Bollinger Bands: how to use them in crypto trading

Bollinger Bands show one thing: how far price has drifted from its own average. The indicator promises nothing beyond that. It doesn't tell you where price is going, it doesn't draw an entry point, and it doesn't replace volume analysis.

The problem is that 90% of articles about Bollinger Bands sell it as an entry system. In reality it's a tool for measuring volatility, and it only works paired with something else: volume, RSI, the order book. We'll break the indicator down the way traders who actually trade crypto use it, not the way a technical analysis textbook does. Below: the full mechanics of the indicator, and how to use Bollinger Bands across different timeframes, from swing trading to scalping.

Bollinger trading without an understanding of these mechanics usually turns into guessing reversals off a line touch, and that's exactly what we're going to get rid of.

What Bollinger Bands are

Bollinger Bands are a volatility indicator developed by trader John Bollinger in the early 1980s. It builds a channel around price whose width changes depending on how actively the market is trading the asset right now.

The idea is simple. When the market is quiet, the bands contract. When a move starts with rising volume, the bands expand. The indicator adapts to the market in real time, unlike static support and resistance levels.

For crypto this matters especially. BTC and altcoins can sit in a lull for hours, then cover a move in 15 minutes that would take equities a week. A static channel is useless in that situation because it can't adjust in time. Bollinger Bands adjust automatically, because the calculation runs off the last N candles.

The components: middle line, upper and lower bands

The Bollinger lines — the middle, the upper and the lower — together form the channel. The indicator consists of three lines.

The middle line is a simple moving average (SMA) over the chosen period, usually 20 candles. It shows the notional "center of gravity" of price over the recent past.

The upper band is the middle line plus two standard deviations of price over the same period.

The lower band is the middle line minus two standard deviations.

All the logic of the indicator lives between those three lines. The harder price swings up and down, the larger the standard deviation, and the wider the bands spread. If there's barely any movement, standard deviation drops and the bands contract almost flush against the middle.

The formula and standard deviation

The formula looks like this:

Middle line = SMA(N)

Upper band = SMA(N) + (K × standard deviation)

Lower band = SMA(N) - (K × standard deviation)

Where N is the period (20 by default) and K is the standard deviation multiplier (2 by default).

Standard deviation (denoted in statistics by the Greek letter sigma) is a measure of how widely price is scattered around the average. If price has moved in a narrow range over the last 20 candles, the deviation is small. If there have been sharp candles with large bodies in both directions, deviation grows and the bands widen.

The multiplier of 2 wasn't picked at random. Under a normal distribution, roughly 95% of values fall within plus or minus two standard deviations of the mean. Price, of course, isn't perfectly Gaussian, especially in crypto with its fat tails and sharp outliers. But the rule holds up well enough: price sits inside the bands about 85-90% of the time, and a move outside them is statistically a rarer event.

Some people use a 1.5 multiplier for a more sensitive reaction, others use 2.5 for a more conservative channel. On heavily volatile coins such as fresh listings, traders sometimes set 2.5-3 to filter out false band touches from random wicks.

Let's take a rough numerical example so the formula doesn't stay abstract. Say that over the last 20 candles on the SOL/USDT hourly chart the closing price ranged between 178 and 186, and the average (SMA20) came out at 182. The standard deviation across those 20 values was about 2.1. The upper band would then sit at 182 + (2 × 2.1) = 186.2, and the lower one at 182 - (2 × 2.1) = 177.8. If the next candle closes at 187, it formally exits the upper edge of the channel, and that's a statistically rare deviation that needs its own explanation through volume or the news background.

Standard deviation isn't calculated off random points but through the root-mean-square deviation of the closing price from the SMA over the same period. Terminals and trading platforms do this automatically, there's no need to calculate it by hand, but understanding the mechanics matters: the more "jagged" the price, the wider the scatter of candles around the average, the wider the bands will be on the chart.

Settings: period 20 and alternatives for crypto

The standard Bollinger settings are period 20, deviation 2. That's the classic setup, which appeared long before the crypto market and was calibrated for stocks and indices with their smoother dynamics.

For crypto the classic setup works, but it needs adjustments depending on the timeframe and trading style.

On the daily and 4-hour charts, period 20 performs fine, because 20 candles gather enough data for stable statistics and the market is less noisy on those timeframes.

On the hourly and 15-minute charts you can keep the period at 20, but sometimes it's shortened to 14-18 so the indicator reacts faster to shifts in the volatility regime that are typical for crypto.

On minute timeframes (M1-M5) used by scalpers, period 20 is often too sluggish. By the time the bands react to a volatility spike, half the move is already done. A period of 10-14 gives a faster reaction, although it increases the number of false signals.

I tested this on several pairs on Bybit and Binance Futures: on BTC/USDT, period 14 on M5 gave noticeably earlier squeeze signals than the classic 20, but it required stricter confirmation through volume, otherwise it picked up extra noise.

There's also a more fundamental reason why the classic settings need adaptation specifically in crypto. The stock market trades 6-8 hours a day, five days a week, and volatility there is smoothed out by overnight pauses and weekends. The crypto market trades 24/7 without stopping, there's no session "open" and "close" here, which means there are none of those natural pauses that would bring down accumulated deviation statistics. On top of that, altcoins, especially low-cap ones, are prone to sharp outliers from single large orders, something that almost never happens with stock market blue chips. As a result, standard deviation in crypto is on average higher and less stable, so many traders treat period 20 not as a universal constant but as a starting point for calibration.

The settings table below gives you a starting point, but the final calibration is something each trader usually tunes to the specific instrument.

Bollinger Bands: how to read them

The point of the indicator isn't the lines themselves but what happens between them. There are four basic states you need to recognize at a glance on the chart.

Band contraction (Squeeze): low volatility before a move

A squeeze is the moment when the bands converge almost flush against the middle line. Channel width drops to its lowest level over the last weeks or months of observation.

The reason is simple. The market has entered a phase of low activity, trading volume has dropped, participants aren't ready to commit. That kind of lull rarely lasts long, especially in crypto: sooner or later a critical mass of orders builds up and the market fires off in one direction.

A squeeze on its own doesn't tell you where price will go. It's only a warning: volatility is at a low, get ready for a move. Direction has to be determined from other signs: the trend on the higher timeframe, volume, where price sits relative to key levels.

I usually wait not for the squeeze itself, but for the first candle with a noticeable rise in volume after it. That candle is what most often points to the real direction of the break.

There's a practical detail that's easy to miss. A squeeze has no fixed duration. On some coins the contraction holds for 3-4 candles, on others it can stretch out for several days, especially on higher timeframes like D1 or H4. Waiting for the break while constantly checking channel width is awkward, so many traders use an additional indicator, BandWidth (band width as a percentage of the middle line), which visually shows whether the current contraction is at a low for the chosen historical period, for example the last three months. If BandWidth is printing a new low for that stretch, the odds of an imminent strong move are higher than with an ordinary local narrowing of the bands.

Expansion: the start of a trend

Expansion is the opposite of contraction. The bands start spreading actively, the middle line changes its slope, and a sustained directional move forms on the chart.

Expansion usually follows a squeeze, but it can also start without an obvious prior contraction, simply on the back of sharp news or a large order that pushed the market. Expansion is the phase in which a trend forms, not a reversal.

One detail matters here: while the bands are actively expanding and price is riding along one of them, trading against the move is statistically unprofitable. Band expansion signals the strength of the trend, not its end.

Touching and bouncing off the bands

The most common beginner mistake is treating a touch of the upper or lower band as an automatic reversal signal. "Price touched the upper band, so it's time to sell." That only works in a range.

In range conditions (sideways movement with no pronounced trend) a band touch does often produce a bounce back to the middle line. The logic goes like this: price has statistically deviated too far from the average, and in a calm market it usually returns to it.

But in a trend a band touch is a completely different thing. In a strong upward move price can "ride the band", meaning it touches the upper edge of the channel candle after candle without pulling back at all. A classic example: a strong altcoin pump after a listing on a major exchange, where price holds at the upper band for hours in a row.

A simple rule helps separate a "bounce in a range" from "riding the band in a trend": if volume keeps rising after the band touch and the tape shows no sign of the buyer fading, it's premature to wait for a reversal.

Moving outside the bands: breakout or reversal

Sometimes price doesn't just touch a band but moves outside it, fully breaking through the channel edge. That kind of move usually means one of two things: either the start of a strong impulsive trend, or an extreme outlier that will soon correct.

The key question is what volume the move happened on. If the band break comes with volume 2-3 times above average, it's most likely the start of a real trend rather than a random outlier. If volume stays low while the candle has a long wick, that's more often a sign of a manipulative move or a stop hunt, followed by a quick return into the channel.

