How to Trade With Bollinger Bands: Volatility, Not a Verdict

Bollinger Bands wrap price in a volatility envelope — a 20-period average with an upper and lower band set two standard deviations away — that shows whether price is stretched or calm relative to its own recent range. This page covers what the bands actually measure, the squeeze and the expansion, mean-reversion versus breakout use, and why a band touch is not a buy or sell signal, the same way an oscillator like RSI isn't. Research and education only — not financial advice.

Bollinger Bands are a volatility envelope, not a trade signal: a touch of the upper band means price is stretched and volatility is high — it never automatically means "sell." The bands describe conditions; the decision to trade still needs a catalyst, a level, and confirmation the bands cannot provide on their own. That distinction is the whole of sensible bollinger bands trading.

What Bollinger Bands actually are

John Bollinger built the tool in the 1980s from three lines. The middle band is a 20-period simple moving average of closing price. The upper and lower bands sit two standard deviations of those same 20 closes above and below the middle. Standard deviation is a measure of how spread out recent prices have been — so the bands are literally price plotted against its own recent dispersion.

Because their width is driven by standard deviation, the bands breathe. When price gets volatile the bands flare apart; when it calms they pinch together. That is the entire point of the design: an envelope that adapts to conditions rather than a fixed percentage drawn around a moving average. A 3% band would be too wide in a sleepy market and too tight in a wild one — Bollinger's bands solve that by re-measuring volatility on every bar.

You will hear that price spends "about 95% of its time" inside the bands. Under a normal distribution, ±2 standard deviations does contain roughly 95% of observations, so most closes land inside the envelope. But market returns have fat tails — extreme moves happen more often than a bell curve predicts — so treat 95% as a rule of thumb, not a law. Price can and does ride outside a band for extended stretches, which is exactly where the naive interpretation gets people hurt.

The squeeze and the expansion

The squeeze is the setup Bollinger Bands are most famous for. When bandwidth contracts to a multi-week or multi-month low, the bands visibly pinch — volatility has compressed. Bollinger's own observation is that volatility is mean-reverting even when price is not: quiet periods tend to be followed by active ones, and vice versa. So a squeeze flags that a directional expansion may be building up pressure behind a narrow range.

The squeeze tells you a move may be coming — never which way. Anticipating direction from a squeeze alone is guessing. Pre-positioning long or short inside a squeeze because "it's about to pop" is the single most common way traders misuse the pattern.

The expansion is the release: volatility fires, the bands flare apart, and price drives out of the range. The event breakout traders wait for is the first decisive close outside a band on rising volume, ideally with the middle band starting to slope in the same direction. But Bollinger himself named the classic trap — the head-fake: price pops one way out of a squeeze, sucks in the traders who were leaning that direction, then reverses and runs the other way. An initial poke that fails back inside the bands is a warning, not an entry.

Two derived readings worth knowing

%B places price on the band scale: 1.0 is sitting on the upper band, 0.0 on the lower, 0.5 on the middle. It turns "near the top" into a number.
Bandwidth is (upper − lower) ÷ middle — a direct measure of how wide the envelope is. A squeeze is simply bandwidth printing at a low relative to its own history.

Mean reversion vs breakout: pick the regime first

Bollinger Bands support two opposite playbooks, and using the wrong one for the market you are actually in is the most expensive error in bollinger bands trading.

Regime	How the bands behave	The playbook	The failure mode
Range / chop	Bands roughly horizontal; price ping-pongs between them	Mean reversion — a touch of a band tends to reject back toward the middle	The range ends and a trend begins
Trend	Bands slope; price "walks the band," closing on the upper (or lower) band repeatedly	Trend continuation — band rides are strength, not exhaustion	Fading the band means shorting the strongest tape

Notice the contradiction: in a range a band touch is a fade, and in a trend the same touch is confirmation to hold. So the first question is never "did price touch a band?" It is "what regime am I in?" — and the bands themselves don't answer that. Trend structure does. Tools like ADX exist precisely to grade trend strength before you decide whether to fade a band or ride it. "Walking the band" in a strong trend is not overbought; it is the market telling you the move has legs.

