The Bollinger Band Mean Reversion Strategy

A Band That Breathes With the Market

Bollinger Bands consist of a moving average, typically a 20-period simple average, plotted alongside two bands set a fixed number of standard deviations above and below it, most commonly two. (The Technical Analysis → Indicators section covers the full calculation and the concept of standard deviation; this page focuses on trading the bands themselves.) Because the bands are built from standard deviation rather than a fixed percentage, they widen automatically during volatile periods and narrow during quiet ones, breathing with the market's own changing character.

The Bollinger Band Mean Reversion Strategy's core idea is that price touching or briefly piercing the outer band, in a market that is genuinely range-bound rather than trending, represents a short-term statistical extreme that tends to snap back toward the middle band. This is conceptually similar to the RSI Mean Reversion Strategy covered earlier on this shelf, but it reads the extreme through price's distance from its own recent average rather than through a separate momentum oscillator.

The single word doing the most work in that description is range-bound. Bollinger Bands behave very differently, and this strategy performs very differently, once a market begins trending strongly, which this page addresses directly and without softening.

Band-Walking: The Single Most Important Concept on This Page

Understanding band-walking is essential to trading this strategy safely, and it deserves its own explanation before any rule is given. In a genuinely range-bound market, price oscillates between the upper and lower bands, touching each one briefly before reverting back toward the middle, exactly the behaviour this strategy is built to trade. In a genuinely trending market, price instead begins riding along one band for an extended stretch, repeatedly touching or hugging the upper band during a strong uptrend, or the lower band during a strong downtrend, without meaningfully reverting toward the middle for many sessions at a time.

That second behaviour, band-walking, looks superficially similar to the mean-reversion setup at first glance, price is touching the outer band, but it is actually the opposite condition entirely. A trader who sells every touch of the upper band during a genuine, strongly trending uptrend, expecting a reversion that a trend actively resists, will be stopped out repeatedly while the trend simply continues.

Distinguishing the two requires checking the broader trend before acting on any band touch, using exactly the same longer-term moving average filter taught elsewhere on this site. A market trading sideways relative to its own 50-day average is a genuine mean-reversion candidate; one trading strongly above or below a rising or falling 50-day average is very likely in a band-walking regime instead, where this strategy's core signal should be avoided.

Setting the Bands for Indian Instruments

The standard 20-period, 2-standard-deviation setting is a reasonable default across most liquid Nifty 500 stocks and the major indices, and there is little reason to deviate from it without a specific, tested reason. For intraday trading on Bank Nifty, some traders shorten the period toward 15 on a 15-minute chart, trading off some smoothness for faster reaction on a shorter timeframe.

India VIX offers a useful, India-specific context layer for this strategy. During periods of unusually low India VIX, Bollinger Bands narrow considerably, and a touch of the outer band during that narrow-band period sometimes represents a genuinely smaller statistical move than the same touch during a period of elevated India VIX, where the bands themselves are already wide. Checking the current band width relative to its own recent history, not just whether price has touched the band, adds useful context before entering.

The Playbook, Rule by Rule

The regime filter comes first in this rule set for good reason; everything else depends on getting that judgement right.

• Setup condition: Confirm the instrument is genuinely range-bound, trading roughly sideways relative to its own 50-day moving average over the past several weeks, with no clear band-walking behaviour on either side visible in that period.
• Trigger: Price touches or briefly pierces the lower Bollinger Band and closes back inside it, showing a rejection rather than a continuation below the band.
• Entry: Enter on the close of the rejection candle, or on the next session's open if that candle closed too late in the session to act on comfortably.
• Stop-loss: Place the stop just below the low of the rejection candle, or a further band-width below entry if the instrument's typical volatility calls for extra room.
• Exit: Take profit at the middle band, the 20-period moving average, or a partial exit there with the remainder trailed toward the upper band if the range appears to be widening rather than closing.
• Position size: Risk one to one and a half percent of trading capital on the distance between entry and the stop-loss placed just below the rejection candle.

