— BBC Newsnight (@BBCNewsnight) May 26, 2017
I spent the whole of yesterday in the sweltering heat of central London discussing with one group after another the same subject – how GE2017 had suddenly become so much harder to predict.
During the morning I was at an academic conference at the British Academy on the
intriguing subject of whether June 8 would see another polling fail. We looked in detail at the quite considerable changes polling firms had made following the May 2015 mess up.
The pollsters are operating very differently to last time with online now predominant and a focus on reaching and getting the views of those who can often be overlooked – those with little interest in politics.
Lunch with a strong LAB supporter from Twickenham who this time will be tactically voting for Vince Cable which I took as an interesting pointer.
In the evening there was the excellent PB pre-general election gathering where everybody had things to say about the extraordinary move in the polls but there was a reluctance to make firm predictions.
If there was a consensus view it was along the lines of that discussed on Newsnight later by the great Sir David Butler (see clip above) who for eight decades has been observing and commenting on general elections. Theresa’s Tories are going to win but we should lower our expectations on the scale of the victory.
There’s never been an election campaign where a party was chalking up 20%+ poll leads only to see a decline to single figure leads in a week and a half and one pollster having it at 5%.