Bollinger Squeeze: No Direction In It
A Bollinger squeeze is band width at a multi-month low, marking unusually quiet trade. Quiet stretches do tend to precede active ones, which is the sound half of the idea. But the reading has no direction in it, so the resolution is as likely to go down as up.
How it works
A Bollinger squeeze is a statement about the distance between two lines. Bollinger Bands sit a fixed number of deviations either side of a moving average, so quiet trade pulls them inwards.
The measurement is band width, and it is a ratio. Upper band minus lower band, divided by the middle band. That division makes the reading comparable across price levels.
The squeeze itself is a percentile, not a fixed threshold. No universal number marks a market as compressed; the condition is band width at its lowest in months.
The claim underneath is volatility clustering. Quiet periods tend to be followed by active ones, and active by more activity. That part is genuinely well established.
The gap between quiet and active is wide enough to matter. On this site’s shared 576-bar history the 14-bar average true range has a median of 0.5994, a tenth percentile of 0.2823 and a ninetieth of 0.7954 — a ratio of 2.82.
What it will not tell you
And it says nothing whatever about direction. The bands narrowing is a statement about volatility, and volatility has no sign. A standard deviation is built from squared distances, which discards the sign of every move.
Which means half the breakouts go the other way. That is the indicator working correctly. The popular presentation shows a squeeze followed by a large rise and omits the mirror image.
Participation on the break is the one filter worth adding. If volume expands as price leaves the compressed area, more people are involved than during the quiet stretch. It is a weak filter with a real mechanism.
In practice
On a daily chart a squeeze can last for weeks. Most of the loss comes from the false starts, not the resolution. Entering early, repeatedly, is how traders reach the real move already down.
And the resolution frequently arrives as a gap. Compression often breaks overnight, so the gap opens beyond the band and the intended entry never trades. Wait for a close, not a touch.
The other side of the range is the only clean stop. A squeeze defines one level worth using: the far edge of the compressed area. Any other stop is a number you chose.
Every false break costs a share of a bar. Round-trip cost here is 0.0098 price units, 2% of a median bar’s range and 45% of the smallest bar.
The bands are computed and nobody is defending them. A band is a moving average plus a multiple of recent deviation. Nobody holds capital at a number that moves every bar.
Why the level you trade is not the band
A band break and a range break are different pieces of evidence. Price crossing a prior high says something about orders resting where people watched. Price crossing a band says only that recent movement exceeded typical movement.
So treat the band as the measurement and the range edges as the levels. The high and low of the compressed area are visible to every chart reader whatever their settings, which is why a breakout from those edges carries more information than a band touch.
Dropping to a faster chart does not create information; it multiplies attempts. The same compression appears on every interval, each offering its own false starts at the same fixed cost.
The squeeze says when to start watching; the range says what to watch. In that order the indicator does the job it can do, and the level comes from structure rather than a formula.
What a Bollinger squeeze is not
- Not a directional signal. It has no sign, and no version of it ever will.
- Not a fixed threshold. It is a percentile of that instrument’s own band width.
- Not a timing tool. It says change is likelier, not that it happens next bar.
- Not a standalone strategy. Without a directional thesis it is a coin flip carrying costs.
When it fails
- In a range, the squeeze resolves back into the range. Volatility expands, price pokes out, and the market carries on between the same two edges. Nothing in the reading separates that from a genuine departure.
- When the break does not hold its level. On this series a breakout above a 20-bar high closed back below the level within ten bars in 85% of 39 events, and above a 55-bar high in 100% of 11 events.
- When the market is already trending. Band width contracts during steady advances too, and a squeeze inside a trend often resolves as a pause, not an acceleration.
- When you scale down to force more signals. Costs stay fixed while bar ranges shrink, and round-trip cost already exceeds 10% of a bar’s range on 15 of the 576 bars here.
- When the resolution gaps past the entry. You chase at a worse price or skip the trade, so a real share of resolutions are never tradeable on the terms implied.
The original data
The measurements come from research/series-measurements.json. Bar ranges span 0.17 to 1.101
between the tenth and ninetieth percentiles, a ratio of 6.5; the smallest is 0.022, the largest
2.338, the median 0.493.
Two figures from the same file describe follow-through. The ten-bar efficiency ratio has a median of 0.34, with 30% of bars above 0.5, and the base rate for a higher close ten bars later is 54% across 566 observations.
A median efficiency ratio of 0.34 means the typical ten-bar stretch gives back most of the ground it covers. Sustained direction is the exception here, which is when a false breakout becomes the likelier resolution.
Coverage in research/corpus-coverage.json shows how thinly this is taught. Across 31,760
videos, “bollinger squeeze” appears in 4 videos, median 4,379 views, 2 channels. Before the next
squeeze break, write the far edge of the compressed area on the chart and size the trade from that
stop.
Related
Bollinger Bands are the construction the squeeze is measured from, so the settings you pick define what counts as compressed. Read it first if you have never set that multiple yourself. The squeeze inherits every assumption the bands make.
Breakout is the event a squeeze is supposed to precede. That page separates a break that holds from one that does not. It is more useful still if you intend to trade the break.
False breakout is the outcome to expect by default here. It sets out the structure a failed break leaves behind. Reading it is worth more than spotting the squeeze.
I stopped trading squeezes on their own years ago, and it was not because they never worked. It was because I could never tell you in advance which way the thing was going to break, and I kept quietly pretending I could. Now a squeeze is just a note that conditions are about to change. The direction has to come from somewhere else.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money. Some links on this page earn a commission if you buy through them. It costs you nothing and it does not decide what appears here or in what order — how these pages are made is set out in our methodology.