And Theres the Fringe

Beyond "star wars," there is a fringe of technical analysts who primarily rely on the tools of a good fortune-teller. Astrology has been gaining a following in recent years, with its adherents claiming that, as other methods fail, it is rapidly moving into the mainstream. Henry Wein-garten, a well-known financial astrologer, organized a conference on astrological investing some years back at New Yorks Vista Hotel. Unfortunately, he had to relocate the conference; the stars had not wamed him that the Trade Center bombing would close the Vista. Astrologers, undaunted either by setbacks or skepticism, claim their "science" predicts price movements much more regularly than either •fundamental research or charts. Some of the leading astrological re-«archers confidently predict the dates and times of these moves. However, they state with perfect candor that they do have problems with market direction-not being sure if it will be up or down.

A budding astrological superstar is Archibald A. H. Crawford, a former technical analyst under Bob Farrell, one of the deans of the field at Merrill Lynch. Crawford, who writes an astrological financial letter, told The New York Times that he cashed in on the 1995 bull market "by deciding to go 200% long on December 20, 1994, because of Satums 45 degree angle to Neptune." However, by March 12, 1995, the planets apparently had walked out of their orbits, because Crawford now stated to the Times that "1995 was going to be a grim, untidy year, with plenty of earthquakes and a market heading South.. . . Prices will top in the first week of April," he stated firmly, "with the Dow ending the year 1,000 points lower at about 3,200."*

He missed it by a hair. The Dow rose 36% for the year, ending at 5,100, a mere 62% above where the declension of Satum indicated. An astrologer should be able to remember that Satum is two-faced.

example, are now used by technical "quants." These tools are cobbled together from the sciences of biology, physics, computer sciences, and statistics to pursue the technicians dream of a system that works consistently. Computer consultants, such as the Santa Fe Institute, a think tank specializing in complex computer systems, hold conferences for , Wall Street quantitative analysts. Computers, claim the researchers, can sift through complex pattems light-years faster than the smartest human chartist. Furthermore they can apply objective statistical tests, free of emotional biases.

The Technicians Moment of Truth

Technical analysis is widespread on Wall Street. The daily comments in market columns of The Wall Street Journal, The New York Times, and the financial sections of most newspapers are liberally sprinkled with its jargon. In fact no major or regional brokerage firm could long survive without a troop, or at least one or two, of these distinguished gurus. So there was considerable astonishment in the sixties when academic research began to demonstrate that no trends, tides, or waves could be

Slight miss or no, Crawford ranked second among 42 market timers monitored by Hulberts Financial Digest in 1994.

If astrology is attempting to become mainstream, what might we find on the outer fringe? Sunspots have their followers. Others read tree stumps. Frederick N. Goldsmith, an unfortunate investment advisor of the 1940s, went to trial when the fuzz found out he got his market sig- nals from a comic strip, put there, he fek, by a famous 19th-century mogul, James R. Keane.

Although technical information has increased exponentially with the advent of computers, and the methods range from charting in the mainstream, to the occult at the fringe, the basic problems remain the same. To quote our friend Joe Granville again: "While the indicators may be crystal clear in definition and theory, they often break down and render false signals." Often, patterns are so complex that even leading chartists disagree on their meaning. Too, the "noise" around the market movements is usually so extreme that the indicators can flash red, yellow, and green-all at the same time.

"There is nothing wrong with the charts, only the chartists," runs another old Wall Street maxim. Maybe, but charting requires a very high degree of inte retive capability.

I leamed this myself as a teenager when I visited my fathers small commodities trading firm. His chief trader, although excellent at executing orders, was a passionate chartist. He was convinced that, if given the capital, he would make zillions. He kept the office covered with m every conceivable type of chart. My father, not wanting to lose thif mans skills, gave him a small amount to trade by the charts at the beginning of each year The money was always gone by the end of January. This meant more intensive study and many more charts before the next yearly stipend came around. When I visited the office many years later, the cycle yet revolved.

shown to exist. Technical analysis, the academic findings stated, was useless.

The first findings that price movements were random emerged at the turn of the century. Louis Bachelier, a brilliant French math student, presented such evidence in his doctoral dissertation, written under the supervision of the internationally famous mathematician Jules Henri Poincare. Bacheliers work, which dealt with commodity prices and govemment bonds, concluded that past price movements could not predict future changes. His work rested peacefully for sixty years, until financial researchers rediscovered it in 1960.

About the same time, other academicians embarked on the same course. One early study showed that randomly chosen numbers, when plotted, closely resembled actual price pattems of stocks. Another found that price fluctuadons resembled Brownian motion, the random movements of microscopic particles suspended in fluids.

A good many researchers tested the proposition that stocks move in discemible trends. Amold Moore conducted important work in the field in 1964,° Clive Granger and Oskar Morgenstem in 1963," and Eugene Fama in 1965. Granger and Morgenstem used the advanced new statistical technique of spectral analysis. Their database included 700 weeks of price information for various industries in the 1939to 1961 period, as well as the Standard & Poors and Dow-Jones indices between 1915 and 1961. Eugene Fama analyzed the price movements of the 30 stocks in the Dow-Jones Industrial Average for intervals of 1 to 14 days ( 1 5 years.

All the studies demonstrated that future price movements cannot be predicted from past changes. Without exception, the finding indicated randomness in price-day to day, week to week, even 1 to month. The efficient market hypothesis originated from this workO

No matter how convinced the technician is about the markets or a stocks next move, he has no more chance of being right than by tossing a coin. When a coin is tossed, even if it comes up heads ten times in a row, there is still a 50-50 chance that it wil] come up tails on the next throw. Or, in market terms, if a stock closebigher for ten consecutive days, the next day the stock has a 50-50 chance of closing down. That markets display identifiable trends, the central thesis of the technical school, had been overwhelmingly dismissed by academic research.