Bollinger Bands
Last updated
Bollinger Bands draw a band around a moving average that is two standard deviations wide on each side, so the band widens when price swings more.
How Bollinger Bands are calculated
Middle band = 20-bar SMA of the close. Upper band = middle + 2 x standard deviation of the last 20 closes. Lower band = middle − 2 x the same. Two derived numbers are %b (where price sits in the band, 0 to 1) and bandwidth ((upper − lower) / middle). John Bollinger developed them in the 1980s.
How it is read
Price near the upper band means it is high relative to its own recent range. The bands are a description of volatility, not a prediction that price will turn at them.
Common mistakes
- Assuming 95 percent of closes must stay inside. That holds only for normally distributed returns, which prices are not.
- Selling at the upper band in a strong trend, where price can ride the band for many bars.
- Using different settings in two tools and comparing results.
Start with 20 days
Take the last 20 closing prices and average them. That is the middle line. Now measure how far the closes have scattered around that average (the standard deviation). Draw one line two of those distances above the middle and one two below. That is the whole indicator: a channel that follows price and breathes with it.
Worked example with made-up numbers
Say a stock's 20-day average is 100 and the standard deviation is 3. The upper band sits at 106 and the lower band at 94. If the stock closes at 95, it is near the bottom of its own recent range, with %b about 0.08. If volatility doubles to 6, the bands move to 112 and 88. Same average, a wider channel. The bands tell you how big recent moves have been, nothing more.
What the bands are good for
They are a quick picture of volatility: narrow bands mean a quiet stretch, wide bands mean a busy one. They also give a consistent way to say that price is high or low relative to its own recent range. They do not say that price must turn back, and a trend can ride the band for weeks.
Could a rule built on them still be luck
Yes. Any band rule can look good on one stretch of data by chance, which is why Tickfloor tests the rules on long histories, with trading costs, and corrects for how many rules were tried. The results are on the linked test pages below, not summarised here.
What Tickfloor tested
How we counted
Tickfloor's research desk has backtested 1,346 strategies, net of modelled trading costs. After correcting for the 1,778 tests run, 0 passed. That does not prove no strategy works. These are results from our older test setup, which was built before our current testing rules. They have not been redone under those rules.
Claim tests on Bollinger Bands
- Bollinger Band Bounce (claim test)
- Bollinger Band Squeeze (claim test)
- Bollinger Band Breakout (claim test)
Lessons that cover it
Related concepts
General information only. It doesn't consider your objectives, finances or needs. Tickfloor holds no financial services licence and never places trades.