This strategy's short, modest target, the middle band rather than the opposite outer band, reflects its purpose honestly. It is built to capture a reversion to the mean, not a full swing across the entire range, and traders who hold for the opposite band routinely give back a meaningful share of the gain this strategy is actually designed to capture.

A Reversion on the ICICI Bank Chart

Illustrative example built to show the mechanics, not a specific historical trade record.

Consider an illustrative example on ICICI Bank during a period when the stock traded largely sideways, oscillating between roughly 1,180 and 1,260 for close to two months, with its 50-day average essentially flat throughout, a clear range-bound regime by this strategy's own filter.

During that stretch, the stock touches its lower Bollinger Band near 1,195, dipping intraday to 1,188 before closing the session at 1,203, a clear rejection candle. The trader enters near 1,206 on the next session's open. The stop is placed at 1,182, just below the rejection candle's low, a risk of roughly 24 points.

Over the following six sessions, the stock recovers toward its middle band, closing near 1,244, where the trader books the position. The trade captured roughly 38 points on a 24-point risk, a reward-to-risk near 1.6 to 1, a modest but consistent result typical of this strategy's short, mean-reversion-focused objective within a genuine range.

Reading Band Width as a Regime Signal in Its Own Right

Beyond the basic touch-and-reject signal, the width of the bands themselves carries useful information. A period of unusually narrow bands, sometimes called a squeeze, often precedes a sharp expansion in volatility, though that expansion can resolve in either direction and is not, by itself, a trading signal for this strategy. A squeeze followed by a decisive break of either band, accompanied by genuinely strong volume, is more often the start of a new trending regime, exactly the condition this mean-reversion strategy should stand aside from, rather than a mean-reversion opportunity.

Recognising a squeeze in progress is therefore useful primarily as a warning to be more selective, not less, about taking mean-reversion signals during that specific period, since the eventual resolution of a squeeze frequently produces exactly the band-walking behaviour this strategy is built to avoid.

Costs, Frequency and the Regime-Switching Risk

This strategy's signal frequency depends heavily on how much of the calendar year a given instrument spends genuinely range-bound versus genuinely trending, which varies considerably across Indian large caps and indices. A large-cap stock or index that spends a meaningful share of the year in a defined range can generate a reasonable number of signals across that period, each carrying brokerage, exchange charges, GST, stamp duty, and slippage on both entry and exit.

The genuinely dangerous cost with this strategy is not ordinary trading cost drag; it is the risk of a range quietly transitioning into a trend without the trader noticing in time, continuing to apply mean-reversion rules to what has become a band-walking regime. Reviewing the regime filter freshly before every single signal, rather than assuming an instrument that was range-bound last month remains range-bound this month, is the single most important cost-control habit for this strategy specifically.

What Bollinger Band Content Usually Skips

Bollinger Bands are among the most widely used indicators in retail trading content, and the band-walking distinction is precisely the part that content most often glosses over. Here is the rest of what usually goes unmentioned.

• Band-walking is not a rare edge case; it is the ordinary, expected behaviour of Bollinger Bands during any genuine, sustained trend, which means this strategy's core signal is fundamentally unsuitable for a meaningful share of any given year's total price action.
• A Bollinger Band squeeze, a period of unusually narrow bands, is frequently presented in social media content as a bullish signal on its own, when it is actually direction-neutral; the eventual breakout from a squeeze can resolve in either direction with roughly comparable frequency.
• The middle band exit target is deliberately conservative, and traders chasing the opposite outer band as a bigger target routinely convert a reliable, modest mean-reversion trade into a much lower-probability bet that the range will fully complete before reverting again.
• Combining Bollinger Bands with RSI, rather than treating either in isolation, has historically filtered out a meaningful share of the weaker single-indicator signals, a refinement that basic introductory content on either indicator rarely covers together.
• The traders who use this strategy well over years are rarely the ones who catch the most touches of the outer band. They are the ones disciplined enough to recognise a band-walking regime early and step aside from the signal entirely until the range genuinely returns.

Before risking real capital, paper trade this exact rule set on ten signals and log every one, win or lose, in your Trading Diary. Patterns you cannot see while trading, you can see clearly while reviewing.