By John Bollinger, CFA, CMT, Creator of Bollinger Bands*

 

Trading bands have been around for many years and come in many varieties. Sometimes they are called bands, sometimes envelopes, and, occasionally, channels. The methods of computation are all over the board, some adaptive, some fixed.

 

Whatever the name or method of computation, all trading bands serve the same general purpose; they define whether prices are high or low on a relative basis. The trader, armed with that information, can, then, make rigorous trading decisions. Those decisions might involve pattern recognition, confirmation (or lack thereof) with other technical indicators / methods, or they can focus on the information generated by the bands themselves.

 

A couple of examples will serve to illustrate the basic use of trading bands. For these illustrations, I'll stick to using my own bands with the default parameters -- a short explanation of Bollinger Bands can also be found at the end of this piece.

Example One, Pattern Recognition with Bollinger Bands

A W bottom or double bottom, as it is sometimes called, consists of a decline to a low followed by a recovery and a subsequent decline and turn back up that sets the stage for a sustainable advance.

 

The relationship between the two lows has been a topic of wide discussion in the technical community. "Should the first low be higher than the second? Should the lows be equal? The second higher? " and so on…I find that the process of rigorously identifying these patterns is much easier, and the patterns are made much clearer, if one considers the lows in relation to the Bollinger Bands and ignores the absolute questions.

 

If the first low is beneath the lower band, and the second low is at or above the lower band, you have a potentially interesting setup, a divergence where the second low is relatively higher than the first, regardless of the absolute levels involved. Add confirmation and discipline, and you have something worth working with, not an argument.

%b, an indicator that tells us where we are in relation to the Bollinger Bands, is the tool we use to clarify the patterns.

Chart 1, Pfizer, PFE

The key here is a relative comparison. Note that the stock makes a new low in July in absolute terms but not in relation to the Bollinger Bands. This is highlighted by %b, which falls below zero at the initial low but remains above zero at the July low.

Example Two, Bollinger Bands and Indicators

It was not pattern recognition that interested me when I first got involved with trading bands; it was technical signals generated by tags of the bands that went unconfirmed by technical indicators. For example, a tag of an upper trading band by an important stock market index that was accompanied by negative stock market breadth data.

Chart 2 (shown subsequently) depicts an advance that was confirmed by a technical indicator until its final stage. Just prior to the decline setting in, we get a tag of the upper band accompanied by a negative reading in the indicator. In this example, the indicator is a variation of David Bostian's Intraday Intensity developed by Marc Chaikin, 21-day Intraday Intensity %. II% is designed to depict the flow of money into and out of a stock

Chart 2, Norfolk Southern, NSC

Note the new high for NSC in May. It is a tag of the upper Bollinger Band accompanied by a negative reading for 21-day Intraday Intensity percent. Also note that the subsequent "throw-back rally" carried NSC right back to its breakdown point and offered a last chance to get out or a second chance to get short.

 

In a nutshell, relative highs and lows as a framework for rigorous decision making are what trading bands are all about. Why the emphasis on rigor? Because emotions are our worst enemies and employing rigorous tools and techniques is an effective way of keeping our emotions in check. But, that, as they say, is another story.

Good trading!

 

*Author’s Note: Bollinger Bands are curves drawn in and around the price structure. The base of the bands is a 20-period simple moving average that serves as an anchor for the bands and an indication of the trend. The bands themselves are spread above and below the average using a measure of volatility called standard deviation.

Standard deviation is calculated using the same data that was used for the moving average. The default period is 20, and the default width of the bands is plus and minus two standard deviations.

 

*Reprinted (and modified) with permission from John Bollinger, Bollinger Capital Management (www.BollingerBands.com or http://www.BollingerOnBollinger Bands.com). All charts created with eSignal and the eSignal Bollinger Bands Tool Kit.