Thursday, September 12, 1996
Divining Stock Prices
or
It's Technically Just a Stick
by MF Runkle

This past week we were looking for a waterline at work. Our drawings are not that accurate, at least not to the nearest foot, which you need when excavation starts. One of my employees took out two bent pieces of wire, walked around a little bit, and pointed to the ground. "It's here," he said confidently. He'd used the wires as a divining rod. I was taught how to do the same thing by my cousin, Clarence the Farmer. Supposedly, two pieces of bent wire or a forked stick have enough intelligence to find water, buried utilities, and your car keys. I can't even find my butt with a divining rod, so I am skeptical.

True, hydrologists agree you can generally find ground water where a divining rod says it is. That's because ground water doesn't travel in underground streams, except where there are caverns beneath you (which happens in limestone). Ground water is everywhere, even under the deserts for the most part. You just have to dig REAL deep. There are some people who are supposed to be really good with a divining rod, so I won't go out on a limb and say its not a valid system. Oh yes I will! A forked stick to find water? how silly can you get? This whole thing sort of reminds me of Technical Analysis.

Before you TA fans fire off a flaming e-mail at me, I do think there is something to this. Technical Analysis is basically stochastics. With stochastics, you attempt to find a trend in data. Linear regression and least squares analysis can be used to fit data to an equation (which really doesn't work with stock prices), and trends can be predicted by graphing data (which does work). The moving average used by technical analysts levels out random movements of price, and helps indicate a trend. The problem with technical analysis of stock, as I see it, is you don't really know what uncertainty you have. If you are a short-term trader, and have a feel for this, your odds are certainly greater than average, and you can make money. That's because you will be right more often than wrong.

Where I get skeptical is when a "Technical Analyst" sees all sorts of strange formations that look like squiggly lines to me. Then they go on to predict what the stock will do with absolute certainty. How many times have we seen in the stock boards, "this stock will go down to 8, where it will see support for a while, then we should see a strong run." I have had professors that would have flunked me for making a statement like that on charted data. Oh, I like the names of all the different formations, such as "upward parabolic formation," "pennants," "cup with handle," and on and on. I confess that I can't make head nor tails of these things. (Is that another formation?)

It looks to me like people try to divine more out of a chart than is really there. I guess my skepticism is fueled by the guy who wrote the book I read on technical analysis. His newsletter has a slightly better return than stuffing money in a pillow case (about 0.5% annualized). Maybe I'm wrong, and stock prices can be predicted with precision with this method. The problem may be with me; I may not be seeing things that are in these charts. If I did, I would probably get it all wrong anyway. Just like Ray with the waterline. He didn't find it, the contractor did. The back hoe cut through that 2" line today like a knife through butter. What a mess! Next time we'll check the drawings, and leave divining rods and technical analysis to those better than we.

Transmitted: 9/12/96