So is it a workable Tory majority or a hung parliament?

Since the budget there have been eight general election opinion polls – three from ICM, two from YouGov and one each from the other pollsters that do regular monthly surveys – MORI, ComRes and Populus.

What’s interesting is that the big Tory surge that’s both affecting the betting markets and the political climate generally has been based primarily on two pollsters, ICM and YouGov – the firms, incidentally, that are giving very different pictures of the London mayoral race.

What’s making me very nervous is that if you key the latest numbers from each of the firms into the seat calculators you get a range on Tory seats from 285 to 387. So a £50 a seat Tory buy bet at SportingIndex’s current 317 could produce a £1600 loss or a £3500 profit.

In the immediate aftermath of Darling’s speech ICM first survey moved from a 3% Tory lead to an 11% one. This has been followed by a 13% margin in its March survey for the Guardian and an 11% one last weekend when the Tories hit a modern day high with the pollster of 43%.

YouGov, meanwhile, moved from a 7% Tory lead at the end of February to an immediate post-budget one of 16% which then dropped a touch to 14%. The last two surveys showing a Tory share of 43% – the same as ICM.

Yet with the other three pollsters Tory progress has been nothing like as marked. Comres at the end of March was showing an 7% Tory lead down four points on its February figure; MORI saw a February to March Tory jump from 2% to 5% while Populus yesterday was also showing a 3% jump from its poll a month earlier.

With the exception of ComRes the Tory trend has been the same – only that the margins and shares have been so much bigger with YouGov and ICM.

Political gambling, particularly on the spread markets, can be an expensive past-time if you get it wrong and although I have a smallish Tory seat buy position I would be much more comfortable if other pollsters were following the big two.

I am a big opponent of poll averaging which I think is a flawed concept. See this excellent recent piece in the Wall Street Journal.

Mike Smithson