Technical Analysis Voodoo: Analysis and Reply

January 9, 2001

In an article discussing the Motley Fool Rule Breaker Portfolio's dismal performance of the year 2000 - down 50% - TMF Otter asks the (for him) more or less rhetorical question, Is Technical Analysis Voodoo?

The Article is full of a lot of mistakes, and deserves close analysis, since a lot of people take what appears on the Motley Fool site as the gospel truth.

First, he begins with a common complaint of somebody who is criticized for not doing well recently. He does not like it when somebody says What have you done for me lately?, and ignores a glorious past. Personally, I think that this is a perfectly reasonable question. Why should anybody think that an investing philosophy that has not worked recently will work in the future? So it worked well when there was a dot com mania, but that is over, at least so far as I can tell. AOL and AMZN, for example, are both down significantly from their highs, and are likely to be evaluated in more conventional ways as the new economy matures. Latching onto a mania is fine, but recognize it as a mania, and jump ship when it subsides.

Then he says

What I find interesting about the witch hunt [he thinks that people are using David Gardner as a scapegoat] is two things: First, David stated at the beginning of 2000 that he was comfortable with the prospect that his portfolio (the Rule Breaker is his money) could lose a significant portion of its value over the year; and secondly, that we are seeing from many correspondents that one of the "lessons" learned from the carnage in 2000 is that investors "ignore the charts at their own peril," or something similar.

Now let us consider what he is saying. First, it is downright goofy to be comfortable with a 50% loss. That would be reason for a good Japanese aristocrat to commit hari kari in the old days. Second, it is very reasonable to say ignore the charts at their own peril, since the charts - the price actions of the stocks - are the ultimate fact of the market. Investing in a stock because it ought to go up, and maintaining a position in spite of the fact that it is going down - those are the two major ways that people lose money in the market.

He then goes on to say that the Rule Breaker Portfolio was known to be a risky enterprise. Fine, but is it not reasonable to limit the risk with some stops, or at least a reality check when stocks tank? Then we have the statement that for venture capital companies such as AMZN, price is less valid than a firm grasp of how the company expects to make money. Now this may certainly be true. However, neither the price nor the price action of the stock is irrelevant. The object of investing, after all is to make money off the price action of the stock, not to find companies with interesting ways of trying to make money.

Then we have a definition of Technical Analysis:

Technical Analysis is a simple science: It states that stocks that are in motion tend to stay in motion, that a stock that rose today is more likely to rise tomorrow. It is a powerful tonic for those who are terrified by the notion that short-term stock market movements are without rationality. Technical Analysis allows investors to say "Ignore the reasons, they are meaningless. Focus on the patterns." The pure technician ignores such basic concepts as stock value, price, or other fundamentals on the belief that the institutional investors leave telltale signs when they are moving into or out of a stock, and that the "smart money" telegraphs its actions by virtue of its sheer size.

This is a powerfully attractive theory, and I do not doubt that there are those who can practice it with some success. But these people are not the "average" technical analysts. They are, in fact, few and far between.

This is hardly all there is to technical analysis. It sounds more like simple momentum investing. I do not think that he as a very good grasp of the field at all.

Later, in fact, he gives the example of a pennant formation, and says that if it were just as simple as finding pennants, then everybody would do it and so the formation would be meaningless.

Now I really want to look at the if it really worked, then everybody would be doing it, hence it cannot work argument. This argument can be made for any way of investing. If it were really true that certain stocks with certain fundamental characteristics always went up, then they would already have gone up, and so it would be pointless to look for those fundamental characteristics. The same can be said about any technical analysis pattern. But the fact is that there are a lot of purely random factors at work. These things, whether fundamental or technical, only work some of the time. All you have is a probability, and one hard to measure at that. Besides, there are always people that are skeptical about anything, so it is unlikely that everybody would ever trade in a particular way.

In addition, I should point out that a pennant formation is price congestion, where the pattern of the chart shows large variation at the beginning and small variations to the end. Then the theory is that if there is a breakout either to the upside or the downside, that breakout is likely to determine the movement of the stock (up or down) for the near future. This does not fit the definition of technical analysis above: stocks that are in motion tend to stay in motion...

