An introduction to technical analysis: is it all hocus pocus?

When I started out as an analyst, I remember being introduced to the Wizard. He was a strange guy. Everyone else in the office did things I could understand. They were economists or analysts, they made spreadsheets, they tracked industries, they looked at numbers from the Bank of England or the government. But this guy played with wiggly lines and different coloured pencils and what looked like huge games of noughts and crosses. His office door said 'technical analysis' and what he did was charts.

Charts get a very bad press sometimes, partly because some investors use them as a short-cut for thinking. It's much easier to look at the wiggly lines and 'read the tealeaves' than it is to do your own analysis.

But there is a real philosophy behind technical analysis, and once you understand it, you see how using charts isn't nonsense - but it also has limitations and weaknesses that you need to understand. Technical analysis is about looking not at the value of the business behind the stock (fundamental analysis) but looking at the way that the market values, and trades, the stock (behavioural economics).

Behavioural economics looks at how a free market is affected by human behaviour, in the aggregate. For instance, we might observe a tendency for people to follow those stocks about which they see the most coverage in the newspapers, and to buy those stocks which are already rising, since they feel they are going with the flow. Behavioural economics explains the failure of efficient market theory (the theory that markets always perfectly reflect all information that is available about a given investment); by applying behavioural principles we can see why individuals and institutions make systemic errors. For instance, why the banks piled into sub-prime lending in 2005-7 and consequently needed bailing out.

Charts are simply one way by which we can assess the behaviour of the market. Rather than looking at the fundamentals of a stock - the company's business, its latest results, price-earning ratio, or balance sheet - we look at the way investors have behaved when trading that stock, as reflected in the movement of the stock price. So for instance we might notice that there is a typical trading range. That might reflect two possible views of the correct valuation, so the stock might swing between 'overvalued' and 'undervalued' points fairly regularly.

One particular factor that behavioural economics points out, which fundamental analysis won't pick up, is that markets typically overreact to information before adjusting to it. (I think the housing market is a very interesting case in point here - lots of people seem to think that it will start to recover once it gets back to a 'normal' house price to income ratio. But if you look at what actually happens, and doing so through a chart is an easy way to do that, you'll see that it usually swings way below the average for a couple of years before finally starting to recover.)

Now technical analysis doesn't just say that the way investors behave will determine the way a market values a particular investment. It also says that this behaviour tends to occur in particular patterns - like overreaction-and-adjustment, boom-and-bust. When a technical analyst looks at a chart, he's trying to find the patterns that repeat themselves over and over again. It's the same concept that lies behind other concepts in economic disciplines such as marketing (the product lifecycle) or statistics (regression to the mean).

Most technical analysis focuses on price alone. However I believe strongly that we can look at any information about a stock in a technical way - and we should look at derived ratios (like the house price / income ratio, or the PE ratio), as well as at trading volumes, not just at prices.

Once you understand the philosophy behind charting, you can easily see its strengths and weaknesses. Its strength is that it shows you what is actually happening in the market, and that it gives you a clear historical record to compare with the present. It tells you what other investors are doing, so you can see the trends clearly.

Its weakness, of course, is that it is value-free. You may look at a price peak on a chart and say to yourself that it looks horribly exposed - but to find out whether it is or not, you'd need to know something about the fundamentals. An exposed price peak might be a tech company in early 2001, or an oil company that's just found a huge new field - in the first case the chart would be a dreadful warning, in the second, it wouldn't. (As long as the oil was actually there, of course!)

So I'd tend not to use technical analysis on its own as a means of analysis. But I do think it's a useful discipline to add to your armoury of techniques when you're looking at the market - and it does actually have a real methodological underpinning, it's not just 'reading the tealeaves'.