A worked squeeze-to-expansion sequence

Spot the squeeze. Bandwidth is at a multi-month low; the bands are visibly the tightest they've been in weeks. This is a radar ping, not a trade.
Wait. No direction is implied yet. Do not pre-position.
Require a reason. The bands don't supply a catalyst. A news item, an earnings reaction, or a break of a level you already marked is what gives the coming move a direction to expect.
Demand confirmation. A decisive close outside the band on above-average volume, with the middle band beginning to turn that way, is the expansion — not the first tick past the band.
Respect the head-fake. An initial poke that closes back inside is a caution flag. Some traders specifically wait for that failed poke to reverse before committing.
Define the trade before entry. A written trigger, a first and second target, a stop (often beyond the opposite band or the middle band), and a size. A risk-reward calculator tells you whether the target justifies the stop, and a position-size calculator keeps the loss bounded if the break is a fake.

False signals: why "touch equals trade" fails

Three false signals recur. First, the band touch in a trend — fading a stock that is walking the upper band, which is fighting momentum. Second, the squeeze head-fake — the false break out of compression. Third, treating an adaptive band as a fixed threshold — because the envelope re-measures volatility every bar, a "touch" is not the same fixed event twice.

The deeper reason sits underneath all three: Bollinger Bands are a transform of past price — a moving average plus the standard deviation of closes that already printed. Like every price-derived indicator, they carry no information price didn't already contain. In our published hypothetical backtest, a raw scanner traded blind — mechanical entries with no catalyst check — produced 161 simulated trades, a 46.6% simulated win rate, and a 0.82 simulated profit factor, i.e. negative expectancy per trade. Indicator thresholds crossing lines are exactly the kind of blind logic that test was running. The full breakdown is on the public record.

Where the bands fit on our desk

Used well, Bollinger Bands are a volatility context gauge and a squeeze radar — not a trigger. They help characterize a name: is it compressed and coiling, or extended and riding a band? That read informs a setup; it does not place the trade. A complete idea specifies a trigger, TP1/TP2, a stop, and a time-stop, posted before the move to a public, timestamped paper record where the losers stay visible — you can see how that discipline reads on live setups in the signals feed. The envelope tells you how stretched or calm price is. What you do about it still has to come from a catalyst, a level, and a written plan — never from the band alone.

Common questions

Is a touch of the upper Bollinger Band a sell signal?

No. A touch of the upper band means price is extended and volatility is high relative to the last 20 bars — it is a description, not an instruction. In a range, band touches often reject back toward the middle, but in a strong uptrend price "walks the band," closing on the upper band repeatedly while it keeps rising. Fading every upper-band touch in a trend means shorting the strongest tape, which is why regime matters more than the touch itself.

What is a Bollinger Band squeeze?

A squeeze is when the bands pinch together because volatility has compressed — bandwidth drops to a multi-week or multi-month low. Since volatility tends to mean-revert, a squeeze flags that a larger directional move may be building. The critical limit: a squeeze tells you a move may be coming but says nothing about direction, so trading a squeeze usually means waiting for a confirmed expansion rather than guessing which way it breaks.

What are the standard Bollinger Band settings?

The default is a 20-period simple moving average for the middle band, with the upper and lower bands set two standard deviations above and below it. These are conventions, not magic numbers — shorter periods react faster and give more false touches, wider standard-deviation multiples contain more price but signal less often. The settings should match your timeframe and purpose, not be copied blindly.

Should I use Bollinger Bands for mean reversion or breakouts?

Both are valid, but they are opposite strategies for opposite regimes, so you pick the regime first. In a range-bound market, band touches tend to fade back to the middle (mean reversion). In a trend, price rides the band and touches are continuation, not reversal (breakout/trend-following). Deciding which one applies depends on trend structure, which the bands themselves don't measure — this page treats them as a volatility context gauge, not a standalone signal, and none of it is advice to buy or sell anything.

See the process with your own eyes. The desk posts trigger-based cards to a public, timestamped record — losses included — and published the backtest where its own raw scanner loses. The scoreboard is free to watch. Join the floor →

Free to join · paid floors optional · research and education only

Last updated 2026-07-11 · ClaudeQuantAlgo Research Desk · research and education only.

Disclosures. ClaudeQuantAlgo is a research and education community. Nothing on this page constitutes financial, investment, legal, or tax advice, or a recommendation to buy or sell any security or derivative. No profit promises are made, ever — trading stocks, options, and forex involves substantial risk of loss; options positions can lose 100% of their value. The public scoreboard reflects a model ("paper") desk — no real money. Past performance — real, paper, or simulated — never guarantees future results. Hypothetical and simulated results have inherent limitations and no representation is made that any account will or is likely to achieve similar profits or losses. ClaudeQuantAlgo is not a registered investment adviser or broker-dealer. You are solely responsible for your own trading decisions. Never risk money you cannot afford to lose.