There is normally relatively little polling for local elections for a variety of reasons, not least the uneven pattern of contestation across the the country. However ComRes have done one, with interesting topline figures. Local voting intention in those areas with local elections on Thursday stands at CON 31%, LAB 24%, LDEM 12%, UKIP 22%.
Now, we need to make a number of important caveats in understanding this:
It is NOT comparable with normal voting intention polls. It only covers the areas with local elections on Thursday, which are most rural Conservative shires and doesn’t include any Metropolitan counties… hence the fact the Conservatives are ahead. Neither is it comparable with the shares of the local election vote that the BBC and Rallings and Thrasher will calculate (the “Projected National Share” and “Equivalent National Vote”). These are both projections on what support would be across the whole country, not just where local elections are happening.
To understand the figures we need to know the votes last time round, which including the two councils (Durham and Northumberland) that actually last voted in 2008 were Con 44%, Lab 13%, LDem 25%, UKIP 5% – so the changes are Con down 13, Lab up 11, Lib Dem down 13, UKIP up 17. This suggests considerable bigger swings than Rallings and Thrasher have predicted.
By my estimates it would produce getting on for 500 Conservative losses and 250 UKIP gains, if it is giving an accurate picture… and local election predictions are not something that there is much track record for. We shall see
UPDATE: Peter Kellner and I have been pondering the number of UKIP seat gains if they do get 22% (the joys of the YouGov office on a morning before an election!) and how on earth you model gains when they are tripling the number of seats they contest. It’s very difficult, but I suspect I have overestimated it a bit… though even assuming a higher base level of support in the areas they didn’t contest in 2009 (and therefore a lower swing in the seats they did) if they do get 22% they should still be looking at well over 100 seats. Suffice to say, how many seats UKIP will get on Thursday is still incredibly hard to predict.