More Options

Randomness

Inklings

Lewis Jones

Volume 15.1, March 2005

In July of 1990, The Wall Street Journal began its famous dartboard contest. Every month, four investment professionals picked stocks that they believed to be winners. And members of the Journal’s staff made selections by hanging the newspaper’s list of stocks on the wall and throwing four darts at it. The results were compared at six-month intervals.

After fourteen years, the players have now put away their darts. In the end, the pros came out ahead, with an average investment gain of 10.2 per cent, while the darts players only managed 3.5 per cent. The really surprising thing is that the stock pickers actually gained an edge over random odds (although there is a suspicion that the pros’ selections may have influenced other people to buy the same stocks, and thus pushed up the price). Not that the pros had things all their own way by any means: there was one wonderful six-month spell in which the darts won six times in a row.

Randomness is of course well known as the psychic’s friend. Persuade people that you have access to the secret pattern in some random sequence, then sit back and wait for the cash to roll in. This has been called the Jeane Dixon Effect, in which the odd hit receives maximum publicity and the thousands of misses are ignored. The recipe is simple: forecast often and don’t keep records. This is a particular favorite of those psychics who claim to help the police by locating the bodies of murder victims.

Random results are also the ally of scammers who offer worthless health remedies. It’s easy to be tempted into thinking that highly surprising results can’t possibly be the result of chance. That’s why random events have been called effects in search of causes. Take Evelyn Adams, the woman who won the New Jersey lottery in 1985 and again in 1986. Not quite the one-in-a-trillion chance claimed by the New York Times. Statisticians Persi Diaconis and Frederick Mosteller put the odds that this would happen to someone, somewhere in America at 1 in 30. This is roughly the same chance as my guessing the playing card you’re thinking of after you've told me it’s a red card.

Nassim Nicholas Taleb held senior trading positions in New York and London before founding his own trading firm, and now specializes in the risks of rare events. He writes: “Outside of textbooks and casinos, probability almost never presents itself as a mathematical problem . . . (in the real world, one has to guess the problem more than the solution).”

He confesses that he once joined the hunt for the secret of entrepreneurs who became millionaires. He found that success simply depended on taking risks (he called these people crisis hunters). Then he realized that if he had done the same study on bankrupts, he would have come up with the same answer. “The first counter-intuitive point is that a population entirely composed of bad managers will produce a small amount of great track records.”

Thus, “if a twenty-five-year-old played Russian roulette, say, once a year, there would be a very slim possibility of him surviving until his fiftieth birthday-but, if there are enough players, say thousands of twenty-five-year-old players, we can expect to see a handful of (extremely rich) survivors (and a very large cemetery).”

After all, if talent were only a matter of good results, firms would go out of their way to hire people who had won the lottery. After listening to the O.J. Simpson trial, Taleb confessed to being truly scared of being arrested for no discernible reason. And “having to fight some glib lawyer in front of a randomness illiterate jury.”

Unfortunately, you can always find some sort of pattern in a random series if you look hard enough. When Carl Sagan studied the cancer cures resulting from a visit to Lourdes, he found that the cure rate was, if anything, lower than the one for spontaneous remission. It was lower than the average for those who didn’t go to Lourdes at all. (So does your chance of survival diminish after you visit the place?)

Sometimes a miss is so egregious that you just can’t hide it: the stock market crash of 1987, the Gulf War, the fall of communist East Berlin. The forecasters missed every one.

The American Meteorological Society has admitted that the limit for weather forecasting is between ten and fourteen days. But as business forecasting consultant William A. Sherden has pointed out, “Any farmer who bets his ranch on weather forecasts going out more than one or two days could just as well use a roulette wheel.” And those who imagine that tinkering with a gas such as carbon dioxide can significantly affect the planet’s climate can have little understanding of the complexity of the random fluctuations of the huge atmospheric engine.

As for money matters, you may recall the old joke: “Why did God create economists? To make weather forecasters look good.”

Alvin Toffler made himself an impressive reputation with such books as Future Shock, but practically all of his predictions have been dead wrong. Herman Kahn, Director of the Hudson Institute, made himself famous for an entire slew of planetary predictions: between 75 and 85 percent have been shown wrong. Professor of Demography Paul Ehrlich (with such books as The Population Bomb) confidently predicted war, pestilence, and famine for the planet, and (as Boston Globe columnist Jeff Jacoby ruefully showed) has been “richly rewarded for his almost perfect record of getting things wrong.”

Why rake over this old ground? Let H.L. Mencken provide an answer: “It seems to me that one of the prime jobs of the educated man on this earth is to denounce charlatans. New ones are always popping up, and the common run of idiots are always succumbing to them.”

Lewis Jones

Lewis Jones is a science writer in the U.K.