In an interview in the Daily Telegraph today, Lynton Crosby breaks cover to claim, inter alia, that “polls have become part of the political process so they’re not an independent measurement that says this is what’s going on, they actually influence what’s going on”. He goes on to argue that “there should be a stay on publishing polls publicly for two or three weeks before an election”.
The first part of Crosby’s case is half-right. Pre-election polls certainly affect “what is going on”, if by that one means media coverage of the election. However, they didn’t decide the result: if they had, Ed Miliband would doubtless be in Downing Street today, propped up in minority government by the SNP. On election day, the voters gave the pollsters two fingers, as Crosby himself might put it.
This being so, the case for a moratorium enforced by law simply hasn’t been proved. But in any event, there must be a strong presumption against bars and bans in a free society. If the pundits and pollsters are wrong – as most were in this case, including me on the central question – then let them be proved so, and be left to wipe the egg off their faces afterwards.
As we never tire of pointing out, one of the exceptions to that norm was Andrew “Punxsutawney Phil” Gimson who, at ConservativeHome’s Christmas lunch, forecast the Conservative total to within a single seat. I understand that Andrew is shortly to combine writing for this site with heading up a new megabucks forecasting operation. Get your hands off Gimson, Lynton.