In my experience this pattern breaks down about 30% of the time, specifically on illiquid alts, where a single large player can push price outside the band for a couple of minutes with no real market interest behind it.

Bollinger Bands trading strategies

Below are four approaches that are actually used in the crypto market. Each is built for its own market context, and mixing them carelessly isn't a good idea.

Squeeze and breakout (Bollinger Squeeze)

This is the basic trend strategy, built on the idea that after a lull the market will inevitably fire off.

The mechanics are simple. We wait for the bands to contract to their narrowest width over the last few weeks. We mark the level of the upper and lower channel edges at the point of maximum contraction. We wait for a candle that closes outside one of those edges with volume at least one and a half to two times above average.

Entry happens on the breakout candle or on the first pullback after it, in the direction of the break. The stop goes beyond the opposite band or slightly deeper than the last local extreme before the break. Take profit is usually calculated as a distance equal to the channel width at the moment of contraction, projected from the breakout point.

A trade example. ETH/USDT on H1, the bands contracted to a range of 0.4% of price (for comparison, the usual channel width on this pair is around 1.2-1.5%). Price sat in the 3180-3192 range for almost six hours. Then a candle printed with volume three times above average, closing at 3211, above the upper band. Entry at 3213, stop at 3175 (below the lower band at the moment of contraction), take profit at 3245 (channel width of about 32 points, projected from the breakout point). The trade closed at take profit 40 minutes later.

The key mistake with this strategy is entering at the moment of contraction itself, without waiting for the breakout to confirm. Bands can "sit" in a squeeze for hours, and an early entry often means a series of small stops before the real move.

Bouncing off the bands in a range

This strategy works exclusively in sideways movement and is absolutely contraindicated in a trend, which you need to understand upfront.

The logic: we establish that the market is in a range (price moves in a horizontal corridor with no pronounced slope in the BB middle line over the last 15-20 candles). On a touch of the upper band we look for a short with a target at the middle line. On a touch of the lower band we look for a long with a target at the middle line.

Getting additional confirmation matters, because a band touch alone isn't a reliable enough signal. Pairing it with RSI works well: if RSI is also showing overbought conditions (above 70) when the upper band is touched, the odds of a bounce are higher. If RSI hasn't reached an extreme yet, it's better to skip the signal.

The stop in this strategy goes outside the band, take profit at the middle line, less often at the opposite band when the channel is especially wide.

This strategy works most consistently on pairs with high liquidity and moderate volatility, where the market can genuinely stay in a range for a long time. On sharply trending memecoins or fresh listings, band bounces are much harder to catch.

A trade example for this strategy. BTC/USDT on H1 moved in the 66400-67800 range for three days straight, and the BB middle line was almost horizontal. Price approached the upper band at 67750, and RSI(14) was reading 74 at that moment. Short entry at 67700, stop at 67950 (outside the band plus a buffer), take profit at the BB middle line, which at the time ran at 67100. The trade closed at take profit in roughly three hours, with a risk-to-reward ratio of about 1 to 2.4.

It's worth spelling out separately when this strategy breaks. If the market gets hit with a strong news trigger while you're holding the position — an unexpected Fed rate decision, say, or a large liquidation event in derivatives — the range can flip into a trend right in the middle of the trade. In that situation the stop has to fire without exceptions; trying to "sit through" a breakout in a counter-trend position almost always costs more than the planned loss.

Bollinger Bands + RSI (double confirmation)

Pairing BB with RSI (the relative strength index) covers Bollinger's main weakness: on its own the indicator doesn't tell you whether the market is overbought or oversold at that moment, only how far price has deviated from the average geometrically.

RSI adds a second dimension, momentum, meaning the speed and force of the price move. The combination gives more reliable signals than either indicator on its own.

The setup is simple: price touches or breaks the upper BB band, and at the same time RSI goes above 70. That's double confirmation of overbought conditions, a signal for a possible short or for taking profit on a long. Symmetrically for the lower edge: a touch of the lower band plus RSI below 30 gives a stronger long signal than a band touch alone.

An important nuance: if price touches the band while RSI diverges from price (divergence, where price prints a new extreme and RSI doesn't), the signal gets even stronger. That's a classic sign of fading momentum.