After this, TMF Otter proceeds to make this totally absurd statement:

In his seminal book The Money Masters, John Train speaks of a standing wager he has with any technical analyst. It goes something like this: He will take a chart from several years ago and remove any identification from it as to company name or time. He then cuts the chart in half, giving the earlier portion to the chartist while keeping the later portion to himself. Since chart-reading is supposed to be prophetic, this should be no hardship to the technical analyst. In order to win the wager of $100 the analyst must be able to tell whether a stock was higher or lower at any point in the second period than the first. And since technical analysis is predictive and requires that the practitioner pay friction costs, Train asks for modest odds to compensate.

How have the takers of this bet fared? Hard to say, since no one ever has taken the bet.

Of course nobody takes the bet. Technical analysis is a day-to-day thing. Nobody in the field has ever claimed to make long-term predictions. In fact, sometimes you look at a chart and the results are inconclusive. Sometimes you look at a chart and see something that makes you believe that the stock is going to move up or down. A lot of times you are right, and if you are wrong, paying attention to the price action should get you out with minimal loss. No credible technical analyst has ever claimed certainty - only that probability is on your side if you read the charts correctly.

We then have the following statement:

As prudence would demand, I expect that proof will be demanded of me in my assertion that technical analysis is an inferior methodology for investing. People will of course roar about one-year performance, and compare it to the Fool portfolios. I ask you this: Who invests only for one year? The reality is that there is no such thing as an investing strategy that outperforms each and every year. Any investor who thinks otherwise, regardless of the strategy she employs, had best prepare for grave disappointment.

This is just such a wonderful hedge! This is why, when I hear somebody say I am in it for the long haul, I can give you pretty good odds that he just lost a bunch of money. A year is a long time. Fifty percent is a lot of money to lose - now you have to double your money in order just to break even.

In the article, TMF Otter also mentions the failures of Joseph Granville and Elaine Garzarelli, both of whom had some good calls followed by some bad calls. Now I do not know anything about Granville's methods, but I do know something about Garzarelli. Whatever is to be said about her success or failure, she was not a technical analyst, at least not when I was subscribing to her newsletter in 1996. She referred to her methodology at techno-quant, which meant that she crunched a lot of numbers to get her results. But the numbers were all things like current earnings, historical and current P/E ratios, future estimates, estimates of related industries, interest rates, etc. She was a highly technical analyst who believed that she could look at the fundamentals of the major companies - those in the S&P 500 - and determine which were undervalued by historical standards. Then the theory was that if you bought the undervalued companies, they would shortly be fairly valued, and you would make money. So whatever is to be said about Garzarelli, her failure to call the market correctly in 1996 was not a failure of technical analysis. Was it a failure of fundamental analysis then? I would not draw that conclusion either. She had a sophisticated model which did not work so well, that is all.

TMF Otter also makes reference to Warren Buffet as an example of a great investor who did not use technical analysis. OK, so what has he done recently?, I ask. Over the last 3 years, up at a compound annual growth rate of abour 13.5%. OK, but hardly legendary.

Finally, I want to go back to

This [description of technical analysis above] is a powerfully attractive theory, and I do not doubt that there are those who can practice it with some success. But these people are not the "average" technical analysts. They are, in fact, few and far between.

That bit about the average may or may not be true. However, suppose you paid your hard-earned money for a class in investing (or anything else, for that matter), and the instructor said

This course is for the average investor...

Do you want to be average, or do you want to be above average? Why not try to succeed by all available means?

Anyway, just for the record, I think that it is necessary to use both fundamental and technical analysis when trading individual stocks, and I think ignore the charts at your own peril is a very good saying.

[ Top ] [Introduction] [The Dogs] [End of Funds?] [Couch Doggie Portfolio] [Pick Good Funds] [Stock Trading Systems] [ Trading Systems ] [Simple and Profitable] [Journal] [ Timing ] [Selection ] [ Common Sense ] [ Statistical Noise ] [ Mechanical Hibernation ] [Best Ever ] [ LTBH Results ] [ Beat Market ] [ Technical Voodoo? ] [ Charts Do Not Lie ] [ Bad Advice ] [ Analysts ] [ Gamblers, etc. ] [ Relax ] [Links ]