recognition, and for much more. Bands are usually thought of as employing a measure of central tendency as a base such as a moving average, whereas envelopes encompass the price structure without a clearly defined central focus, perhaps by reference to highs and lows, or via cyclic analysis. We'll use the term trading bands to refer to any set of curves that market technicians use to define high or low on a relative basis.
 
The earliest example of trading bands that I have been able to uncover comes from Wilfrid Ledoux in 1960. He used curves connecting the monthly highs and lows of the Dow Jones Industrial Average as a long-term market-timing tool. After Ledoux the exact sequence of trading band development gets foggy. In 1960 Chester Keltner proposed a trading system, The 10-Day Moving Average Rule, which later became Keltner bands in the hands of market technicians whose names we do not know. Next comes the work of J. M. Hurst who used cycles to draw envelopes around the price structure. Hurst's work was so elegant that it became a sort of grail with many trying to replicate it, but few succeeding. In the early '70s percentage bands became very popular, though we have no idea who created them. They were simply a moving average shifted up and down by a user-specified percent. Percentage bands had the decided advantage of being easy to deploy by hand. At any given time a 7% band consists of a base moving average, an upper curve at 107% of the base and a lower curve at 93% of the base. (Arthur Merrill suggested multiply and dividing by one plus the desired percentage.) When I started using trading bands percentage bands were the most popular bands by far. Along the way we got another fine example of envelopes, Donchian bands, which consist of the highest high and lowest low of the immediately prior n-days. Those are the main types of band/envelopes that I know of. Over the years there have been many variations on those ideas, some of which are still in use. Today the most popular approaches to trading bands are Donchian, Keltner, Percentage and, of course, Bollinger Bands.
 
Percentage bands are fixed, they do not adapt to changing market conditions; Donchian bands use recent highs and lows and Keltner bands use Average True Range as adaptive mechanisms. Bollinger Bands use standard deviation to adapt to changing market conditions and thereby hangs a tale. When I became active in the markets on a full time basis in 1980 I was mainly interested in options and technical analysis. Information on both was hard to obtain in those days but I persisted; with the help of an early microcomputer I was able to make some progress. At the time we used percentage bands and compared price action within the bands to the action of supply/demand tools like David Bostian's Intraday Intensity. A touch of the upper band by price that was not confirmed by strength in the oscillator was a sell setup and a similarly unconfirmed tag of the lower band was a buy setup. The problem with that approach was that percentage bands needed to be adjusted over time to keep them germane to the price structure and the adjustment process let emotions into the analytical process. If you were bullish, you had a natural tendency to draw the bands so they presented a bullish picture, if you were bearish the natural result was a picture with a bearish bias. This was clearly a problem. We tried reset rules like lookbacks with some success, but what we really needed was an adaptive mechanism. I was trading options at the time and had built some volatility models in an early spreadsheet program called SuperCalc. One day I copied a volatility formula down a column of data and noticed that volatility was changing over time. Seeing that, I wondered if volatility couldn't be used to set the width of trading bands. That idea may seem obvious now, but at the time it was a leap of faith. At that time volatility was thought to be a static quantity, a property of a security, and that if it changed at all, it did so only in a very long-term sense, over the life of a company for example. Today we know the volatility is a dynamic quantity, indeed very dynamic.
 
After some experimentation I settled on the formulation we know today, an n period moving average with bands drawn above and below at intervals determined by a multiple of standard deviation (We use the population calculation for standard deviation). The defaults today are the same as they were 35 years ago, 20 periods for the moving average with the bands set at plus and minus two standard deviations of the same data used for the average. But they weren't “Bollinger Bands” yet, that would come later when Bill Griffeth, an on-air host for the Financial News Network, asked me what I called my bands on air. I had presented a chart showing an unconfirmed tag of my upper band and explained that the first down day would generate a sell signal. Bill then asked me what I called those lines around the price structure, a question that I was totally unprepared for, so I blurted out the alliteratively obvious choice: “Bollinger Bands.”
 
So what are Bollinger Bands? They are curves drawn in and around the price structure usually consisting of a moving average (the middle band), an upper band, and a lower band that answer the question as to whether prices are high or low on a relative basis. Bollinger Bands work best when the middle band is chosen to reflect the intermediate-term trend, so that trend information is combined with relative price level data.
 
Soon the Bollinger Bands had company, I created %b, an indicator that depicted where price was in relation to the bands, and then I added BandWidth to depict how wide the bands were as a function of the middle band. For many years that was the state of the art: Bollinger Bands, %b and BandWidth. Here are a couple of practical examples of the usage of Bollinger Bands and the classic Bollinger Band tools along with a volume indicator, Intraday Intensity: