How does mass psychology work in the financial markets? Two basic emotions drive the markets: greed and fear. The balance between those two is what keeps the markets on a stable course, because people usually veer about somewhere in the middle. (You can exclude the other great emotion, love, as most people don't have an emotional attachment to their shares. At least, they shouldn't, because that really does get in the way of rational decision making.)

Looking at the way markets invariably over-react to almost any stimulus best proves this greed/fear dichotomy. Once somebody's spotted a bargain and started running in one direction, you can bet your life the greedy herd isn't far behind. Closely followed, of course, by all the fearful ones who follow the greedy people simply because they don't dare to be left out. That situation's all fine and dandy until the day somebody realises that the herd has now run too far in one direction, and prices are getting much too high to be sensible. So a 'correction' occurs, in which the herd retreats and prices swing back towards a sort of equilibrium.

At this point, some of the latecomers probably start to realise that they've lost money on the correction, and they may very well attempt to persuade other investors to launch a renewed push for the 'summit' price so that they can recover their positions. A price recovery may then happen. And the rally may even succeed, if enough genuine reasons exist for the market to think it can squeeze a bit more growth out of this investment. Sometimes you can transform even the most exhausted marathon runners by shouting encouragement at them.

Nevertheless, as often as not, the rally quickly falters and fades. Maybe it goes into a sulk while everybody waits to see what happens next. But more probably, it pitches steeply back downwards. People start saying things like: 'Well, I've had a good run out of this share, and yes, it was very nice when that last-minute rally came along to save it from a decline. But if even the late rally didn't hold the share up for long, that tells me this share really hasn't got enough support from other investors. I'm out of here before it falls any further.'

At that point fear comes surging back into the equation and everybody else joins the rush for the exit door.

Does that story have a ring of truth? And surely a behaviour pattern like that does create certain unmistakable patterns on a share price chart? If so, perhaps people shouldn't be quite so dismissive of technical analysts' belief that they can identify these trends from looking at a chart.

Observing behaviour patterns isn't necessarily proof that you can use charts in a predictive way. Knowing what's happened in the past (and why) doesn't give you a cast-iron guarantee that you can figure out what may happen next. But if a chart can teach you something about the mass psychology of the mob, that's useful information.

Lots of extraneous reasons also account for why a rally, for example, fails to stick. Maybe the whole economy's gone into shock and everybody's selling shares, full stop. Or maybe something really exciting's going on somewhere else, and people are pulling their money out of one stock to buy another. A chart can't tell you that. So, on the clear understanding that a chart's a bit of a blunt instrument, the next section describes how to recognise a range of patterns and, more importantly, how to interpret them.

A double top is what you get when a share reaches a peak and falls back by 10–20 per cent, and then does exactly the same again some time later: hitting exactly the same top price the second time round and falling away at the same rate. In effect, the chart describes an inverted W, as shown in Figure 10-1.

Figure 10.1. A double top pattern.

What a double top says to an investor is that this stock probably doesn't have very much growth potential left and the time's come to sell. The market's made two attempts to beat a particular price level, but somehow just not enough people are willing to give it their backing. The share has hit a ceiling – that is, a 'resistance level', described in 'Rising floors, support levels, and resistance levels', later in this chapter.

The decline that follows this second peak may be very sharp indeed, but a few small rallies probably occur along the way, during which cannier investors may manage to offload some of their holdings.

I hardly need to tell you that almost nobody's thinking about the company itself – only about the curve! That's how technical analysis works, at least in its purest form. You may say it isn't very analytical at all. And it drives 'fundamentals' investors to distraction. I've seen quite a few serious altercations on Internet discussion forums between TA fans and 'fundies', and as a rule the TA fans end up retreating into their own little corners where they all speak the same language. I think this is rather a shame, because the best results by far are obtained by comparing and then synthesising the strengths of the two approaches. Your views might vary, of course.

A head and shoulders pattern is what you get when a share rises, dips, rises to a peak, then falls back and has another go at reaching the peak – but fails. What you're left with is a pattern that looks, well, like a head and shoulders. And this is the surest sign you're likely to get that the share's reached its sell-by date. See Figure 10-2.

Figure 10.2. A head and shoulders pattern.

But what's interesting is that head and shoulders patterns come in an inverse formation too. In this example, a share's fallen, recovered, fallen to an even deeper depth, and then started a bigger recovery that takes it pretty much back to its starting level. But as soon as it dips for a third time, a body of investors think 'hang on, this share isn't as bad as the pessimists have been making out'. A renewed buying push starts, and eventually you get to something called a breakout.

A breakout is what every TA investor dreams of spotting. That's when all the previous assumptions about what the market can and can't bear go out of the window, and some pent-up trend gets unleashed as the magic spell of the resistance level gets broken. (I cover resistance levels in the section 'Rising floors, support levels, and resistance levels' later in this chapter.) You're more than welcome to tell me that the whole idea of having floors and ceilings to a share price is strictly for the fairies in the first place. But wearing my TA hat just for now, a sizeable number of people think that a breakout's a development on the general scale of Jupiter and Mars aligning with each other. A breakout is a big event.

