Technical Analysis

First a definition - Technical Analysis is a method of evaluating securities by analyzing statistics generated by market activity, such as past prices and volume. Technical analysts do not attempt to measure a security's intrinsic value but rather use charts to identify patterns that can suggest future activity.

But most often than not so called technical analysts will say it works unless it doesn't.

There are traders who say they are successful with Stochastics. Others swear by moving averages. Still others track open interest, seasonal patterns, cycles, or changes in volume. Technical analysis is all based on indicators and charts. My belief, which is likely ridiculed and discounted by technical analysts everywhere, is that technical analysis indicators are useless to the average trader when used by themselves. They are information filters, not trading systems. Allow me to illustrate, for those of you who haven't left yet.

See, my definition of a useful standalone indicator is one that works so reliably that it is a big surprise when it doesn't work, and I can instantly recognize that failure and do something about it. To me, "useful" is not a label I'd apply to an indicator that works 40% to 60% of the time, and has long periods of indeterminence where it neither signals a clear entry or a clear exit.

Look at any book that illustrates the Stochastic oscillator. It will provide a chart and highlight how at overbought levels, the fast line penetrates the slow line above the 75% line, and sure enough, a downward surge in the market ensues. Look ahead of or behind that example on the same chart, and I almost guarantee that you will see examples of the same indicator doing the same thing, but without any kind of similar market motion occurring. This is a good skill to practice; instead of looking at the examples an author provides for confirmation of the signal, go looking for places where the signal didn't work. If you find a fair number of them, then you can assume that you will lose at least as often as a failed signal occurs.

Crossing moving averages? They cross all the time, back and forth, without a cross being very meaningful to market action. Think about the underlying fundamental statement that the indicator is making, rather than just looking at the picture. An 18/40 crossing pattern means this: "When the average closing price for the last 18 days becomes less than the average closing price for the last 40 days, then the market is going to travel downwards for a while." Really? Why? What fundamental market force does this point to? And shouldn't you have gotten in 18 days ago, when the market covered all that bearish ground? What if the downswing only has 20 days worth of motion in it? From where I sit -- and it's not a crowded room, believe me -- you need an indicator like this to be startlingly accurate in order to profit from it on its own merit. Even 60% accuracy is nowhere near good enough. You want 80%, or better than that.

Trend lines in technical analysis? Well, I like trend lines, but I think the only really good application of them is in the support (or resistance) of a trend. If you use them as top and bottom pickers, you are asking for trouble. And the big reason they're somewhat useful as a tool is that your risk is definable. If the market penetrates the line, you're out, no reversal, no messing around. Do I trade trend lines by themselves? No.

Momentum measurements, moving averages, seasonal patterns, cycles, trend lines, median lines; all of these technical analys indicators come under the heading of "it works unless it doesn't". This is one of my favorite futures phrases (well, DUH, you say, if you've read more than one of my articles). If you are enamored of one of these indicators, go grab some charts and look for exceptions. If you don't spend more than twenty seconds before you find your first one, then you've hit something that's very likely to fail too often.

Does that mean you should throw away all technical indicators and never use them? Heck, no. What it means is that you should be extremely wary of using any indicator BY ITSELF, without additional thought, and you should be extremely wary of anyone who is selling something they bill as the ultimate standalone indicator. Some folks do use Stochastics successfully, but they also use its signals under restricted market conditions, and have a clear strategy for what to do when it doesn't work out. A pattern or a moving indicator or some kind of arithmetic relationship is a filter to help you block out market information you don't want. But unless you have a one-in-a-million indicator that I've never seen before, using it as the sole decision maker in your trading world will just lead to losses.

Yes, many experts extol and defend a variety of indicators, but it also seems that a lot of the experts and market gurus have been inhaling from the Kreem Wip can a little too often, except when they occasionally and accidentally make a good prediction. The conventional wisdom and advice in the marketplace is causing 90% of the new traders to lose what they have. Reliance solely on demonstrably useless signal generators is part of that conventional wisdom and practice. If you really want to throw your money away, why not just send me a check instead? I promise to enjoy myself, and I'll even write you while I'm on vacation.