Follow the loser and become a winner
by Vivek Kaul/ DNA Money
Ketan Mehta, an Indian residing in New York, had been receiving stockmarket newsletters in his mail for five consecutive weeks.
For each of the five weeks, the newsletters had correctly predicted the direction in which the Dow Jones Industrial Average (DJIA), a 30-share stock market index, was headed. The newsletter claimed to be using sophisticated software and Wall Street connections to give the right prediction, week after week.
A couple of days later, another newsletter from the same company arrived. It said that Mehta could continue receiving the newsletter, if he paid a certain subscription charge. Mehta thought that the money would be well spent and decided to subscribe. The correct predictions continued for the next two weeks, but after some days it went kaput.
Mehta simply could not figure out as to why this was happening. When the newsletter had got the direction of the market right all along, how did it suddenly go wrong?
Mehta had become a victim of the stockmarket newsletter scam, which has happened in the past in the US. And here’s how it works.
In the first week, a newsletter was sent out to 128,000 individuals, picked up at random from a database, let’s say a telephone directory.
To half of the 128,000 individuals, the newsletter predicting that the DJIA will go up was sent and to the rest the newsletter predicting that the DJIA will go down was sent.
Whatever happens to the index by the end of the week, 64,000 people would have received a correct prediction. The same process would be repeated again, but this time with the individuals who got the right prediction.
This process will be repeated a few times more. By the end of the fifth week, 8,000 people would have got correct predictions for five consecutive weeks. To these people another letter will be dispatched, pointing out the good performance of the newsletter.
And from now on, if individuals wanted to continue getting the newsletter, they would need to pay for it. And many individuals like Mehta will fall for it and subscribe to the newsletter.
This scam can be executed because of the existence of survivorship bias in investors, wherein they see only the winners and hence get a distorted view.
Nassem Nicholas Taleb, in his book, Fooled by Randomness, explains this phenomenon. He says, “If one puts an infinite number of monkeys in front of (strongly built) typewriters, and let them clap away, there is a certainty that one of them would come out with an exact version of the Iliad”.
Having said this, Taleb asks “Now that we have found that hero among monkeys, would any reader invest his life’s savings on a bet that the monkey would write the Odyssey next?”
So the question to be asked in this case is “Does past performance predict future performance?”
We cannot deny the fact that to some extent it does. But this presumption might be very weak. The initial sample size matters a lot. The greater the sample size, the greater the chance that someone might do well just by luck. Let’s take the case of stockmarket analysts, the fact that there are so many of them out there, the greater is the chance of one of them performing well over a longish period, simply by luck.
As John Allen Paulos points out in his book, A Mathematicians Plays the Stock Market, “Are stock analysts in the same profession as the newsletter publisher? Not exactly, but there is scant evidence that they possess any unusual predictive powers”.
(The example is hypothetical)
