How to pick a winner: Follow sport, sip Bordeaux and watch your hems

Can't wait until Wednesday to find out who'll be the next president of the United States? Can't make head or tail of the opinion polls? Well here's some good news for you: there are plenty of ways to predict the outcome without listening to the pundits or crunching the contradictory poll numbers.

Can't wait until Wednesday to find out who'll be the next president of the United States? Can't make head or tail of the opinion polls? Well here's some good news for you: there are plenty of ways to predict the outcome without listening to the pundits or crunching the contradictory poll numbers.

You could look at the performance of the Boston Red Sox baseball team, for a start. Or wonder whether this has been a good year for Bordeaux. You could abandon psephology as a load of fuzzy science and put your faith in the self-proclaimed sages who claim never to have called a race wrong (most of whom seem to be propping up the bar at Harry's in Paris).

Or, if you really insist on doing it scientifically, you could buy one of the many complicated mathematical formulae on offer, in which everything is factored in from the president's popularity rating last July, to the percentage increase in Gross National Product from the fourth quarter of last year until the second quarter of this.

Curiously, almost all of these superstitious indicators point to the same result: a victory for George W Bush. Probably. Give or take a couple of extenuating factors.

Take the Red Sox. According to tradition, a victory in the team's final home game before the election favours the incumbent party, which is to say Al Gore's Democrats. This year, the Sox lost.

Meanwhile, over at Harry's Bar near the Place de l'Opera in Paris, the punters have been giving their opinion on who will win in a straw poll whose results are on prominent display next to the cocktail shakers. Again, Mr Bush is the favourite, by a 55-45 per cent margin. Harry's has got it wrong only once since 1924; then again, the ex-pat businessmen who frequent the place are overwhelmingly Republican, and Republicans have dominated the White House since the end of the Second World War. Can they really be trusted? (The one mistake they made, in 1976, was a close race in which the Democrat, Jimmy Carter, sneaked home.)

The fool-proof profs with their mathematical models offer even less certainty. Michael Lewis-Beck of the University of Iowa says he predicted Bill Clinton's margin of victory in 1996 to within one-tenth of a percentage point; this time, he's bucking the trend of his fellow prognosticators and giving Al Gore 55.4 per cent of the main-party vote.

Could he possibly be right? An outfit called the American Enterprise Institute thinks it is wise to look at a wide array of factors: everything from hemlines (rising or falling?) to the quality of this year's Bordeaux to the latest movements in the stock market. This year, the wine has been good (favours Democrats) but the hemlines have fallen (good for Republicans). Their survey also gives Mr Bush the edge; as a teetotaller, we know however that if he wins he won't be uncorking any claret, whatever the vintage.