CON and UKIP edge up too; Others and LDs take a knock
The one-time received wisdom that August polls should be ignored because summer sun and silly season made sampling and support unstable took a good hit last month when Mike noted the accuracy of ICM’s August polls in the run-up to three of the last four elections. Of course, polls are snapshots not predictions and are bound by a political equivalent of Newton’s First Law: polling taken outside periods of turbulence will remain steady unless subjected to events, dear boy.
Put another way, in the absence of significant shocks to the system, the movement in polling over a month is more likely than not to give a good indication of how underlying opinion is moving. On that level, Ed Miliband will be a happy man. The Politicalbetting poll average figures for August were:
CON 32.1 +0.3
UKIP 14.0 +0.8
LD 8.5 -0.4
OTH 10.0 -1.2
Those are not particularly big movements of themselves but what may be more significant is that it’s the third consecutive month that Labour has put on around half a point, which adds up to a mini-trend, bouncing back from their slump in support in May (which it has to be said was a singularly poor time to have such a drop and the implications of that coincidence of timing and movement can’t be wholly ignored for 2015). It was also the third month in a row that the Lab-Con gap has grown, edging out from 2.2% to 3.3% over the period.
Elsewhere, UKIP recovered a little from their own slump last month while the Tories also nudged upwards slightly, though I wouldn’t read too much into either movement: UKIP’s figures have been quite erratic over the summer, if in a solidly good range, while the Conservatives’ movement is small and remains in a band centred on 31.5%, as it has for over a year now. By contrast, the Lib Dems went backwards, falling to their second-lowest monthly score in the series. The minor parties too dropped back, though in their case it was from their highest point in the parliament.
Given that we know that there are two big political events coming up, one of which (Clacton) could be quite significant historically, the other of which could be overwhelmingly so, do the polls taken during a quieter time before then matter? That depends on both the outcome of those votes and on the various parties’ reactions to them. If nothing else, it gives an idea of where the underlying momentum is and if the two votes do turn out to be damp squibs in the medium term, that’s not inconsequential.
— Mike Smithson (@MSmithsonPB) September 2, 2014