The reverse situation is just as critical to understand. If price breaks the upper band while RSI keeps climbing along with it, never showing overbought above 70, that speaks to the strength of the trend, not a reversal. In that case it makes more sense not to short but to look for entries with the trend.

A divergence example in practice. ETH/USDT on H4 printed a new local high, closing at 3420, slightly above the upper BB band. But RSI(14) at that high read 68, whereas at the previous local peak a week earlier, at a similar price, RSI had climbed to 79. Price higher, RSI lower. That's a classic bearish divergence, and combined with the upper BB band touch the reversal signal came out noticeably stronger than if the trader had looked at just one of the two indicators.

Here the Bollinger indicator works not on its own but as part of a pair. For more on RSI itself and its settings, read the article "RSI indicator in crypto: how to set it up and use it".

BB + the order book

This is no longer a classic technical analysis strategy but an approach specific to active intraday trading using the order book.

The idea is not to rely on BB as the only source of a signal, but to use it as a filter: the indicator tells you where price is statistically overheated or undervalued, while the final decision is made off the actual orders in the book.

For example, price approaches the lower Bollinger band against the backdrop of a general range. On its own that's only a hint at a possible bounce. Next we open the order book and look at whether there's a meaningful density level below the current price, a large limit buy order. If the density level is real (not spoofing, meaning the order doesn't disappear as price approaches), and buys start appearing on the tape as price comes into it, the odds of a bounce off the lower band are noticeably higher than if we were going off the indicator alone.

The reverse example: price moves outside the upper band on volume, which by classic BB reading is a sign of a strong trend. But if there's a heavy sell density level in the book directly above price, and the tape starts to "fade" as price approaches it (large buys stop coming in), the move outside the band may turn out to be false, with a quick return into the channel.

This pairing removes the main problem with classic technical analysis in the crypto market: indicators are built on historical prices, while the order book shows real money sitting there right now. BB suggests a statistical probability, the order book confirms or refutes it through the actual interest of participants.

Bollinger Bands for scalping

Scalping places completely different demands on an indicator than swing trading does. On M1-M5 timeframes price makes several dozen moves an hour, and the classic BB settings often turn out to be too slow.

Settings for M1-M5

For scalping, the period is normally shortened to 10-14 instead of the classic 20. The reason is that over 20 one-minute candles the market can change its volatility regime several times, and an indicator built on such a long window will lag behind the market's real state.

The deviation multiplier is sometimes lowered from 2 to 1.5 as well, so the bands react more sensitively to short spikes. The flip side of that setup: the number of false signals grows, and the bands start "flickering" more often between contraction and expansion.

A practical compromise many scalpers use: period 10, deviation 1.5-2 on M1; period 14, deviation 2 on M5. Beyond that, calibration goes per coin, because the volatility of BTC and of an illiquid altcoin differ by multiples, and one setting won't fit both at once.

There's an important limitation worth stating honestly: on M1, BB throws off a lot of noise signals simply because of the nature of the minute timeframe, where random price swings of a fraction of a percent look like full-blown indicator moves. Using BB on M1 as the sole entry signal is risky. The indicator works better here as a context filter than as a standalone system.

BB + the tape in Secret Terminal

In practice scalpers rarely look at Bollinger Bands in isolation. The indicator gives statistical context: "price is far from the average right now, the odds of a correction are higher" or "volatility has contracted, a move is coming soon." The specific entry point and the confirmation of the move's strength come from order flow tools.

In Secret Terminal this is done by pairing the chart with Bollinger Bands and the order book, tape and cluster modules opened in an adjacent window of the workspace. The logic is the same as described above in the section on BB and the order book, just adapted to the speed of scalping.

Price approaches the lower band on M5. The indicator suggests that the market is statistically oversold on that timeframe. Next we look at the tape: if large buys appear as price comes into the band, the tape "accelerates" on the green side, and the cluster shows delta favoring the buyer, that confirms a potential bounce. Terminal lets you set a minimum trade size filter on the tape to cut out market noise and see only the trades that actually move price, which is critical specifically for scalping, where it comes down to seconds.

You can also use the terminal's global density module to check whether there's a large limit order near the Bollinger band that could either strengthen the indicator's signal or, conversely, refute it, if the density level sits on the opposite side from the expected move.

This approach removes the main weakness of classic indicators in scalping: they react to price that has already happened, while the order book and the tape show what's going on with real orders right in the moment.

