Bollinger Bands: Settings, Squeeze and What the Data Shows
Bollinger Bands explained: the 20/2 formula, %B and bandwidth, the squeeze, walking the bands and a W-bottom, with hit rates from our own hourly forex data.

This site contains affiliate links. We may earn a commission if you sign up through our links. How we research
Bollinger Bands are a volatility indicator made of three lines: a 20-period simple moving average in the middle, and an upper and lower band set 2 standard deviations away from it. The bands widen when price becomes volatile and narrow when it goes quiet. In our own forex data they were excellent at showing volatility and poor as a stand-alone "price touched the band, so reverse" signal.
Disclosure: OnForex may earn a referral commission from Exness links on this page. This is general education, not investment advice.
This guide covers the formula and settings, how to read %B and bandwidth, the squeeze, what happens when price walks the bands, the W-bottom idea, a worked example on a real chart, and what we measured on EURUSD, GBPUSD, USDJPY and gold.
How Bollinger Bands are built
John Bollinger developed the bands in the 1980s. The calculation has three steps:
- Middle band = simple moving average (SMA) of the last 20 closes.
- Standard deviation (SD) = how far those 20 closes typically sit from their average. Large swings give a large SD; small, tidy moves give a small one.
- Upper band = middle + 2 x SD. Lower band = middle - 2 x SD.
Because the SD is recalculated every candle, the bands "breathe". If closes were normally distributed, about 95% of them would sit inside the bands. Real price is not that tidy, so the figure is only a rough guide, but it explains why a close outside the bands feels unusual.
| Setting | Typical use | Effect |
|---|---|---|
| 20, 2 | Default for most charts | Balanced; outside-band closes are fairly uncommon |
| 20, 2.5 | Fewer, more extreme touches | Wider bands, fewer signals |
| 50, 2.1 | Slower trend view | Smoother, reacts later |
| 10, 1.9 | Short-term trading | Narrow and busy |
There is no "correct" row. Longer periods smooth more and react later; larger multipliers make touches rarer. Change one number at a time and judge the change on the time frame you actually trade.
%B and bandwidth: the two numbers behind the squeeze
Bollinger also defined two derived values, and many platforms can plot them in a separate panel.
- %B = (close - lower band) / (upper band - lower band). It is 1.0 on the upper band, 0.0 on the lower band and 0.5 on the middle band. Values above 1 or below 0 mean the close is outside the bands.
- Bandwidth = (upper band - lower band) / middle band. It rises when volatility rises and falls when it drops.
A squeeze is a stretch when bandwidth is at a low relative to recent history. In our test we defined it as bandwidth in the lowest 10% of the last 120 bars.
Reading the bands in a trend
The most common mistake is treating the bands as fixed support and resistance. In a strong move, closes can sit on or outside one band for many candles in a row, because the moving average keeps catching up. This is called walking the band.

The chart shows both behaviours on GBPUSD. In mid-October candles hug the upper band while the market climbs: each close near the upper band was a sign of strength, not an automatic sell. Later the bands pinch to their narrowest point around 22-23 October, and the next large move starts from that quiet stretch.
A practical way to read it:
- Price walking the band + widening bands: momentum is strong. Fading it means trading against the trend.
- Price back inside the bands, near the middle band: the move has cooled; the middle band often acts as a pullback reference (see pullback trading).
- Narrow bands: a quiet period; larger movement is more likely soon, direction unknown.
Strategy ideas people search for
These are descriptions of how traders use the bands, not recommendations.
1. Mean reversion to the middle band
The idea: after a close outside the bands, price tends to return toward the average. The idea only suits ranging markets. In a trending market the same setup is a series of small gains followed by one large loss, because price keeps walking.
2. The squeeze breakout
Traders wait for a squeeze, then watch for the first close outside the bands and trade in that direction, with a stop on the other side of the narrow range. The problem is false first breaks: the first close outside can reverse within a candle or two. Many traders wait for a retest or a close beyond a recent swing instead.
3. The W-bottom (and M-top)
John Bollinger described a pattern in which price makes a low outside the lower band, rallies, then makes a second low that stays inside the lower band. The second low shows weaker selling momentum, and a move through the rally high confirms the pattern. An M-top is the mirror image at the upper band. The pattern is a way to combine the bands with price structure, which is why it is more reasonable than a plain "touch and reverse".
4. Trend following with the middle band
In an uptrend some traders buy pullbacks to the middle band while price stays in the upper half of the bands, and exit if closes drop into the lower half. This uses the bands as a trend filter rather than an overbought/oversold tool.
What our data shows
We tested the bands on hourly and 4-hour candles for EURUSD, GBPUSD, USDJPY and gold. Source: OnForex analysis of hourly prices from Yahoo Finance, January 2024 - October 2026 (gold: June 2024 - October 2026, COMEX gold futures). 4-hour candles are the hourly data resampled. Settings: 20, 2.
A close outside the band: does price come back to the middle? Yes, nearly always.
| 1-hour chart | EURUSD | GBPUSD | USDJPY | XAUUSD |
|---|---|---|---|---|
| Close below lower band: touched middle band within 24 bars | 95.4% | 92.9% | 92.2% | 94.1% |
| Close above upper band: touched middle band within 24 bars | 92.2% | 92.1% | 93.6% | 90.9% |
That reads like an edge, but it is mostly arithmetic. The middle band is a moving average of price; price oscillates around its own average, so touching it again within a day is the normal state. It does not tell you the trade was profitable, because you still need to know how far price went against you first.
Direction 24 bars later, 1-hour chart. To judge a signal we compare it with the baseline: the share of all bars where price was higher (or lower) 24 bars later. USDJPY and gold rose over the period, so their baselines sit above 50%.
| 1-hour chart, 24 bars later | EURUSD | GBPUSD | USDJPY | XAUUSD |
|---|---|---|---|---|
| Baseline: price higher | 47.7% | 51.5% | 56.5% | 53.8% |
| Close below lower band: price higher | 51.6% | 51.6% | 54.0% | 57.1% |
| Baseline: price lower | 52.3% | 48.5% | 43.5% | 46.2% |
| Close above upper band: price lower | 50.7% | 45.5% | 45.2% | 45.0% |
On the 4-hour chart, a close below the lower band was followed by a higher price 24 bars later 48.6% | 53.0% | 59.8% | 60.4% of the time, against baselines of 48.7% | 49.8% | 58.0% | 57.1%. The gaps are one to four points on samples of a few hundred events. With that sample size, differences of a few points are noise.

