you want to make your fortune publishing a market newsletter. You need first to convince potential subscribers that you have talent worth paying for. But what if you have no talent? The solution is simple: start eight newsletters. In year 1, let four of your newsletters predict an up- market and four a down-market. In year 2, let half of the originally optimistic group of newsletters continue to predict an up-market and the other half a down-market. Do the same for the originally pessimistic group. Con- tinue in this manner to obtain the pattern of predictions in the table that follows (U prediction of an up-market, D prediction of a down-market). After three years, no matter what has happened to the market, one of the newsletters would have had a perfect prediction record. This is because after three years there are 23 8 outcomes for the market, and we have covered After the fact, the one newsletter that was always right will attract attention for your uncanny foresight and investors will rush to pay large fees for its advice. Your fortune is made, and you have never even re- searched the market! WARNING: This scheme is illegal! The point, however, is that with hundreds of market newsletters, you can find one that has stumbled onto an apparently remarkable string of successful predictions without any real degree of skill. After the fact, someones prediction history can seem to imply great forecasting skill. This person is the one we will read about in The Wall Street Journal; the oth- ers will be forgotten.         wsletter Predictions all eight possibilities with the eight newsletters. Now, we simply slough off the seven unsuccessful newsletters, and market the eighth newsletter based on its perfect track record. If we want to establish a newsletter with a perfect track record over a four-year period, we need 24 16 newsletters. A five-year period requires 32 newsletters, and so on.       The Lucky Event Issue In virtually any month it seems we read an article about some investor or investment company with a fantastic investment performance over the re- cent past. Surely the superior records of such investors disprove the efficient market hy- pothesis. Yet this conclusion is far from obvious. As an analogy to the investment game, con- sider a contest to flip the most number of heads out of 50 trials using a fair coin. The expected outcome for any person is, of course, 50% heads and 50% tails. If 10,000 people, however, compete in this contest, it would not be surprising if at least one or two contes- tants flipped more than 75% heads. In fact, elementary statistics tells us that the expected number of contestants