Mike Smithson  of politicalbetting.com has posted the results of IPSOS-MORI analysis, which shows the swing to the Tories is greater in the LAB-CON marginals than in the country as a whole.

Unlike Smithson, I do not find this surprising. It is not due to an anti-Labour bias (though I confess, indeed proclaim this), but due to a simple analysis. There is considerable resentment of the current Labour government, that may surpass in the polls that of the Tories in the late nineties. That is, where they are able people will vote to get the government out. In a Tory or Lib Dem seat this will not matter. In a solid Labour seat, there is a de-motivating factor. But in a Labour marginal seat – and that can include seats with greater than 10% majorities, voting against the Labour candidate may help remove the government.

This is why I do not believe it when pundits say the Tories getting 40% of the vote against 30% for Labour will result in a hung parliament. We may not get a reverse of 2005, where Labour were on 35%, just 2% ahead of the Tories and still with a working majority. But the resentment factor is now mostly directed at Labour, and they will get punished accordingly, with the Tories being the principle beneficiaries.