One Time Bets

I was offered a bet recently.

It was the kind of thing that would be a real money maker if I could duplicate its essential features again and again. I cannot. It was only a one-time opportunity.

The Bet

Using one methodology, symmetry, I can reasonably predict that P/E10 will fall from today’s 27 to 15 in 2011. Placing (coarse) confidence limits around this estimate, I expect to see P/E10=15 between 2009 and 2013.

My lower confidence limit is that P/E=10 before the end of 2009.

I was offered $100 against my $1 that P/E10=15 would not occur before the end of 2009.

This is an extremely attractive bet, considering the payoff. The likelihood that P/E10=15 before the end of 2009 is in the neighborhood of 5% to 10%. I can identify reasons that the probability might be even higher. If you place 100 such wagers, you will receive, on average, $500 to $1000 for an original investment of $100 (or less, depending upon the exact wording of the bet).

If I could make lots of bets like this, I would open a casino.

Nassim Taleb, who wrote Fooled By Randomness, earns a fortune by making lots of bets like this in the derivatives market. He makes a large number of bets that unlikely events will happen. Almost all of his bets lose. But those that win have huge payoffs that cover his losses many times over. It is not simply a matter of the odds. It is the combination of the odds and the payoff.

Most Likely Outcome versus Expected Return

The most likely outcome is that I would lose my bet. It would take, on average, ten to twenty such bets for me to win a single time.

The expected return is very high, $5 to $10 per $1 invested (or better, depending upon the exact wording of the bet).

A wise investor follows Nassim Taleb’s example. He considers the combination of the odds and the payoff. He also makes sure that he has enough bets for statistics to take effect.

Have fun.

John Walter Russell
November 25, 2006