Bollinger Bands, developed by John Bollinger in the early 1980s, are a volatility envelope plotted at a fixed number of standard deviations above and below a moving average. Their core idea is that volatility is relative and mean-reverting: by making the band width adapt to recent price dispersion, they create a dynamic definition of "high" and "low" that tightens in quiet markets and expands in turbulent ones. The central tension a swing trader must hold is that the bands describe volatility, not direction — they tell you how stretched price is, never which way it will resolve, and treating a band touch as a trade signal is the single most common way they are misused.

How it's calculated / formed

Three lines from the same inputs (StockCharts ChartSchool, Fidelity):

• Middle band = 20-period simple moving average (SMA)
• Upper band = SMA + (2 × 20-period standard deviation of price)
• Lower band = SMA − (2 × 20-period standard deviation of price)

Bollinger's defaults are 20 periods, 2 standard deviations (his Rule 9). He notes Fidelity-cited variants: 10-period / 1.5σ for shorter horizons, 50-period / 2.5σ for longer. Because σ is recalculated every bar, the bands breathe with volatility automatically.

Two derived indicators do most of the analytical work:

• %B = (Price − Lower Band) / (Upper Band − Lower Band). %B = 1 at the upper band, 0 at the lower, 0.5 at the SMA, and can exceed 1 or go negative outside the bands. It normalizes "where is price within the channel."
• BandWidth = (Upper − Lower) / Middle. It normalizes the width of the channel and is the engine of the Squeeze.

Caution Bollinger himself stresses (Rule 14): make no statistical assumptions from the σ calculation. The "~95% of price falls within 2σ" rule of thumb assumes a normal, stationary distribution; real returns are fat-tailed and serially correlated, so the bands routinely contain far less than 95% of bars in trends.

How it's used in practice (the swing setups)

A swing trader keys on four concrete configurations:

1. The Squeeze (volatility contraction → expansion). When BandWidth contracts to a multi-month low, energy is coiling for a directional move. The standard trigger: wait for the first close clearly outside the contracted band, ideally with an expansion in volume, and trade in that direction. Failure mode: the head-fake — Bollinger explicitly warns price often pokes one way first, then reverses and runs the other way. Confirmation rule of thumb: a follow-through close, not just an intraday tag.

2. W-bottoms (reversal long). Bollinger's adaptation of the Arthur Merrill "W" pattern. Price makes a low that pierces the lower band, rallies, then forms a second low — in Bollinger's canonical form the second low can be lower in price than the first yet still holds inside/above the lower band (a higher %B on the second dip = positive divergence). Trigger: break above the reaction high between the two lows. M-tops are the mirror image for shorts.

3. Walking the band (trend continuation). Repeated closes hugging the upper band are strength, not exhaustion (Bollinger's Rule 7: price walks the band). The swing edge here is to stop fading band tags in a confirmed trend and instead buy pullbacks toward the middle band, using the SMA as dynamic support. A trend-up "walk" is typically confirmed by a separate momentum/volume tool, not by the bands alone.

4. Band tag as a target, not a trigger. After an entry off another signal, the opposite band is a logical profit zone in a ranging instrument — but only when no trend "walk" is underway.

Across all four, Bollinger's Rules 6 and (per Fidelity) the bands' explicit design intent apply: a band tag is not by itself a buy or sell signal, and the bands are meant to confirm signals from a non-correlated indicator (e.g. volume, RSI, MACD), never to stand alone.

Adoption, debate & evidence

Bollinger Bands are among the most widely deployed indicators in retail and many institutional charting stacks, and BandWidth/Squeeze concepts are baked into mainstream platforms. The mechanics are uncontested; the edge is where honesty is required.

The measured record is mixed and strategy-dependent. Backtest aggregators (e.g. QuantifiedStrategies) report that mean-reversion entries (buy a close below the lower band) show some profitability in mean-reverting single stocks but generally fail on stock indices. Day-trading-style band-touch systems on intraday charts have been reported with low win rates (cited backtests around 23–30% on 1- and 5-minute charts — LiberatedStockTrader); treat such precise figures as illustrative of one dataset, not a universal base rate. There is no broad peer-reviewed consensus establishing a standalone Bollinger edge; a single-strategy academic study (Su, Atlantis Press) examines mean-reversion variants but is not generalizable. Folklore-vs-measured: the popular "price reverses at the bands" belief is contradicted by Bollinger's own walking-the-band rule and by trend data — fading every tag is a known losing pattern. What survives scrutiny best is the Squeeze as a volatility-regime signal (low BandWidth reliably precedes higher realized volatility) — though it predicts magnitude, not direction.

Strengths & limitations

Strengths: self-adapting to volatility regime; excellent for framing (defining relative high/low, clarifying W/M patterns, spotting %B divergences); the Squeeze is a genuinely useful timing filter for "a move is coming."

Limitations: lagging (SMA-based); directionally agnostic; the σ math invites false statistical confidence; and in strong trends naïve fading is actively harmful. The #1 misuse: treating a band tag as an overbought/oversold signal — the bands measure stretch relative to recent volatility, not absolute over-extension, and trends will tag and walk the band for weeks.

Sources

• StockCharts ChartSchool — Bollinger Bands (formula, %B, BandWidth, Squeeze, W-bottoms)
• John Bollinger's official 22 rules — bollingerbands.com (Rule 6 tags-are-not-signals, Rule 7 walking the band, Rule 9 defaults, Rule 14 no statistical assumptions, Rule 15 %B)
• Fidelity — Bollinger Bands indicator guide (calculation, settings, "confirm with other indicators")
• QuantifiedStrategies — Bollinger Bands backtests (mean-reversion works on stocks, fails on indices)
• LiberatedStockTrader — Bollinger Bands real test results (intraday low win-rate figures — single-dataset, treat as illustrative)
• Su — Bollinger Band mean regression study (Atlantis Press) (single-strategy academic test, not generalizable)

Dispute flags: precise win-rate percentages come from individual retail backtests, not peer-reviewed work — qualified as illustrative. No robust academic consensus establishes a standalone Bollinger edge; the strongest evidence supports the Squeeze as a volatility (not directional) signal.