A while back zigfred at The Polymath Investors wrote a 2-part piece (Part 1, Part 2) sharing his views on why technical analysis is of no use in stock market trading – at least by itself. His reasons are three:
- Its nature
- Its tools are flawed
- A lot of credible long term studies reveal that it does not work
OK, I have to address the last one first as being a non-argument.
Basically, he’s saying that technical analysis doesn’t work because studies have shown it doesn’t work. That’s not a causal statement at all. It’s like saying, “I can’t run a 100m dash in under 10 seconds because I’ve never been able to run it in under 10 seconds.” It’s providing evidence of the fact, not a reason why it’s true. As such, you can basically toss that out, but I won’t quite do that because the evidence needs to be addressed, which I do later.
Tackling things in their proper order, though, let’s start with #1.
The flawed nature of technical analysis
As zigfred rightly points out, the basis of technical analysis is market psychology. Taking that as given, he then presents the argument that on this basis, using technical analysis to trade the markets is a kind of recursive effort in that it turns back on itself because the act of using market psychology to trade influences that market psychology.
While it is certainly true that a definite issue with technical analysis is that it can create a kind of self-fulfilling market dynamic, zigfred presents things as if everyone is trading on technical analysis. Obviously, that’s not the case.
He also seems to be implying that trading on technical analysis is the cause for markets being more volatile than fundamental valuations would seem to suggest. In a truly efficient market – which zigfred seems to think one driven only on fundamentals would be – price would change relatively infrequently, only when new information arrives. Reality is far, far different. Even in the absence of technical analysis there are market mispricings. It’s the under/over-reactive nature of markets driven by individuals who are not perfect in their analysis of information and forecasting of future events. This isn’t even mentioning well-known psychological biases and other factors.
One need only look as far as the housing bubble for a major example. You can’t tell me that technical analysis was the main driver of that!
From a more market-specific perspective, what about the way prices react to data and news with sometimes extreme volatility? You cannot attribute that to technical analysis.
So while I agree that a market overly populated by technical analysis will tend to see TA losing its effectiveness, where fundamentals are still a major factor it remains a useful way to view prices.
Flawed technical analysis tools
The second argument against technicals zigfred makes is that the methods of analysis are basically no better than throwing a dart at a board. His major point is that even technicians don’t agree on which techniques are best or how to interpret charts and indicators.
Hard to disagree. There are a great many indicators out there that are derived from the fields of math and statistics and such which are either poorly understood or incorrectly interpreted. The same can be said of chart patterns and what the underlying causality of their development means. To my mind, this is largely a function of people failing to do the work and the study to really know what it is they are using to analyze the markets.
As flawed as the technical tools may be, let’s not suggest there aren’t major issues with the way fundamental analysis is applied.
Studies show that it doesn’t work
In zigfred’s post he specifically mentions a couple of studies which suggest that technical analysis methods don’t work. I’m not really surprised because I personally believe that rote application of the techniques aren’t really effective in the long run. The markets are too dynamic and changing for things to hold their usefulness consistently.
That said, however, academic research has consistently pointed out a momentum effect in the markets. I don’t have a reference at hand, but it came up a lot in the readings I did while developing my PhD thesis. Momentum in the academic usage of the term is basically trending. If there are trends that can be measured and anticipated, then at least one element of technical analysis has firm grounding in the research.
Now, this post is not me saying that technical analysis is the best thing. As you’ve seen, I’m quite willing to admit it’s problems. I just want to make sure the discussion is done on balanced terms. In my own stock trading I combine it with fundamental analysis. In other markets, and especially in shorter time frames, though, I rely on technicals more heavily.