What Bollinger Bands measure

Bollinger Bands are a volatility-based indicator built from a moving average and two standard-deviation bands. The classic setup uses a 20-period simple moving average (SMA) as the middle band, with the upper and lower bands placed at:

  • Upper band = SMA(20) + 2 × standard deviation(20)
  • Lower band = SMA(20) - 2 × standard deviation(20)

Because the bands expand when volatility rises and contract when volatility falls, they help traders see whether price is stretched relative to its recent range. That makes them useful for two very different strategy families: mean reversion and breakout trading.

Mean-reversion: betting on a snap back

A mean-reversion approach assumes that when price moves far from its recent average, it is more likely to drift back toward the middle band. A simple version is:

  • Buy when price closes below the lower band
  • Exit when price returns to the middle band or upper band
  • Optionally short when price closes above the upper band

This works best in range-bound markets, where price oscillates around a stable mean. Traders often add filters to avoid fading strong trends. Common examples include:

  • Only trade when the slope of the 20-SMA is flat or mild
  • Require a low ADX reading, such as below 20 to 25
  • Use an RSI filter, for example buying only when RSI(14) is below 30

A key pitfall is assuming every band touch is an opportunity. In a strong trend, price can “walk the band” for a long time, meaning repeated touches of the upper band during an uptrend or the lower band during a downtrend. Fading those moves too early can lead to large losses.

Breakout: using volatility expansion

A breakout strategy treats the bands differently. Instead of fading extremes, it looks for expansion after compression. The logic is that narrow bands often precede a large move. One common setup is:

  • Identify a squeeze when band width is unusually low
  • Enter long on a close above the upper band
  • Enter short on a close below the lower band
  • Use a stop below the breakout level or the opposite band

Band width can be measured as:

  • Band width = (Upper band - Lower band) / Middle band

When band width is low relative to its own history, volatility is compressed. Traders may compare current width to a rolling percentile, such as the lowest 20% of the last 100 bars. Breakout systems often work better when paired with volume expansion, trend filters, or a higher-timeframe bias.

The main risk is false breakouts. Price may pierce a band briefly and then reverse. To reduce whipsaws, many traders wait for a full candle close beyond the band, or require confirmation such as a higher close on the next bar.

How the same indicator supports opposite ideas

Bollinger Bands do not predict direction by themselves. They only show how far price is from a recent average, adjusted for volatility. That is why the same indicator can support both mean-reversion and breakout logic:

  • Mean reversion says: “Price is stretched, so it may return to the mean.”
  • Breakout says: “Volatility is expanding, so a new trend may be starting.”

The market regime matters more than the indicator. Range-bound conditions tend to favor mean reversion; trending or news-driven conditions tend to favor breakouts.

Typical parameters and what they change

The default 20-period, 2-standard-deviation setting is popular because it balances responsiveness and noise. But it is not sacred.

  • Shorter lookback, such as 10 or 14 periods: more sensitive, more signals, more noise
  • Longer lookback, such as 50 periods: smoother, fewer signals, slower response
  • Smaller deviation multiplier, such as 1.5: tighter bands, more touches
  • Larger multiplier, such as 2.5: wider bands, fewer touches

Changing parameters changes the trade distribution. For example, a 10-period, 1.5σ setup may generate many more mean-reversion entries, but also more false signals. Always test whether the edge survives after commissions and slippage.

Backtesting both approaches correctly

A good backtest should separate signal logic from risk management. For example, a mean-reversion strategy might enter at the lower band and exit at the middle band, but the position should still have a stop-loss, such as 1.5 to 2.5 times ATR(14) below entry. Breakout systems often need trailing stops or volatility-based exits to capture trends.

Useful metrics include:

  • Win rate
  • Average win / average loss
  • Profit factor
  • Maximum drawdown
  • Expectancy per trade

Expectancy can be approximated as:

  • Expectancy = (Win rate × Average win) - (Loss rate × Average loss)

A strategy with a modest win rate can still be profitable if winners are much larger than losers, which is often the case for breakout systems. Mean-reversion systems often have higher win rates but smaller average wins, so losses must be tightly controlled.

Common mistakes to avoid

  • Using the same rules in every market regime
  • Optimizing parameters until the backtest looks perfect
  • Ignoring transaction costs on lower timeframes
  • Entering on intrabar band touches without confirming candle closes
  • Failing to test across stocks, forex, and crypto separately
  • Assuming a single asset’s historical behavior will generalize

A robust process usually includes out-of-sample testing, walk-forward analysis, and paper trading before risking capital. That helps reveal whether the edge is real or just curve fit.

Practical takeaway

Bollinger Bands are best thought of as a volatility framework, not a complete strategy. If you believe markets revert to a mean, use the bands to identify stretched conditions and add regime filters. If you believe volatility compression leads to expansion, use them to spot breakouts and confirm momentum.

The real question is not “Do Bollinger Bands work?” but “Under what market conditions, with what rules, and after costs?” Backtesting can answer that far better than intuition alone.

Educational content only, not financial advice.