Squeezes do precede bigger ranges. After bandwidth fell into its lowest 10% of the previous 120 bars, the median high-to-low range over the next 24 one-hour bars was 6.67 | 6.74 | 5.56 | 5.97 ATR, compared with 5.48 | 5.47 | 4.92 | 5.00 ATR across all bars. That is roughly 15-25% larger. (ATR is the average true range, a standard measure of typical candle size; see ATR indicator.)
But squeezes do not tell you the direction. When we took the first close outside the bands after a squeeze, price continued in that direction 24 bars later in 52.1% | 48.7% | 54.3% | 50.9% of cases. That is a coin flip.
What to take from this:
- The bands are a good volatility and location tool. They tell you the market is quiet, stretched or normal.
- "Outside the band means reverse" was not better than baseline in our sample.
- A squeeze is a reason to prepare for movement, plan a stop that matches it, and avoid sizing up in a quiet market that can suddenly expand.
Worked example: sizing a stop from the bands
Suppose you plan a trade after the GBPUSD squeeze above. You do not know the direction, so you wait for a close above the narrow range and a retest. Say you buy at 1.3350 and place the stop 40 pips below, under the retest low. On a USD 1,000 account, risking 1% means USD 10 at risk.
Lot size = risk / (stop in pips x pip value per lot) = 10 / (40 x 10) = 0.025 lots, which most brokers round down to 0.02 lots (USD 8 risk). Numbers are illustrative; spreads widen at news and opens, and a stop can be filled worse than planned (slippage). See position sizing and stop-loss placement.
Because a squeeze is followed by larger ranges, a stop sized for the quiet period is more likely to be hit by normal noise once volatility expands. Sizing a stop in ATR rather than in a fixed number of pips handles this better.
Common mistakes
- Selling every touch of the upper band. In a trend, the band is a ceiling only until it is not. Check the trend before fading.
- Reading the squeeze as a direction signal. It signals possible expansion, nothing more. Our first-break continuation rate was about 50%.
- Trading the first candle outside the bands. The first break often reverses. Wait for confirmation or accept the lower hit rate.
- Ignoring the stop. "Price returns to the middle band 90%+ of the time" does not mean a trade wins 90% of the time; the loss when it does not return can be much bigger than the gain.
- Over-tuning settings. Trying dozens of period/multiplier pairs until the past looks perfect produces an overfit rule; see backtesting.
- Using the bands alone. Combine with market structure, support and resistance and a fixed risk per trade.
How to practise it
- Add Bollinger Bands (20, 2) and a separate bandwidth panel to a chart.
- Mark 20 past squeezes on one pair and record: what happened to the range, and which way did the first close outside the bands go?
- Mark 20 times price walked a band, and note how long it lasted.
- Only then test any entry rule, with a stop and a position size, on a demo account, with a risk rule from risk management, before using money.
Related guides: RSI, moving averages, MACD, technical analysis overview and trend trading.
Summary
Bollinger Bands answer a simple question: is price quiet, normal or stretched compared with its recent past? They are very good at that. They are not a reversal detector. In our data, price returned to the middle band about 90-95% of the time, but closes outside the bands were not followed by reversals more often than the baseline, and squeezes preceded 15-25% larger ranges without revealing the direction. Use the bands to understand volatility, size your stop to it, and let structure and risk rules make the decisions.
Practise on a demo account first
Mark squeezes, band walks and stops on a free Exness demo account before you risk money. Forex and CFD trading carries a high risk of loss.
Continue to ExnessForex and CFD trading involves significant risk. You may lose your invested capital.
Risk warning: Forex and CFD trading carries a high risk of loss and is not suitable for everyone. Past statistics do not predict future results, and nothing here is investment advice or a signal.
Related Articles

Parabolic SAR: Settings, Formula and Backtest Results
Parabolic SAR explained: the formula, 0.02 / 0.2 settings, how to read the dots, why it whipsaws, and what a flip-every-signal test showed on four markets.

Ichimoku Cloud Explained: Five Lines, Signals and Settings
Ichimoku cloud explained: the five lines, how to read the cloud, sanyaku kōten and gyakuten, the 9-26-52 settings and what our daily data shows.

ADX Indicator: Settings, +DI/-DI and What Our Data Shows
ADX indicator guide: how ADX and +DI/-DI are calculated, what 20, 25 and 40 mean, how to use ADX to pick a method, and what our data shows on four markets.

RSI Indicator: Settings, Strategies and What Our Data Shows
RSI indicator guide: Wilder's formula, the 14 setting, overbought and oversold levels and divergence, with tests on four markets of what RSI can do.

Moving Averages in Forex: SMA, EMA, 200-Day Data
Moving averages in forex: SMA vs EMA formulas, 50/200 golden and death crosses, common periods, and what 9 years of daily data show about the 200-day line.

Pin Bar: How to Spot, Trade and Test a Pin Bar Candle
Pin bar explained: bullish and bearish pin bar rules, where they matter, a worked trade with stop and sizing, plus how pin bars performed in our 4-market data.