Bollinger Bands Strategy Guide for Trading Bots

Bollinger Bands wrap a moving average in two bands set a number of standard deviations away. They visualize volatility: the bands widen when the market is volatile and contract when it is calm.

The Three Lines

• Middle band: a moving average (default 20).
• Upper band: middle + (2 × standard deviation).
• Lower band: middle − (2 × standard deviation).

Roughly 95% of price action stays inside 2-standard-deviation bands, which is why touches of the bands draw attention.

Two Opposite Uses

1. Mean reversion. In a range, price tends to revert from the bands to the middle. Buy near the lower band, sell/exit near the middle or upper band. Works in sideways markets; dangerous in trends.

2. Breakout (squeeze). When the bands contract sharply (a "squeeze"), volatility is low and a big move often follows. Trade the breakout when price closes outside the bands after a squeeze.

Knowing which regime you are in is everything — the same band touch means "buy the dip" in a range and "trend is accelerating" in a breakout.

Building It in Setup.Cash

Add the Bollinger Bands indicator in the builder, with inputs for length and standard deviation. Then:

• Mean reversion: long when price touches the lower band and turns up.
• Breakout: long when price closes above the upper band after a low-volatility period.

Or describe it to text-to-strategy: "Bollinger breakout on BTCUSD 1h after a squeeze, ATR stop, trail the winner."

Combine With a Trend Filter

Bollinger mean reversion is much safer with a higher-timeframe trend filter: only fade the lower band when the bigger trend is neutral or up. In a strong downtrend, "oversold" band touches keep getting more oversold.

Tuning

• Length 20, 2 SD: the classic default.
• Tighter SD (1.5): more signals, more noise.
• Wider SD (2.5–3): only extreme moves trigger.

Build a Custom Variant

In the Indicators Lab you can build a custom band system — for example, bands plus a squeeze-detection event you can wire straight into entries. Generate it with AI or write it in the formula language.

As always, backtest the strategy across multiple periods and confirm with paper trading before going live. Bollinger Bands describe volatility — they do not predict direction on their own.