Being a bit less sceptical, just for once, you can reason that a breakout signifies that something has indeed changed in the psychology of the market. People have suddenly abandoned whatever was holding them back. And if you're quick enough, you can profit from the emerging change in the conditions. Figure 10-3 illustrates a breakout.

Figure 10.3. A breakout.

In the same way, a double bottom (no, really), looks like a normal W. In a double bottom, the share price has tried twice to plunge below a certain level, but the market isn't playing ball and the price has rebounded strongly. Once the price has recovered for the second time to the level that prevailed at the start of the W, a TA investor feels pretty sure that another breakout has happened, and the likelihood is that the share is now free to rise much more quickly.

Look at a typical TA chart and you're likely to notice that it's covered with straight diagonal lines that run roughly in parallel with the main trends over a given period. One soars above the curve and one's below it, so that the actual share price line is sandwiched between the two.

Why? TAs believe that at any one time the market includes a fairly fixed level of tolerance that dictates just how high a share's price can reasonably be allowed to go, and another level of tolerance for how low it can go. In other words, if a share goes outside those limits, a fair number of investors say no thanks. The share has hit an invisible wall.

You're probably thinking that this is close to mumbo-jumbo. Why on earth should the 'collective psychology' of the market decide, all at once, that an absolute ceiling applies to a share's price beyond which it can't go? And an absolute floor (the 'support level') below which nobody's tempted to sell it?

These are good questions, and I honestly can't answer them. But sheer experience dictates that a great many shares seem to run into these frustrating barriers, so maybe they do exist after all.

I can suggest rational reasons for some of these phenomena, but not all. If a share has been floated on the stock market at 200p but then drops alarmingly back to 100p, a lot of investors may say, 'Yes, I made a mistake, but I'm jolly well not going to sell that share until it reaches the price I bought it at – and then I'm out of here.' So the price hits an invisible wall at 200p.

Other investors may leave instructions with their broker to sell if ever the dividend yield goes lower than the bank rate, or if the price/earnings ratio (see Chapter 2 for more on these) gets to be more than 50 per cent above the sector average. You can't ever really know, which is why the mumbo-jumbo accusations don't stop coming.

What you can say, though, is that if (big if) these ceilings and support levels are genuine, their very consistency suggests that the market's a bit more logical than some people suppose. Not every day does a fundie accuse a TA devotee of being excessively logical, but these things do happen.

The apotheosis of the support and resistance line theory (see the preceding section) comes when an investor reckons that these two lines are converging. As a general rule they tend to hover around in parallel to each other like the outer layers of a sandwich. But when you get a chart like Figure 10-4, a TA specialist says, 'Whoa, this market is getting surer about exactly what it wants.'

And indeed it is. The chart shows a share that's got hardly any more room for manoeuvre. The market's ideas about its top and bottom price are converging so progressively that all the doubt is being squeezed out as the weeks go by. Perhaps the company's in the final stages of a takeover bid, and everyone's pretty sure what the final price may be?

Figure 10.4. A triangle pattern.

More types of triangles exist than I have the patience or the room to describe in this chapter. You can get soft bottoms, triple bottoms, flags, pennants, wedges, and cups and handles. If you've got the time and the inclination, you can find plenty of information about these TA trends (try the Stockcharts website, www.stockcharts.com). Whether you choose to believe in them or not is entirely up to you. But at least now you understand what people are yakking on about in the discussion forums. That's good enough for me, because by bringing them to your attention I've primed you for the day you meet these peculiar terms in the flesh.

Forget about watching tiny share price movements over short spaces of time, and turn your gaze to the heavens instead for the bigger picture. The giant of TA theories is the Elliott Wave.

The Elliott Wave is half-philosophy, half-superstition. It dictates that nearly all really important stock market developments come in predictable waves. And that something mystically called Fibonacci numbers determine these waves.

Bull markets (long-term trends of stock market growth) are characterised by five major waves, according to Elliott Wave theorists: a first wave, followed by a small relapse, then a third wave and another fallback, and then the fifth really big wave where all the action really happens. You may hear TA enthusiasts arguing for months on end as to whether the fifth wave has really started, or whether the third wave has merely had a fit of indigestion and thrown a minor wobble that makes you think the first wave has started. Chapter 4 covers bull markets.

But bear markets (longish periods of stock market decline) give you less warning. They have only three waves, according to Elliott Wave followers: a first falling wave, then a small recovery, then the big drop! See Chapter 4 on bear markets.

I'm oversimplifying this theory quite horribly, and I apologise to all genuine Elliott Wave people for my unforgivable heresy. If you want the real lowdown, the information section on the Stockcharts website (www.stockcharts.com) gives you plenty to think about.