[
The Truth About Randomness
Discover The Amazing Power Of Point & Figure Charting
Why The Random Walk is Mostly Wrong, Most Of The Time
To download a pdf copy of this article click on the following link
Why The Random Walk is Mostly Wrong, Most Of The Time
A Case Study In Randomness
An actual histogram of the daily percent changes in the BBY's stock price over a one year period is shown below. This chart is from a few years ago. It should noted that the number of up days during this period of time was 54% and the number of down days was 46%. The balance between the up days and the down days was very close to flipping a coin (50/50) so a lot of investors would say that this was very close to being random. However, the average daily percent change in price for the full year was +0.33% per day. In other words, this random stock gained a little over 80% during this 253 day period of time. The up days were up about 2.33% on average and the down days were down about -2.0%. The difference between the up days and the down days was evidence of a persistent trend in the price of that stock. This trend of stock price changes accumulated to a very significant profit for investors in this stock even though the daily changes had a random component.
Randomness or meaningless variation (noise) in the stock price is suggested by the bell shaped curve of the histogram of the daily percent price changes. If there was no trend, the average daily change would be zero and the distribution would be perfectly balanced and symmetrical on both sides of the mean. This histogram did not show a zero mean and it was unbalanced with more days above the mean than below, and that indicates a degree of skewness. The non-zero mean is primary evidence of a persistent long-term trend in the stock price data.
The non-zero mean is irrefutable evidence that a trend existed over the time period under study. The randomness of daily price changes does not in any way preclude a trend from being present in the data. Maybe we cannot predict the future movements of the trend, but the existence of a trend suggests trend following methods to track the trend and generate a signal when the trend changes direction in a meaningful way. We want to remove the random variation from the price data so that the trend will show up more clearly. We don't want to make investment decisions based on the noise.
The three box method of point and figure charting provides just such a method for removing the small and random variation from the data. The point and figure charting method requires a minimum movement of three boxes to show a reversal of trend. On a price chart, a box is one point on the price. This acts as a substantial filter on the price data and it only allows significant price changes to show up on the chart. This filters out the small random noise from the chart. The long-term P&F charts only show the intermediate movements of price back and forth on the chart. Engineers have been removing the noise from data with mathematical filters for over a hundred years and the point and figure application does the same for investors.
The changes in stock prices, day-to-day, are not the result of some physical or electrical system but are due to the actions of human beings in buying and selling stocks. These changes may be due to many factors, some fundamental but also emotional and psychological factors as well. The point and figure analyst accepts the fact that he may not be able to determine precisely which factors are in control at any point in time - he does not predict the trend as much as he just follows it. He follows the trend so he will know when it changes direction. Knowing when the trend changes direction is the next best thing to an accurate prediction of the trend.
This type of charting is perfectly suited to the needs of long-term investors. The lore of Wall Street indicates that P&F charting evolved as a means for outside investors to keep track of a stock that was being manipulated by an inside pool. It was very important for the outside investor to know when the manipulation was finished and the stock had started down. When the trend stopped, he knew the inside manipulation was over.
The P&F chart can also be used to track relative strength in a filtered format and this removes the influence of the market. Many investors use the relative strength P&F charts to manage the performance of their portfolios. A primary objective for many investors, professional as well as individual, is to own stocks that outperform the market.
It seems that the academic community has become overly enthralled with the mathematical proofs of randomness and has "thrown the baby out with the bath water" in terms of using charts to follow trends and manage a portfolio's performance. The results of attempting to predict the stock market by any means have been found wanting but the trend following methods of P&F analysis show continuing positive results and have for many years.
The truth is that the randomness of stock price movements is a very insignificant part of the business of investment management and the conclusion that charts are of no value whatsoever is totally wrong. How else can an investor effectively follow the trend of a stock price so he will know when that trend (manipulation?) stops? A simple trend following discipline could have saved billions for the investors who owned ENE, WCOM, ABIZ, Q, CNC, TYC and many other stocks that may not have been outright frauds but were investment disasters, nonetheless.
