Combine Bollinger Bands

What the bands measure

The construction is worth knowing because it explains why the outer lines move and what a touch of one actually represents.

A middle line is a moving average. The outer lines sit a multiple of the recent standard deviation above and below it. Standard deviation is a measure of dispersion, so when recent closes have been spread out the bands are wide, and when they have clustered the bands are narrow. Everything else follows from that.

Volatility envelope

The width is the primary reading. Wide bands say the market has been moving; narrow bands say it has not. Both states are informative, and the transition between them is more informative still, because volatility tends to alternate rather than persist indefinitely.

Standard-deviation width

Because the outer lines are a statistical distance from the average, price spends most of its time inside them by construction. That is a property of the arithmetic rather than a discovery about markets. It also means a touch of the outer band is a relatively ordinary event, not a rare one, and treating it as rare is the source of most band-based losses.

Mean reversion idea

The intuition that price returns to its average is reasonable in a market with no direction and misleading in one with a direction. In a range, the middle line sits in the middle of the range and price oscillates around it. In a trend, the middle line follows the trend, so a return to it is a pullback rather than a reversal, and the two look identical on the indicator alone.

• Width tells you the regime. Read it before anything else.
• Touches are common. The bands are drawn so that they are.
• The middle line is a moving average and carries every property one has, including lag.

Read the width first. The bands describe volatility, and their level meaning is secondary.

Read band behaviour

Three behaviours cover almost everything the tool has to say, and each maps to a different market condition.

Those behaviours are compression, expansion and the walk. Learning to name which one is in front of you is more valuable than any entry rule built on the bands.

Squeezes and expansion

A squeeze is a stretch where the bands narrow noticeably: recent closes have clustered, and participants are agreeing on price. Agreement is temporary, so a squeeze is usually followed by an expansion. What a squeeze does not tell you is the direction of that expansion, and this is the most frequently overstated claim about the tool. The honest version is that a squeeze tells you to expect movement and to have your levels drawn, not which way to lean.

Band touches

A touch means price has reached a statistical distance from its recent average. In a quiet, range-bound market that is often the edge of the range and worth watching. In an expanding market it is simply where a strong move takes price, and it will be touched repeatedly as the move continues.

Walking the band

The walk is the behaviour that ends fading strategies. During a strong directional move price can ride along the outer band for many intervals, touching it repeatedly without turning. Each touch is a signal by a naive rule, and each one is a counter-trend entry into the strongest part of the move. Recognising a walk in progress and standing aside is worth more than any refinement of the entry.

Name the behaviour first: squeeze, expansion, walk or oscillation. The right response differs for each.

Avoid the touch trap

The touch trap is the belief that reaching the outer band is by itself a reason to trade against the move, and it is the most expensive misreading of this tool.

It is an easy belief to acquire, because in quiet conditions it appears to work repeatedly. Then the market develops a direction and the same rule produces a run of losses that feels inexplicable.

Touch is not a signal

A touch is a location. It tells you where price is relative to its recent average, and nothing about whether the move has finished. Converting a location into a trade requires an event: a rejection candle, a close back inside the band, a failure to make a new extreme. Any of those turns "price is at the band" into "price reached the band and was pushed away", which is a different observation.

Trend continuation

In a directional market, band touches are a symptom of the direction. They occur because the move is strong enough to carry price a statistical distance from its own mean, and they will keep occurring for as long as that is true. Fading them is fading strength, and it produces the same distribution of results as fading extreme oscillator readings, for the same underlying reason.

Needing confirmation

• Require a close back inside before treating a touch as a rejection.
• Check the width. Fading in expanding conditions is fading a move that is still accelerating.
• Check structure. A touch against the higher-timeframe direction is the weakest version of the setup.
• Count the touches. A third or fourth consecutive touch is evidence of a walk, not of exhaustion.

Those four checks cost you most of the trades a naive band rule would take, and that is the point. You can watch a squeeze resolve on the practice charts and log both versions side by side for a week; the difference in trade count alone usually settles the argument.

Never trade a touch. Trade the rejection of a touch, confirmed by a close back inside.

Combine with other tools

Bands pair well with tools that measure something they do not, and badly with anything else derived from the same moving average.

The pairing test is the same as everywhere on this site: does the second input come from an independent observation, or is it arithmetic on the prices that already produced the first?

Candles at the band

This is the strongest combination available here. The band supplies a statistically meaningful location; the candle supplies an event at that location. A pin bar rejecting the outer band in a market with no clear direction is a two-part setup with independent components, which is more than most band rules offer.

Oscillator confluence

A momentum read adds something the bands do not measure, and it is a fair pairing as long as you remember that both will show extremes during a strong move. Requiring divergence rather than a threshold reading makes the combination considerably more selective, because divergence describes fading force while a threshold describes present strength.

Trend filter

The most valuable addition is the one that decides whether band-fading is permitted at all. Read structure on a higher chart, record trending or ranging, and allow counter-move band trades only in the ranging case. In the trending case, the bands still contribute: the middle line becomes a pullback zone, and touches of the outer band in the trend direction confirm strength rather than inviting a fade.

Pair bands with candles or divergence. Never with a second tool derived from the same average.

Bollinger takeaways

The tool is dependable when read as a volatility gauge and unreliable when read as a set of levels to fade.

A volatility lens

The width is the reading that survives across market conditions. Narrow bands say expect movement; wide bands say movement is already here. Neither says which direction, and a rule set that respects that limitation will avoid most of the trouble this indicator causes.

Touches mislead alone

Reaching the outer band is an ordinary event that the construction guarantees will happen regularly. In quiet conditions it can mark a range boundary. In directional ones it marks strength. Since both look identical on the indicator, the regime read has to come from somewhere else, and that requirement is not optional.

Confirmation matters

• Width before location. Regime first, entry second.
• Close back inside as the minimum evidence of a rejection.
• Squeeze means prepare, not predict. Have levels ready; let the expansion pick the side.
• Log the walks. They are the condition that breaks the rule, so they are the ones worth counting.

As with every tool on this site, no success figure accompanies any of this. How often a squeeze precedes a usable expansion, or how often a confirmed band rejection resolves in your favour, depends on the instrument, the horizon and your definitions. This desk has measured none of it. Fix your definitions in writing, keep them constant for a month, and let your own record answer the question.

Bands tell you how much the market is moving. Direction and timing have to come from somewhere else.