New Statesman: Racing card - United Kingdom election forecasts

The Bet: How many seats will the parties win at the next general election?

Labour 370-380

Tories 212-222

Liberal Democrats 37-41

(Source: IG Index)

If you don't fancy Labour at 1-5 on a straight bet, why not have a punt on the spread? There is much more money to be won if you get it right -- but don't forget, you can also lose a lot if you get it wrong.

For the uninitiated: if you think Labour will win more than 380 seats, you buy; if fewer than 370, you sell. Last time we looked at the spread, in June, Labour was 350-360. In May, it was 345-355, and I suggested that you buy large. Plenty of people must have taken my advice, because the difference between the two main parties has increased from 110 to 158. Even at 380, I would still buy Labour seats because, although 158 may seem a large gap, last time it was 253.

The main reason for the latest movement in the spread is the by-election results. However, even when the Tories briefly went ahead in the polls after the fuel crisis, they only went as high as 230. The bookies have taken more money on the next leader of the Tory party, which may give some indication as to why the spread has been lengthening in Labour's favour. Ladbrokes still has Portillo as favourite at 6-4, despite his claim that he doesn't want to be leader. Ann Widde-combe has seen her odds shorten to 3-1. Despite, or because of, her barmy views on cannabis, she is still very popular with the rank and file. Entering the odds for the first time since the election, at 7-2, is Ken Clarke; at a GMTV party, he told the Independent's political editor, Colin Brown, that he was willing to run.

The Tories are so bereft of credible candidates that Archie Norman is quoted at 5-1. I'm keeping my hands in my pockets on this one and saving all my money to put on Labour.