Bollinger Bands: volatility and reversion.

Bollinger Bands, created by John Bollinger in the 1980s, solve a problem that fixed indicators cannot: they adapt to volatility. Instead of static levels, the bands breathe with the market — they widen when it gets agitated and tighten when it calms down. That makes them excellent for measuring relative extremes and anticipating explosive moves.

01What Bollinger Bands are

They are three lines:

• Middle band: a simple moving average (usually 20 periods). It is the reference "fair value".
• Upper band: the average + 2 standard deviations.
• Lower band: the average − 2 standard deviations.

The key concept is standard deviation — a statistical measure of volatility. When the market is volatile, the deviation rises and the bands move apart; when it is calm, they move closer together. Statistically, about 95% of prices stay within 2 standard deviations, so touching a band means price is at a relative extreme.

02The main uses

1. Mean reversion (ranging market)

In a market with no trend, price tends to oscillate between the bands. Touching the lower band can signal a buy ("cheap" price); the upper band, a sell ("expensive" price). That is the basis of mean reversion strategies.

2. The squeeze (the most powerful use)

When the bands tighten sharply, it means low volatility — the market is "compressed". And here is the insight: periods of low volatility tend to be followed by periods of high volatility. The squeeze does not tell you the direction, but it warns that a strong move is building. Traders use it to get ready for a breakout.

3. "Walking the band" (strong trend)

Here is what wrecks beginners: in a strong trend, price sticks to the upper band (in an uptrend) and keeps climbing. Anyone who reads that as "overbought, I'll sell" is trading against the trend and loses. Walking the band is a sign of strength, not of reversal.

03Coding Bollinger Bands

//@version=5
indicator("Bandas de Bollinger", overlay=true)

periodo = input.int(20, "Período")
desvios = input.float(2.0, "Desvios padrão")

media = ta.sma(close, periodo)
desvio = ta.stdev(close, periodo)
superior = media + desvios * desvio
inferior = media - desvios * desvio

plot(media, "Média", color=color.blue)
p1 = plot(superior, "Superior", color=color.red)
p2 = plot(inferior, "Inferior", color=color.green)
fill(p1, p2, color=color.new(color.blue, 90))

// detects a squeeze: band width at its lowest recent level
largura = (superior - inferior) / media
squeeze = largura < ta.lowest(largura, 50) * 1.1
if squeeze
    alert("Squeeze — possível movimento forte chegando")

import pandas as pd

def bollinger(precos, periodo=20, desvios=2.0):
    media  = precos.rolling(periodo).mean()
    desvio = precos.rolling(periodo).std()
    superior = media + desvios * desvio
    inferior = media - desvios * desvio
    return media, superior, inferior

media, sup, inf = bollinger(df["close"])

# relative band width (to detect squeezes)
largura = (sup - inf) / media
df["squeeze"] = largura < largura.rolling(50).min() * 1.1

ultimo = df["close"].iloc[-1]
if ultimo <= inf.iloc[-1]:
    print("Preço na banda inferior — avaliar (só em lateral!)")

04Building strategies

• Reversion (ranging): buy at the lower band + RSI < 30, sell at the upper band + RSI > 70. Only in a market with no trend.
• Squeeze breakout: detect the squeeze, wait for a break of one of the bands, enter in the direction of the break. Captures the start of strong moves.
• Bollinger + trend: in an uptrend, use the lower band (or the middle average) as a re-entry zone on pullbacks.

05Frequently asked questions