ICM on Peter Mandelson

The tables for ICM's poll in the Sunday Mirror are now up here, so we can have a proper look at those Mandelson figures that the Sunday Mirror got so excited over.

Firstly, the figures in the Sunday Mirror were repercentaged to exclude don't knows. Explaining the rather high numbers when normally "who you you like to see as the next leader of the X party" normally gives a thumping victory for "Don't know".

Indeed, that was the case here - 55% of people said don't know. If you include don't knows in the percentages, the headline figures become David Miliband and Jack Straw on 12%, followed by Mandelson and Johnson on 7%. 4% would like to see Harriet Harman as the next Labour leader, 3% Ed Balls and 1% Jon Cruddas. Amongst Labour voters, the top four candidates are all very even - 13% for Miliband, 12% for Straw, Johnson and Mandelson.

There was also a question asking if people would be more or less likely to vote for Labour were it led by Peter Mandelson instead of Gordon Brown. Regular readers will know that I am not a fan of questions like this - people can use it to indicate support or opposition to a candidate or policy when there is no chance of them actually switching their vote, and you invariably get lots of people saying it would make them more likely to vote X who already vote X, and lots of people saying it would make them less likely to vote Y who wouldn't vote Y anyway.

For the record though, 7% said they would be more likely to vote Labour with Mandelson in charge, but 20% said they would be less likely. 14% of current Labour voters said it would make them less likely to vote Labour, but 5% of current Conservatives and 8% of current Lib Dems said it would make them more likely.