Archive for March, 2008

Revisiting the Prime Inconsistancy

March 23, 2008

If you have not read the previous post involving the prime inconsistency, things will probably make more sense if you do. So take a break and read it – otherwise I would need to summarize it which I really don’t feel like just yet.

After writing that particular piece, I ask myself how would we know if such a thing were going on again? No doubt the business of “guessing the future” has moved on quite a bit in twenty years. The systems used to play the markets are no doubt much more sophisticated. The soft bot on someones desk top computer in 1991 is probably now an integrated server. In my own experience I saw these things used to manipulate and correct engineering systems and their databases in the early 1990’s. A project I worked on then involved watching a database analyst do his work day by day and building a mainframe computer program to replace him as he was retiring. When “client server” computing arrived in the data center, there were no longer any obstructions to building these systems and algorithms. It might be more accurate to call them “meta” algorithms. They would most likely be implemented in a “rules” engine, server, whatever, which would use an underlying set of predictive algorithms. With time and experience, one could perhaps construct a “stack” of these systems, which would generate perhaps extremely “useful” results.

Presuming this has indeed happened, I have looked at a few market graphics. It is my opinion this sort of thing probably goes on from time to time. Since these systems depend on identifiable trends, patterns, etc., the trends should be visible in the market graphics. Look at the typical daily trading trends. Start low peak before lunch, down slightly at lunch, peak in the afternoon, drop to close. If you are using one of these systems, all you need is an established trend. Either day traders are using these systems, or their trades are tracking such systems. It may very well be that most day traders are not sophisticated enough to be doing this, but if they are not they are essentially using the same set of algorithms perhaps working them by hand. I somehow suspect that most day traders have access to a lot of computer power these days.

If there were not some form of manipulation in progress, there would be trends in particular stocks, but the aggregate would not make radical departure from flat – and boring. I don’t see flat or boring lines anywhere. I see peaks and valleys. I see sets of peaks and valleys. I see waves marching along together. That tells the data analyst locked up inside of me the same sort of situation exists today as did in 1987. Why else would you have waves marching along together? Why would you want such a scenario?

If prices are moving up, you buy in today, and cash out tomorrow, make a few bucks, no worry. Then you repeat the process until it doesn’t work anymore. If the market is trending down, you sell out before it reaches the price where you bought in, take your profit to the bank, go short, or get into futures. If your shares are already below your buy-in price, collect your dividends, or cut your losses, go short, play the futures, or find a real job – good luck with that. This changes however if the market is “unsettled” or is ramping up or down slowly. Traders make their money from fees and by taking advantage of the rise and fall of prices. If things are stable there is no money to be made. If the markets are unsettled, you have to know where things are going, to make money. ¬†Perhaps why insider trading is so popular. Unless there are established trends, and¬†volatility, winners will tend to match losers.

A trader who wants to make serious cash in a floundering or otherwise flat market requires volatility. A manipulator’s first problem is establishing volatility. His next problem is recognizing any patterns his volatility produces. To find the manipulator, peel the volatility onion. First find the recognizable pattern. Some recent market trends appear to have such a pattern. It is characterized by a set of double spikes. If you look at a market graphic of several months, it is recognizable. If you want to develop a short list of manipulators, look at who is buying before the first spike, and selling on the second. In this particular case the original manipulator may be lost in the noise, since after a while everyone else catches on and starts doing the same thing.

The only reason a surfer would create a wave is so he can ride it to shore.

Advertisements