Valentine's vapourisation for doomed Dugdale after devastating Panelbase poll

Someone said to me a couple of weeks ago that I'd be kept busy as May approaches, because of a flurry of opinion polls for the local elections.  I had to explain that dedicated polls for local elections are very rarely published, and that I wouldn't be surprised if there were none at all.  So well done to Stuart Campbell for proving me wrong as early as Valentine's Day by commissioning a Panelbase/Wings local election poll.  What leaps out straight away is that there's precious little difference between Westminster voting intention and local council voting intention - the figures are almost identical to Panelbase's recent Westminster poll, apart from the fact that the Greens and UKIP are both doing a bit better.

Panelbase poll of local council voting intentions (percentage changes are from last local elections in May 2012) :

SNP 47% (+15)
Conservatives 26% (+13)
Labour 14% (-17)
Liberal Democrats 5% (-2)
Greens 4% (+2)
UKIP 3% (+3)

Now you're probably thinking that those percentage changes don't make a great deal of intuitive sense - both the SNP and the Tories seem to be making a lot of their gains out of thin air.  The explanation, of course, is that many of the people who say they plan to vote SNP or Tory voted for an independent candidate in 2012, and whatever they may currently believe or tell a pollster, a lot of them will do exactly the same thing again this time.  I would very confidently predict that the SNP will fall short of 47% in the popular vote, and I would also predict with a reasonable amount of confidence that the Tories will fall short of 26%. Essentially people weren't thinking properly about the question they were actually asked, and were giving a 'parliamentary' answer instead.

The good news for Labour is that they suffer less than the other parties from competition with independent candidates, because their strength (such as it is) is to be found mainly in urban areas.  So this poll may not be overstating their true position.  The bad news is that 14% is an absolutely desperate position.  Their vote has seemingly more than halved in the last five years, and even assuming they pick up a reasonable amount of transfers from people who give their first preference votes to the Tories, there surely isn't a cat in hell's chance that they're going to retain majority control of any council at all.

Incidentally, Panelbase are blazing a trail with this poll - as far as I'm aware, this is the first Scottish poll from any firm to put people born in other EU countries into their own distinct category, and presumably weight them separately.  As it turns out, they didn't need to be either upweighted or downweighted significantly in this particular poll, and even if an adjustment had been required, it wouldn't have had much impact on the party political numbers.  But this innovation should lead to greater accuracy in future independence polls - because we know anecdotally that there has been a particularly strong swing to Yes among EU citizens.  Hopefully other polling firms will now follow Panelbase's good practice.

UPDATE : I've just caught up with Stuart Campbell's tweet from 24 hours ago, which implies the poll also asked an independence question that has yet to be released.  Panelbase have apparently said it shows "no real change" in public opinion - that's measured from a 46% Yes vote in the last poll from the firm.  I'll indulge in some wild speculation here and suggest that this probably means there has been a 1% or 2% change since the last poll, and that a small increase in the Yes vote is more likely than a small decrease.  (My reasoning is that a small decrease would take Yes to 45% or 44%, and as that would be unusually low by post-indyref standards, I doubt if that would be casually dismissed as "no real change".)

*  *  *

If you've enjoyed my writing in recent months and feel a strange inexplicable urge to 'buy me a hot chocolate', bear in mind that my fundraiser from two years ago is still open for additional donations - it can be found HERE.