Numbers

ERHAAB, the favourite, won the Derby while Plato's Republic, the record outsider at 2000-1, came last. An estimated 4 million people had a flutter and about pounds 40m passed into bookmakers' offices. Most of it passed out again too. A favourite winning is bad news for the bookies. But was it good for the punters?

According to Graham Sharpe of William Hill, the Derby is a good race to bet on the favourite. With uninformed punters spreading their once-a-year bets evenly through the 24-horse field, the bookies can be generous in the odds offered on the favourite.

A good, if long-odds bet, is Pascal's wager, after the 17th-century French philosopher Blaise Pascal, who applied probability theory to religion. His argument was that belief in God is a good bet since the potential gain is eternal life, and the possible risk is eternal damnation, all for the stake of a little corporeal denial.

Apart from that, good advice on gambling is hard to come by. Psychological studies have concluded that the best policy is to back the favourite in the last race on a Saturday (when increasing desperation moves more money on to the unfavoured runners). Another study suggested that picking your horse with a pin is as good a policy as any, and better than most if you place large sums on short- odds horses and small amounts on outsiders. Betting tax diminishes potential rewards, so the whole thing must be, statistically speaking, a bad idea. Pascal, however, would have considered putting a few francs on Plato's Republic.