System Trading Made Easy With John Bollinger
John Bollinger is an analyst, author, and president and founder of Bollinger Capital Management, an investment management company that provides technically driven money management services. Bollinger Capital Management also develops and provides proprietary research for institutions and individuals. He may be best known as the creator of Bollinger Bands, which he developed in the mid-1980s. Since then, investors and traders have come to view Bollinger Bands as a reliable tool for assessing expected price action, and the bands are featured on most financial charting software and websites.
Bollingerís Capital Growth Letter, which has been published since 1987, provides market commentary, charts, and investment advice for the average investor employing a technically driven asset allocation approach. He has also developed several investor websites, among which are BollingerBands.com, EquityTrader.com,
BollingerOnBollingerBands.com, BBScript.net, and a forex website, BBForex.com.
John Bollinger, a longtime proponent of rational analysis, the combination of technical and fundamental analysis, was the first analyst to earn both Chartered Financial Analyst (CFA) and Chartered Market Technician (CMT) designations.
S&C Editor Jayanthi Gopalakrishan and Staff Writer Bruce R. Faber interviewed John Bollinger via telephone on January 4, 2012.
John, as you and your work are well known in our industry, can you begin by telling us the most recent things youíve been working on?
A little more than two years ago, I had the pleasure of spending time with Ian Woodward, who has been doing some interesting things with Bollinger Bands. It was an inspiring time for me. Over the next few weeks, I realized that I could develop an entire framework of indicators around Bollinger Bands that covered virtually all of the traditional technical analysis tasks.
Where did you start?
The first result was an indicator called BB Impulse, which quantifies price action in relation to the bands, and from there I went on to develop a trend indicator. I next developed a momentum indicator, and kept expanding the horizon until I had a complete suite of indicators that were all logically consistent and drawn from the same basic pool of technical information. Iíve ended up with an array of new tools. I have been able to leave behind some older tools and replace them with better ones. It has been a great two years.
The two projects I worked on went way back for me. When I created Bollinger Bands, I created two indicators to go with them, %b and BandWidth. %b tells us where we are in relation to the bands. If prices are at the upper band, %b is 1 and if prices are at the lower band, %b is zero. If prices are at the middle band, %b is 0.5. BandWidth tells you how wide the bands are. In order to measure that, we use the middle band, or the moving average, as a reference point. Those were the two building blocks we had to start with.
And what was the second project?
About that time I started thinking about a language in which we could express these ideas and exchange these ideas with other technicians and others who are interested in developing systems and approaches to the market. Using as a basis a logic we employed, I created a simple computer language and called it ďTrade.Ē To get there we tried to overcome the disadvantages of traditional logic for traders. In traditional logic, everything is black or white, one or zero, yes or no, long or short.
But that does not really model how traders think about the markets. So I worked with an engineer friend and developed a tristate logic, which was 1, 0, and ‑1. We could map that better onto the trading process ó that is, long, flat, or short. With that, we had a logic that modeled what traders did better. I wrote Trade to be able to communicate those ideas. All of these parts and pieces started to come together in the past couple of years.
At some point I abandoned the tristate logic model because I realized that even that was not sufficient to describe what traders really did. Traders often scale in or out of positions, or are 20% long, or 20% short, or some variation. So we combined the ideas of fuzzy logic from the original tristate model and produced a smooth-valued logic that describes how traders function in real life.
Then the final piece was to write the language as a computer interface to express our ideas in a straightforward way. We wrote a compiler that would interpret those ideas and put them on the screen.
Itís been a very, very active two years, during which I pulled together a large number of threads from past years and combined them into a powerful framework for traders.
Sounds like a lot of work. So how would a trader use something like that?
It was an awful lot of ideas in a very short time, so it might be better to focus on a few of the indicators and how they might be used.