“Bollinger Bands are a technical analysis tool used to measure market volatility and identify potential overbought or oversold conditions by plotting three lines on a price chart: a 20-day simple moving average (middle band) and two standard deviation lines (upper and lower bands).” – Bollinger Bands – Technical analysis
Trading decisions often fail when a relative price signal is mistaken for a forecast. A close near an outer boundary tells an analyst that price is unusually far from its recent average after adjusting for recent dispersion; it does not, by itself, establish that a reversal is due. The same observation can describe a powerful trend, a short-lived shock, or a market whose statistical behaviour has changed. This distinction gives the indicator practical value while placing firm limits on what it can say.
How the calculation works begins with a moving average over a selected look-back period. For observations indexed by t, a simple moving average with period n is M_t=\frac{1}{n}\sum_{i=0}^{n-1}P_{t-i}, where P is the chosen price, usually the closing price. The corresponding standard deviation measures how widely those prices have dispersed around the average. If the multiplier is k and the deviation is \sigma_t, the upper and lower boundaries are U_t=M_t+k\sigma_t and L_t=M_t-k\sigma_t. The familiar default uses 20 periods and a multiplier of 2, although these are conventions rather than universal rules.1,2,3
The calculation combines direction and dispersion without confusing them. The central average slopes upwards when the recent price path has a positive bias and downwards when it has a negative bias. The distance between the boundaries expands when recent prices become more variable and contracts when they become quieter. A useful normalised measure is band width, expressed as W_t=\frac{U_t-L_t}{M_t}. This makes the indicator more comparable across instruments with different price levels, though it remains sensitive to the look-back period, the price field, and the chosen volatility estimator.
Relative high and relative low are the proper starting points for interpretation. A price above the upper boundary is high relative to its recent distribution, while a price below the lower boundary is low relative to that distribution. It is not automatically overbought or oversold in the stronger sense of being destined to reverse. In a persistent advance, prices can remain close to, or repeatedly exceed, the upper boundary as volatility and trend reinforce one another. In a persistent decline, repeated contact with the lower boundary can reflect continued selling pressure rather than a bargain.
That is why experienced users distinguish between mean-reversion and trend-following readings. A mean-reversion approach looks for evidence that an excursion is losing force, such as a failed continuation, a return inside the boundary, or confirmation from another measure of momentum. A trend-oriented approach treats sustained movement along an outer boundary as evidence of strength or weakness and uses the central average, recent swing structure, or a separate risk rule to define exposure. The same band contact can therefore produce opposite decisions under different market regimes. The indicator supplies context; the trading rule supplies the interpretation.
Volatility regimes and the squeeze
A pronounced narrowing of the boundaries is commonly called a squeeze. It signals that recent dispersion has fallen, not that a particular direction is guaranteed. Quiet conditions can precede a large move because compressed prices leave room for volatility to expand, but the eventual break may be upward, downward, or false. A disciplined process therefore waits for directional evidence, evaluates the size and persistence of the move, and considers whether volume, market breadth, macroeconomic news, or correlated assets support the signal. Narrow bands can also persist during uneventful markets, so timing a breakout solely from width is hazardous.
The opposite condition, an expansion, indicates that recent movement has become more variable. It can occur during a genuine trend, after an information shock, or during disorderly price discovery. Expansion is not synonymous with opportunity: wider bands imply larger potential movements and therefore greater position risk. Traders who use a fixed number of shares or contracts may unintentionally increase portfolio risk when volatility rises. Position sizing, stop placement, and exposure limits should respond to the changing range rather than treating the chart distance as a guaranteed forecast.
Parameter choice creates a trade-off. A shorter period reacts quickly but produces more frequent changes in the average and more unstable boundaries. A longer period smooths noise but may respond slowly when the market regime changes. A larger multiplier creates wider boundaries and fewer crossings; a smaller multiplier creates narrower boundaries and more signals. The default settings are useful for communication and comparison, but they should not be presented as optimal for every asset, timeframe, or objective.1,2 Parameters selected after inspecting the same data used to evaluate performance can create overfitting, especially when many combinations are tested.
Statistical intuition also needs care. If observations were independent and normally distributed around a stable mean, a band placed two standard deviations from that mean would have a familiar approximate coverage interpretation. Financial prices rarely satisfy those assumptions: returns can be skewed, heavy-tailed, autocorrelated, and subject to volatility clustering. Moreover, the bands are calculated from a rolling sample and are applied to a price series whose level is often non-stationary. The result is a descriptive envelope, not a reliable probability forecast. A price outside a boundary may be unusual within the recent sample without being extreme in the broader market process.
Evidence, debate, and practical use
Research on technical trading strategies has produced mixed conclusions, and the apparent usefulness of a rule can vary by market, sample period, transaction costs, execution assumptions, and data treatment. One study reported that the predictive power associated with band strategies diminished after their publication and largely disappeared in later international evidence, a result consistent with competition reducing a once-visible anomaly.4 Other analyses find that outcomes depend heavily on the asset and regime, with false signals particularly common when a method is applied without a trend or volatility filter.5 These findings do not make the indicator useless; they show that a chart pattern is not the same as a durable economic edge.
A robust workflow begins by defining the signal before looking at results. Specify the price field, period, multiplier, entry condition, exit condition, transaction costs, financing, slippage, and treatment of gaps. Test the rule on separate development and evaluation samples, then examine performance across assets and regimes rather than relying on one successful chart. Avoid treating a boundary touch as an executable price when the signal is only known at the close. For live use, record whether the trade followed a breakout, a confirmed rejection, or a broader portfolio rule, because otherwise a discretionary process can appear more systematic than it really is.
Used carefully, Bollinger Bands remain a compact way to organise three questions: where price sits relative to its recent average, how variable recent prices have been, and whether a change in dispersion is altering the trading environment. They become more informative when paired with an explicit trend measure, momentum confirmation, liquidity checks, and risk controls. They become misleading when upper and lower boundaries are treated as automatic sell and buy levels. Their lasting relevance lies less in predicting the next price than in forcing analysts to distinguish relative extremity, volatility, and direction before committing capital.
References
1. What Are Bollinger Bands? – Fidelity – https://www.fidelity.com/learning-center/trading-investing/technical-analysis/technical-indicator-guide/bollinger-bands
2. A complete explanation of Bollinger Bands – https://www.bollingerbands.com/bollinger-bands
3. Bollinger Bands | ChartSchool | StockCharts.com – https://chartschool.stockcharts.com/table-of-contents/technical-indicators-and-overlays/technical-overlays/bollinger-bands
4. Bollinger Bands – Wikipedia – 2005-03-08 – https://en.wikipedia.org/wiki/Bollinger_Bands
5. Mastering Bollinger Bands: Indicating Market Volatility – 2003-11-23 – https://www.investopedia.com/terms/b/bollingerbands.asp
6. John Bollinger’s Official Bollinger Band Website – https://www.bollingerbands.com/home
7. ijfans international journal of food and nutritional sciences – https://www.ijfans.org/uploads/paper/4ce93d6481d5f510b74c38e830c74a41.pdf
8. Bollinger Bands Explained: Formula & Trading Strategies – Britannica – 2025-01-01 – https://www.britannica.com/money/bollinger-bands-indicator
9. [PDF] NEPSE in Bollinger Bands – AIMS Press – https://www.aimspress.com/aimspress-data/nar/2021/4/PDF/NAR-03-04-023.pdf
10. The Accuracy Test of Technical Analysis of Moving … – https://media.neliti.com/media/publications/411842-the-accuracy-test-of-technical-analysis-9f092941.pdf
11. Buy the Panic: What 20 Million Bollinger Tests Found – TradingView – 2026-04-04 – https://www.tradingview.com/chart/ES1!/FwfqO55g-Buy-the-Panic-What-20-Million-Bollinger-Tests-Found/
12. Profitability of Technical Trading Strategies – arno – https://arno.uvt.nl/show.cgi?fid=136773
13. Bollinger bands explained: definition, examples and strategies – 2023-05-03 – https://www.stonex.com/en-gb/news-and-analysis/bollinger-bands/
14. Popularity versus Profitability: Evidence from Bollinger Bands – 2014-08-20 – https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2484322
15. Bollinger Bands Explained. Rules to use … – https://www.bollingerbands.com/bollinger-band-rules
