Skip to main content

Overview

Bollinger Bands can be understood as: a “stretchable price channel” drawn above and below a moving average using volatility (standard deviation). It aims to solve two problems:
  1. Not only to see whether the price is “high or low” relative to the moving average, but to see—under the current volatility regime—whether the deviation is “outlandish”;
  2. To make “support/resistance” not a rigid line, but a channel that automatically widens and narrows with volatility.
Typical uses of Bollinger Bands:
  • Observe volatility contraction/expansion (before “a long sideways move leads to a breakout” or “a long move leads to consolidation,” changes in band width often appear first)
  • Judge whether price is repeatedly moving up and down within a range, or trending along the rails
  • Assist with:
    • Buy low, sell high in range markets
    • Breakout trend-following after tight consolidation
    • In trend markets, walking the band + exit timing
But also remember:
Bollinger Bands are not a “price cage.” Price often pierces the upper/lower bands—what matters is the environment and what happens next.

Principles of Bollinger Bands

Components

A standard Bollinger Band consists of three lines:
  1. Middle Band Typically an N-period Simple Moving Average (SMA), with a common default of the 20-day SMA.
  2. Upper Band Upper Band = Middle Band + K × Standard Deviation (Std)
  3. Lower Band Lower Band = Middle Band − K × Standard Deviation
Where:
  • N: the lookback period used for the moving average and standard deviation; commonly 20 (adjustable by timeframe)
  • K: the standard deviation multiplier; commonly 2 (theoretically covering about 95% of data under an approximate normal distribution)
What standard deviation (Std) means:
  • Larger Std → recent price swings around the mean are more violent
  • Smaller Std → price is more “well-behaved,” staying close to the mean
So:
  • High volatility → bands automatically expand (wider channel)
  • Low volatility → bands automatically contract (narrower channel)
You can think of Bollinger Bands as:
A channel with “adaptive width”: it narrows when markets are quiet and widens when markets are wild.

Trading Signals

Several common key signals of Bollinger Bands revolve around three words: “squeeze, expansion, crossing.”

1. Squeeze: Volatility Contraction

Characteristics:
  • Upper and lower bands keep moving closer to the middle band
  • Bollinger Bands become noticeably narrower, sometimes almost like a single “strip”
  • Candles often show small bodies and small ranges, with shrinking volume
Meaning:
  • The market enters a low-volatility consolidation
  • Bulls and bears are both hesitant to act aggressively; supply changes hands in a tight range
Practical significance:
  • A squeeze is often the “prelude to a big move,” but it does not tell you the direction:
    • A high-volume breakout above the upper band → may start an uptrend
    • A high-volume breakdown below the lower band → may start a downtrend
  • Many Bollinger strategies focus specifically on “breakouts after a narrow band” for trend-following trades

2. Expansion: Volatility Expansion

Characteristics:
  • Previously narrow bands suddenly open up
  • Upper band curls upward, lower band presses downward, and the channel quickly widens
  • Often accompanied by large candles (big bullish/bearish bodies) + rising volume
Meaning:
  • The market goes from “asleep” to “awake”:
    • One side clearly gains the upper hand; price rapidly deviates from the mean in one direction
  • During expansion, it is often the trend start or acceleration phase
Practical application:
  • If it is a breakout upward + expansion:
    • It can be treated as a signal of bullish trend initiation/acceleration
  • If it is a breakdown downward + expansion:
    • It can be treated as bearish trend initiation/acceleration
Note:
Don’t interpret “wider bands” as “an immediate reversal.” More often it tells you: a trend is running now—this is no longer small noise.

3. Price Crossing: Touching/Breaking the Upper or Lower Band

The relationship between price and the bands is often used to judge “extreme positions” and “trend strength.” Two common approaches (many people mix them—this is the risky part): (1) Range mindset: treat the bands as “overbought/oversold”
  • In a clearly sideways range:
    • Price touches the upper band → can be seen as “short-term high,” some traders short-term sell or reduce exposure
    • Price touches the lower band → can be seen as “short-term low,” some traders short-term buy or add exposure
  • Core logic:
    • In range markets, price repeatedly “springs” up and down within the channel
    • “Buy the lower band, sell the upper band” may work for a while
(2) Trend mindset: “walking the band”
  • In a one-sided trend:
    • Uptrend: price frequently hugs the upper band or even runs outside it, while pullbacks usually only reach around the middle band
    • Downtrend: price frequently hugs the lower band or even runs outside it, while rebounds usually only reach around the middle band
  • In this case:
    • Touching or even breaking the upper band is no longer “overbought, must drop,” but rather “bulls are strong”
    • Likewise, touching the lower band does not mean “an immediate rebound”
Therefore:
“Short when price hits the upper band, long when it hits the lower band” only fits range conditions and must be combined with other signals; once a trend begins, this usage can easily lose money by trading against the move.

