is labour being seduced by the "uniform swing"?

American number-genius Nate Silver has applied his very considerable smarts to the British election. He writes to warn of the inadequacies of translating vote shares into seats via the notion of a "uniform swing". The BBC and other media outlets like the notion of the uniform swing because it's very easy to calculate, and makes it simply to turn polls into headlines. You just take the projected share of the vote, work out the swing from the last election (eg Labour seven points lower) then apply it to every constituency and see what comes out the other end. This works quite well when you're talking about transfers of votes from one party to the other. But when you're in a three-way-race - actually a four-way-race, with "others" taking up much a larger share of the vote - it can become very unreliable.

Silver's initial conclusion is that Labour's structural advantage in this election - which many in the party seem to be hoping will save them from devastation on May 6th - may be being overestimated because of overconfidence in projections based on uniform swings.