Fool on the Hill
An Author's Response

Format for Printing

Format for printing

Request Reprints


By TMFOtter
January 11, 2001

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!

Is Technical Analysis Voodoo?

[Bill Mann wrote on the subject in a Fool on the Hill column, and invited responses from our community. You may access the story from the link above. What follows is his synopsis of that conversation.]

I'd like to thank all of you for coming and speaking about your experiences with technical analysis. I have gotten some excellent information as to where to go to look for deeper insight into a confounding subject.

And to those who chose not to be so polite - relax. I was quite clear about my ignorance about the mechanics of charting. I was also clear about my bias. You could do worse than to lighten up and consider why it was I was asking the question. I was seeking to open a dialogue on whether there is relevance in charting as a skill that can be taught reliably to investors as something to use.

In particular I'd like to thank dgander for referencing some fairly rigorous studies, dbphoenix for his patient correspondence with me, AngryCandy for actually reading a piece of Fool content, and Padavona for his cogent critique.

My bias that charting does not help the average investor comes not from some study of the methodology, rather it comes from a deep belief that efficient trading philosophies will win out for the largest number of people. And just as I don't need to understand physics to figure out how and why a watermelon will fall if dropped out of a window, I do not believe that there is too much to be gained by going through the study. Recall this, the point that I was making is that the general investor, after all trading costs and capital gains, must have a higher net return that the averages (let's say the S&P 500).

So, where does my bias come from? There are three basic reference points.

The first is myriad studies in behavioral finance, most notably by Terrence Odean, who has found that, in analyzing 158,000 online trading accounts over two multi year periods, that the 20% of accounts that had the highest number of transactions performed significantly worse than less active accounts. In multiple posts people have chided me for using the word investors, when they prefer traders. Fine. Let's assume for the moment that there is a much higher incidence of "traders" in the accounts with the most activity, a point which I doubt will be that controversial. Of course, there are some yahoos in that top 20% who simply do not know what they are doing, but there are also some long term holders who are also fairly incompetent and have picked some crappy companies. A point which many of you seem really eager to point out. So traders ipso facto are among the most active accounts, and these same accounts, by measure of several tests, are shown to have worse returns than less active ones.

Here is one such study. I've got plenty more.

OK, point two. The Buffett example. Many have given me several names of highly successful traders, which I deeply appreciate. It takes some of the edge off of this certain example, if in fact their returns can be verified. (And I'm assuming they can be). Still, over a long period of time, I am unaware of anyone in technical analysis has achieved the returns of a Warren Buffett, a Bill Miller (the mutual fund manager with the best 10 year returns and the longest streak of years beating the S&P), a Phillip Fisher (bought Motorola in 1958. Never sold it).

This was not put out as a straw man. Of course it does not follow that if someone using TA can't beat Buffett, then the theory is worthless. What I am interested in is bell curve theory.

As many of you know, random things typically have a bell-shaped distribution. If one were to plot out the distribution of returns on a histogram using the returns of all of the returns of people using cash flows and other fundamental analysis techniques, the bell curve would be quite wide, running high in the middle with some outliers on either end. Warren Buffett would be on the far end of the histogram, as would Peter Lynch, Osceola Moore and others. People whose portfolios consistently under performed would be toward the far end, to be manned at the end, presumably, by someone who was consistently given money and lost 100% over and over again.

Now, do the same thing with chartists. There HAS to be chartists who have earned million percent returns on their capital at the top end. Particularly when matched with the above data point, that states that the most active accounts under performed, for charting to be a superior way for a large number of people to invest, the bell curve would have to be even wider for chartists than it is for pure fundamental analysts. In other words, for the random distributions to follow, there would have to be an even wider bell curve for chartists, with higher percentages of even better returns. Just so, there would also have to be additional chartists with returns that approach unbelievably poor levels.

Issue number three is a simple one. Neither of the first two issues even include taxes. With most trades any profit is going to be taxes at the short term rate, your ordinary income rate. This means that, to beat my returns on something I never sell (theoretically) you would have to beat my returns over, say, three trades in two years by 56% to equal my returns. That's not adding in trading costs, that's just with taxes (which I'm assuming are 39.4% for the ordinary rate).

Several people have rightly criticized Buy and Hold. I'm pretty sure that those who buy and hold all crappy companies are doing nothing but failing. Buying and holding is a poor excuse to avoid reviewing one's stock picks. And tax avoidance is a poor reason not to sell a stock which has changing fundamentals.

The combination of these three issues, plus the myriad studies sited by Malkiel, the teachings of Fisher and Graham on valuation, and studies like Odean's and John Train's among many others, make me believe that, for the ordinary investor, technical analysis is a highly inefficient, more risky way to invest. As I stated in the article, I do not doubt that it works extremely well for some over a long period. I also do not doubt that it works well for many others over a shorter period, but I find the evidence of more trading = poorer performance, fewer and less successful upper range participants, and the sheer inefficiency of trading costs to be strong arguments why technical analysis, and with it the tendency to trade more, to be an unattractive way to manage money.

Padavona, among others, rightfully called me out on the fact that the Rule Breaker uses relative strength, a true technical indicator, as one of its criteria for investibility. Truth be told, just depending on RS as a function of investibility strikes me as a bit of a cop-out on behalf of the RB managers. However, I understand why they do it, and it does have roots in both technical and fundamental analysis. It does not, however, presume to determine what the charts are saying that the stock is going to do next, it is rather a sign of increased belief in fundamental strength by the company, manifested in increased enthusiasm about the stock.

I'm looking forward to continuing this discussion. I'm no zealot by nature, and I'm a firm believer in the benefit of not having any sacred cows. If i can find evidence that weighs strongly enough toward the benefit of charting, I'd certainly like to look to consider its use in the future. But given my confidence that the above points are true, the case for improved performance by the majority of chartists would have to overcome a great deal.

My best to all of you, and thanks for chiming in.

Bill Mann