Core Concepts

1. “Most of the time inside the bands, a small part of the time outside”

With the common 20 + 2 standard deviations parameters, if price distribution is approximately normal:
  • Theoretically about 95% of prices fall between the upper and lower bands
  • The remaining ~5% of the time, price runs outside the bands
So:
  • Outside the bands ≠ immediate reversal —it only indicates “price is relatively extreme versus recent volatility,” and it may continue outside for a while or quickly revert inside.
  • What matters is:
    • The band structure (expansion/squeeze)
    • Whether the market is in a range regime or a trend regime
    • Whether there is confluence from volume, patterns, support/resistance, etc.

2. BandWidth = Volatility

You can use BandWidth as a simple volatility indicator:
BandWidth = (Upper Band − Lower Band) ÷ Middle Band  or simply (Upper Band − Lower Band)
  • BandWidth keeps shrinking → volatility keeps dropping → a potential “setup/accumulation phase”
  • BandWidth expands quickly → volatility spikes → likely a trend leg or major swing in progress
Many trading systems:
  • Watch for breakouts when BandWidth is extremely small
  • When BandWidth is extremely large, and the trend appears late-stage with other risk signals, consider reducing exposure and guarding against drawdowns

3. Middle Band = Dynamic Mean-Price Reference

The middle band is essentially a moving average, commonly the 20-day SMA. In practice:
  • Upper/Lower bands: observe extremes & momentum
  • Middle band: observe trend direction & dynamic mean price
Simple examples:
  • In an uptrend:
    • Price pulls back to around the middle band and then resumes rising → a normal pullback
  • Once price fails to regain the middle band for several consecutive days and the upper band flattens or bends down:
    • It suggests the uptrend may be entering consolidation or reversal

4. Do not treat Bollinger Bands as the only indicator

Advantages of Bollinger Bands:
  • Adaptive width
  • Combines “moving average + volatility” information
But they cannot see:
  • Volume structure
  • Fundamental changes
  • Positioning/cost distribution, capital flow
So a more practical approach is:
Use Bollinger Bands as a visual tool that compresses price + volatility information, then layer in trendlines, support/resistance, volume, patterns, etc., to form your own multi-dimensional trading conditions.

Practical Applications

Case 1: Breakout Trading After a Tight Squeeze

Background:
  • An index has been ranging between 3000–3050 for a long time
  • Under Bollinger Bands (20, 2):
    • Upper/lower bands gradually tighten and BandWidth shrinks noticeably
    • Candles become a cluster of small-bodied dojis; volume dries up
Observation:
  • BandWidth is at an obvious multi-month low → volatility compression
  • Sentiment is calm; inventory changes hands repeatedly in a tight area
Trading idea (example):
  1. Don’t guess direction; wait for the breakout and follow:
    • If one day a high-volume long bullish candle breaks above the upper band and the bands open:
      • Treat it as a bullish breakout signal and try a trend-following long
    • If instead a high-volume long bearish candle breaks below the lower band:
      • Consider the bearish direction
  2. Stops and risk control:
    • Stops are often placed below half or one-third of the breakout candle
    • Or near the middle band / a key support level inside the band
  3. After the breakout, if over the next few days:
    • Bands keep expanding and price “walks” along the upper or lower band
    • Bulls/bears can keep holding until there are clear signs of a drop back to the middle band or an opposite-band break

Case 2: “Channel Bounce” and “Channel Pullback” in a Range Market

Background:
  • A stock has long traded sideways in the 10–11 range
  • BandWidth is relatively stable; upper and lower bands are roughly parallel
  • No clear trend; average volume
Observation:
  • Repeatedly:
    • Near the upper band → next day / within a few candles, price falls back
    • Near the lower band → next day / within a few candles, price rebounds
Trading idea (short-term bias):
On the premise that it is a confirmed range environment, you can attempt a range strategy of “buy near the lower band, sell near the upper band.”
Illustrative rules:
  • Price approaches or slightly breaks the lower band + a reversal/hold pattern appears (hammer, engulfing, etc.)
    • Probe long with a small position; first target the middle band or near the upper band
  • Price approaches or slightly breaks the upper band + clear rejection appears
    • Reduce short-term exposure / flip to short (if shorting is allowed for the instrument)
Emphasize again:
  • It must be a clear box range. Once you see one-sided expansion + price walking the band, exit the “buy low, sell high” mindset and shift to trend thinking.

