Bollinger Bands
Understand Bollinger Bands in crypto: moving-average centreline, standard-deviation bands, bandwidth, band walks, squeezes and volatility-regime limitations.
Reading progress — saved on this device
Bollinger Bands place volatility-scaled bands around a moving average. They show where price sits relative to its recent average and dispersion, but touching a band is not automatically overbought, oversold or a reversal signal.
How Bollinger Bands are constructed
A common configuration uses a 20-period simple moving average as the middle band, with upper and lower bands two standard deviations above and below it.
Upper = SMA(20) + 2 × standard deviation(20)
Lower = SMA(20) − 2 × standard deviation(20)
The exact coverage is not guaranteed to match a textbook normal distribution because crypto returns are not normally distributed, observations are serially dependent and volatility clusters.
Band width and price location
| Observation | What it may describe | What it does not prove |
|---|---|---|
| Bands contracting | Lower recent realised dispersion | That a breakout is imminent at a known time |
| Bands expanding | Higher recent realised dispersion | That the move will continue directionally |
| Price at upper band | Price high relative to recent mean/dispersion | That price is overvalued or must fall |
| Repeated upper-band touches | Strong upside trend / band walk possible | That every touch should be faded |
Squeeze, expansion and band walks
A squeeze is a period of unusually narrow bands. It describes low recent volatility, often before volatility later normalises or expands. Direction is not contained in the squeeze itself.
A band walk occurs when price repeatedly trades near one band during a sustained trend. In that regime, the band can behave more like a trend-strength reference than a reversal zone.
Practical Bollinger workflow
- Define lookback and deviation multiplier in advance.
- Separate two questions: where is price relative to the bands, and are the bands expanding or contracting?
- Classify trend/range conditions using price structure.
- If studying squeezes, define a quantitative bandwidth threshold rather than selecting visually impressive examples.
- Track what happens after expansion: continuation, reversal or two-sided volatility.
This avoids the common hindsight error of labelling only successful contractions as “squeezes” after the breakout is already visible.
Worked example: a band walk
BTC trades around £76,000 with 20-period bands at £72,500 and £77,500. A breakout lifts price to £78,000, the upper band expands to £79,200, and subsequent closes remain near the rising upper band for several sessions.
The first close above the band is not evidence that BTC must mean-revert immediately. The combination of rising centreline, expanding bands and repeated upper-band closes describes a high-momentum/high-volatility regime.
A reversal thesis would need additional evidence such as failed continuation, structural breakdown or a change in volatility behaviour.
Common mistakes and misunderstandings
- Assuming an upper-band touch means “sell” and lower-band touch means “buy”.
- Interpreting a squeeze as a directional forecast.
- Assuming two standard deviations imply a fixed 95% containment rate.
- Ignoring band walks in persistent crypto trends.
- Changing parameters until the bands fit historical turning points.
Knowledge checkpoint
Q1. Why can price remain near the upper band for several candles?
Q2. What information does a Bollinger squeeze contain about direction?
Q3. Why is “two standard deviations” not a guaranteed 95% containment rule for crypto prices?
Q4. How does changing the lookback or multiplier alter band behaviour?
FAQ
❓ Does touching the upper band mean overbought?
Not by itself. It means price is high relative to the chosen moving average and recent standard deviation. Strong trends can repeatedly touch or exceed the upper band.
❓ What is a Bollinger squeeze?
A period of relatively narrow bands, indicating low recent dispersion. It does not specify breakout direction or timing.
❓ Are 20 periods and 2 standard deviations mandatory?
No. They are common defaults. Different settings change sensitivity and should be defined before evaluation.
❓ Do Bollinger Bands predict volatility?
They primarily summarise recent realised volatility through standard deviation. A contraction can be useful context for future expansion research, but it is not a certainty.
Summary
- Bollinger Bands combine a moving average with volatility-scaled envelopes.
- Band width describes recent dispersion; price location describes distance from the centreline.
- Band touches are not automatic reversal signals.
- Squeezes describe compression but not future direction.
Technical analysis describes observed price, volume and volatility behaviour. It does not remove market, execution, liquidity or model risk, and its usefulness depends on data quality, timeframe and regime.
Want this in a personalised order?
Take the crypto assessment and get a custom path of 10 modules matched to what you already know. Free, no card required.
Build my path →