I don’t want in any way to suggest that there isn’t a problem with VL and this type of advertising. But there are real problems with the conclusion Carole is trying to draw from this graph.
First, the graph shows an upwards turn well before the period she’s talking about (the week before the vote).
Second, the graph might just show that campaigning (generally) works.
Third, I’m not sure what the source figures are here. The polls understated Leave till the end. What retrospective corrections are being made?
(The point there is that since the actual Leave vote was in fact higher than most of the preceding polls predicted, there must be a danger of seeing a last minute switch to Leave, when all that happened was that the polls the week before were underestimating Leave support.)