Case 3: “Walking the Band + Exit Signals” in a Trend

Background:
  • A commodity rallies from 100 to 140 with a clearly rising channel
  • Price repeatedly “grinds” upward along the upper band; pullbacks reach at most around the middle band
  • Bands slope upward; BandWidth is medium-to-large
Trend-following plan (example):
  1. Entry:
    • Probe long on a squeeze breakout + first close outside the upper band
    • Or add on the first strong pullback to the middle band that holds
  2. Hold:
    • As long as price spends most of the time above the middle band and occasionally walks the upper band, keep holding
  3. Exit signals:
    • Several consecutive days failing to regain the middle band, with the middle band flattening or bending down
    • Or price breaks below the lower band and the bands start opening downward
    • With high-volume long bearish candles and other signals, consider:
      • Taking profit and closing
      • Or at least cutting exposure significantly and waiting for a new structural confirmation

FAQs

Q1: If price touches the upper band, should I always short; if it touches the lower band, should I always go long?

Absolutely not.
  • In range markets:
    • The bands can be viewed as “range edges,” and “upper band = relatively high, lower band = relatively low” can be somewhat informative.
  • In trend markets:
    • Price can walk the upper/lower band for a long time, and touching the upper band is often not a “sell signal,” but a sign of strength.
Put simply:
Only after you first confirm it is a range market can you consider “sell high, buy low”; once the bands open clearly and the trend is obvious, you cannot force this approach.

Q2: Do Bollinger parameters have to be 20 + 2? Can they be adjusted?

20 + 2 is the most commonly used “default combination,” and it’s also recommended by the inventor:
  • 20: roughly one month of trading days (on daily charts)
  • 2: two standard deviations, theoretically covering most fluctuations
But parameters are not a holy grail. You can adjust based on:
  • Timeframe (intraday/daily/weekly)
  • Instrument volatility characteristics (high/low volatility)
  • Your strategy (more trend-focused vs. more range-focused)
For example:
  • If you want more sensitivity:
    • Shorten the period (e.g., 10, 12)
    • Or slightly reduce the deviation multiplier (e.g., 1.5)
  • If you want more smoothing:
    • Lengthen the period (e.g., 30, 50)
    • Or increase the deviation multiplier (e.g., 2.5)
Suggestion:
Before changing parameters, do review/quick backtesting to compare performance under different settings on the instruments you trade, then decide whether to adopt—rather than switching arbitrarily.

Q3: Can Bollinger Bands tell me the breakout direction in advance? After a squeeze, which way will it go?

Bollinger Bands can only tell you:
  • Volatility is contracting (the market is storing energy)
  • There is a high probability of a “release” (a large move) ahead
But you cannot infer direction from band shape alone. Direction is usually driven by:
  • Fundamentals/news (bullish/bearish catalysts)
  • The broader trend (after extended rises, downside is more likely; after extended drops, upside is more likely—only probabilistic)
  • Positioning/flow (long/short exposure, sentiment)
So a more robust approach is:
  • Use Bollinger Bands to find candidates/zones of volatility compression
  • Then wait for the actual breakout direction to appear (price, volume, pattern confirmation)
  • Trade in the breakout direction, rather than “betting” ahead of time

Summary

  • Bollinger Bands = an adaptive price channel built from a moving average + standard deviation:
    • Middle band: the moving average (trend & dynamic mean price)
    • Upper/Lower bands: MA ± K × Std (volatility range)
  • Core uses:
    • Use BandWidth to read volatility: squeeze → big move may be brewing; expansion → big move in progress
    • Use walking the band to judge trend strength: hugging the upper/lower band = strong momentum
    • In range markets, use bands to assist buy-low/sell-high; in trend markets, use “walking the band + middle band” to follow and time exits
  • Key points in use:
    • The bands are not a “price cage”; being outside does not mean immediate reversal
    • Range logic (sell high/buy low) and trend logic (walk the band) must never be mixed
    • 20 + 2 is the classic default, adjustable moderately by instrument and strategy
    • Combine with trendlines, support/resistance, volume, patterns, etc., rather than using it alone as a trading basis
In one sentence:
Bollinger Bands are not for “guessing tomorrow’s direction,” but for seeing volatility structure + positional context, helping you rein in greed in ranges and avoid fighting the trend when it’s time to follow.

Further Reading

  • Related resources
    • In investor-education sections of major brokers/trading platforms, you can find articles and videos on “Bollinger Bands,” “Bollinger Channel,” and “volatility channel trading,” which pair well with practicing on real charts to draw bands and observe squeezes/expansions.
    • Search Bollinger Bands on technical analysis learning sites or communities; you’ll often find many diagrams and live-market cases, making it very intuitive to compare how the bands behave under different regimes.
  • Recommended books or articles
    • John Bollinger: Bollinger on Bollinger Bands (Chinese editions may use different translated titles) A systematic explanation by the creator covering principles, parameter choices, and how to combine with other indicators—often the first go-to reference for deeper understanding.
    • Technical Analysis of the Financial Markets — John J. Murphy Provides a brief introduction to Bollinger Bands and similar volatility channels in sections on channels, volatility, and indicators, helping you place Bollinger Bands within a broader technical analysis framework.