For more on other indicators often combined for scalping, read the article "Indicators for crypto scalping". The general principles of scalping as a trading style are covered separately in the piece "Crypto scalping: the basic principles."

By the way, if you're still working out how the order book, clusters and the tape work in practice rather than in theory, we have a free trading course on YouTube. Working with the terminal interface and the order book + clusters + tape combination is covered in detail in lesson 3, and how professionals read the market through the order book and clusters is in lesson 4.

Mistakes when working with Bollinger Bands

Let's go through the typical mistakes traders make when they're just starting out with this indicator.

The first and most common: treating a band touch as an automatic entry signal without accounting for the broader trend. In a range that can work; in a trend that approach systematically drains the account, because price calmly rides along the band while the trader keeps opening positions against the move.

The second mistake: ignoring volume. Bollinger Bands are built on closing price alone, volume doesn't enter the formula at all. Which means band width says nothing about whether a move is backed by real interest from market participants or whether it's a wick from one large order. Without a look at volume or the tape, any reading of the indicator stays incomplete.

The third: using the same period and deviation across all timeframes and all coins with no calibration. Settings that work great on BTC on H4 can produce nothing but noise on an illiquid altcoin on M5. The indicator has to be calibrated to the specific instrument and trading style.

The fourth: confusing a band squeeze with a signal to enter immediately. A squeeze only tells you that potential for a move is building, not its direction and not the exact moment it starts. Entering before the breakout is confirmed by volume often leads to a series of stops before the real move gets going.

The fifth: trading BB in complete isolation from market context, ignoring news, major events like listings or token unlocks, the overall trend in BTC dominance. The indicator reflects only the price statistics of the last N candles; it knows nothing about fundamental factors that can sharply change volatility at any moment.

The sixth, specific to scalping: blindly trusting BB signals on M1 without confirmation through order flow. On a timeframe that short, random noise looks statistically identical to a real signal, and the only way to tell them apart is with additional tools like the order book and the tape.

The seventh mistake, often overlooked: rigidly tying take profit to the BB middle line without accounting for the fact that the middle line itself is constantly moving. If a trader opened a trade off the lower band with a target at the middle line, and while the position was held SMA20 shifted upward on the back of a general market rally, the final target ends up further away than assumed at entry. That isn't always a bad thing, but ignoring the movement of the middle line when calculating risk isn't wise, especially on volatile timeframes like M5 and M15, where SMA20 can shift noticeably in a matter of a few candles.

The eighth, psychological: sitting through a losing position hoping that price "has to come back into the channel." More than one deposit has been blown on exactly that logic by a trader who confuses statistical probability with a guarantee. The indicator tells you what happened on average over the last N candles, not what's obliged to happen in the next five minutes. A stop set by the rules of the strategy has to fire without any negotiation with yourself.

FAQ

• What do Bollinger Bands show in crypto trading?

  Bollinger Bands show the level of current volatility and the statistical range within which an asset's price usually fluctuates. The indicator doesn't forecast direction; it measures how far price has deviated from its average over the chosen period.

• Which Bollinger Bands period is better to use for crypto?

  The classic period of 20 works well for the daily and hourly charts. For scalping on M1-M5 the period is usually shortened to 10-14 so the indicator reacts faster to the sharp changes in volatility typical of the crypto market.

• Does a touch of the upper Bollinger band mean a sell signal?

  No, a band touch on its own isn't a sell signal. In a strong trend price can ride along the upper band for several candles in a row without reversing. You need additional confirmation through volume, RSI or order book data.

• What is a Bollinger Squeeze?

  A Bollinger Squeeze is a state where the bands contract to their narrowest width over the recent observation period. It's a sign of low volatility, which is usually followed by a sharp price move, though the indicator doesn't show the direction of that move.

• Can you use Bollinger Bands without other indicators?

  Technically yes, but the risk of false signals is noticeably higher. In practice BB is more effective paired with RSI, volume or order book analysis, which confirm the real strength of a move at the channel edges.

• Do Bollinger Bands work for scalping?

  Yes, but on M1-M5 the indicator often throws off a lot of noise signals. Scalpers usually shorten the period to 10-14 and always confirm BB signals through the tape and density levels in the order book, rather than relying on a single indicator.

• Why does price sometimes move outside the Bollinger Bands and not reverse?

  A move outside the band against a strong trend and rising volume is generally a continuation, not a reversal. The signal only becomes a reversal signal when volume and order book activity start visibly fading after the break.