John Bollinger
The Bollinger squeeze: a signal that tells you when, and refuses to tell you which way
Volatility is more predictable than direction. That single asymmetry is what the squeeze exploits: when Bollinger BandWidth reaches a multi-month low, a large move becomes likely, and nothing about the setup indicates which way. Traders find that intolerable and supply a direction anyway — which is why the head fake, a first break that reverses, is the most documented behaviour of this pattern.
- Style
- Breakout
- Approach
- Mechanical
- Difficulty
- Intermediate
- Horizon
- Swing (days to weeks)
- Holding period
- Days to weeks
- Time needed
- 20 minutes a day
- Markets
- Single stocks · FX · Futures
- Source
- Bollinger on Bollinger Bands — John Bollinger
The rule set
- Compute Bollinger BandWidth — the distance between the bands as a fraction of the middle band — and flag readings near their own multi-month lows
- Take no position during the squeeze itself; the compression is a state, not a signal
- Enter on a close beyond a band accompanied by expanding BandWidth
- Place the stop on the far side of the squeeze range, not at the middle band
- Take profits as BandWidth reaches the high end of its range and price returns toward the middle band
What makes it distinctive
- It turns 'low volatility precedes high volatility' — one of the few genuinely reliable regularities in markets — into something executable
- The squeeze itself is measurable rather than visual: BandWidth at the low end of its own history is a number, not an impression
- Direction comes from the breakout rather than from a forecast, so nothing has to be predicted in advance
When it works
Instruments where compressed volatility meets an actual catalyst — an earnings report, a macro release, a policy decision. The compression supplies the coiled spring and the catalyst supplies the release.
When it fails
The first break out of a squeeze is frequently false and reverses straight back through the range — Bollinger named this the head fake. Without a catalyst, an instrument can remain compressed for far longer than a trader's patience lasts.
How a decision moves through it
Input
Daily closing prices
Closes only. Standard deviation is computed on closes, which is what makes the bands respond to sustained repricing rather than to intraday range.
Measure
Bollinger Bands: 20-period SMA ± 2 standard deviations
Bollinger's own defaults, and unusually well-chosen: 20 periods is about a month of trading, and 2 standard deviations captures roughly 89% of observations for a 20-period sample.
Measure
BandWidth — the compression measurement
Upper minus lower, divided by the middle. It normalises the band distance so readings are comparable across time and across instruments.
Decide
Is BandWidth near its own multi-month low?
Relative to its own history, never to an absolute threshold. A quiet currency pair and a volatile small cap have completely different BandWidth ranges.
Act
Trade the expansion, in whichever direction it comes
The entry is agnostic about direction by construction, and that is the part traders most consistently override.
The one regularity this is built on
Volatility clusters and cycles. Quiet periods are followed by active ones and active periods by quiet ones, far more reliably than price direction persists. This is one of the most robust empirical findings in finance and it holds across markets, instruments and decades.
That gives an unusual kind of signal: one that predicts the magnitude of what is coming without predicting its sign. Most trading tools claim the opposite and deliver neither.
Everything difficult about trading this follows from the missing direction. A setup that says 'something big is coming and I don't know which way' offers no story, no conviction and nothing to be right about in advance — and traders resolve that discomfort by inventing a bias.
BandWidth makes the squeeze a number
Bollinger defined BandWidth as the distance between the upper and lower bands divided by the middle band. Dividing by the middle band is the important part — it turns an absolute distance into a proportion, so a reading is comparable across price levels and across instruments.
A squeeze is then defined relative to that instrument's own BandWidth history: the lowest reading in six months is Bollinger's commonly cited threshold. There is no universal number, and any teaching that gives one has skipped the normalisation that makes the measure work.
Bollinger's own warning about band touches
Bollinger has stated repeatedly that a touch of the upper band is not a sell signal and a touch of the lower band is not a buy signal. The bands are a relative measure of high and low, not levels.
This matters here because the squeeze system does the opposite of the popular reading: it buys a close above the upper band rather than fading it. In a strong move price walks the band for weeks, and a trader who has learned to fade band touches will spend that entire move on the wrong side.
Five ways into this system
- Finding a squeeze, waiting through it, and the head fake at the endIdentifying the squeeze is mechanical. Everything difficult happens in the few days after it ends.8 min read
- Sizing for a setup whose first signal fails oftenThe squeeze produces the tightest stop of any setup in this library, and that is exactly what makes it dangerous.6 min read
- A squeeze needs a catalyst, and catalysts have calendarsCompression alone produces a coiled spring with nothing to release it. The best applications of this setup pair a squeeze with a scheduled event.5 min read
- Squeezes that never resolve, and the direction you supplied yourselfTwo mechanical failures and one human one. The human one does more damage than both mechanical failures combined.6 min read
- Bollinger Bands for beginners: what the bands measure and what a squeeze meansTwo of the most commonly misused lines on any chart, explained properly — including the mistake almost every beginner makes with them first.6 min read
The ideas behind it
This system assumes you already know these. Each one is explained from scratch in Investing 101.
Compare with
- Keltner Channel BreakoutDraw a moving average with bands set by recent volatility, and trade closes outside the bands in the direction the average is already pointing.
- RSI(2) Mean Reversion (Connors)Inside a long-term uptrend, buy the sharpest short-term pullbacks and sell into the bounce a few days later.
These are documented methods described for study. Nothing here is investment advice, a recommendation, or a claim about future returns — every system on this page has losing periods, and the pages say where.
Reading about a system is not having one.
Plutux is where you write your own rules down, test them against real data, and keep the record your memory would otherwise rewrite. Join the waitlist for early access.