, you're in a squeeze. Compressed volatility tends to expand, and that expansion is the thing the squeeze is actually pointing at.

The squeeze doesn't tell you which direction. Most squeezes resolve in the direction of the prevailing trend, but enough resolve against it that you can't trade the squeeze alone. The pattern most traders use is: wait for a squeeze, wait for a breakout (price closes outside the band), then trade the breakout direction with a stop on the other side of the recent range.

The Honest Limitations

Bollinger Bands are a moving average plus volatility. They are not predictive. The lower band is not a support level. The upper band is not resistance. Trading bands is trading the past, dressed up in statistical language.

The signal-to-noise ratio is poor on most stocks at most timeframes. Most touches of the bands don't lead to anything actionable. Most squeezes don't produce big moves. Most band walks end without a clean exit signal. The indicator is honest about being a context tool. The trouble starts when traders treat it as a prediction tool.

How Bollinger Bands Pair With Other Indicators

Almost nobody serious uses Bollinger Bands alone. The usual stack:

Bands plus RSI. Price touching the lower band while RSI is oversold (under 30) is a more credible mean-reversion setup than either alone.
Bands plus volume. A breakout above the upper band on heavy volume is a real breakout. The same breakout on weak volume is often noise.
Bands plus price structure. The bands plus a horizontal support or resistance level give you a more reliable read than the bands floating in space.

Takeaway

Bollinger Bands are a 20-day moving average with two standard-deviation envelopes. They're a context indicator, not a prediction tool. The squeeze, the walk, and the extreme close are the three patterns worth knowing. Anything more than that is overfitting.

The Take

Bollinger Bands are useful for the same reason MACD is: a lot of other people are looking at them. The reflexivity is real. A widely-watched indicator generating a clean signal can move price simply because many traders react to the same pattern. Use the bands as a context overlay on a chart, not as a standalone trading system. The math is high-school statistics. The edge, if there is one, is in how you combine them with everything else you're looking at.

Bollinger Bands: The Indicator That’s Mostly Standard Deviation

Key takeaways