The poll in today's Scotland on Sunday has been discussed elsewhere, but not the Holyrood regional voting intention. It's the best bit, and Kenny F sent me out to the shops this morning especially for it.Holyrood constituencies (2007 result in brackets)
Labour: 33% (32%)
SNP: 28% (33%)
Conservative: 16% (13%)
Liberal: 16% (14%)
Others: 6% (8%)
Holyrood regional
Labour: 31% (29%)
SNP: 26% (31%)
Conservative: 17% (14%)
Liberal: 14% (11%)
Green: 7% (4%)
Others: 6% (11%)
By Scotland on Sunday's calculations (and mine) that would put us up to seven Green MSPs again. The pic shows what that looked like last time. It's a crude summary, but if this were the actual vote shares in May next year, we'd get a result roughly like the 2003 election, except where the six former SSP MSPs were replaced by the SNP. The obvious post-election arrangement would be another Labour/Liberal coalition, too, although Labour have watched the SNP's minority administration enviously.
Cheering as this poll may be for Greens, it's even more A Bit Of Fun than usual. No Holyrood voting intention will be any kind of worthwhile prediction until the UK election results have bedded in. Will Cameron woo or alienate Scots? Could Gordon Brown even hang on? Might the Liberals get the hung Parliament they crave? Could an AV referendum become a true PR election? Might UKIP get beaten by the Monster Raving Loony Party?
The Holyrood polls will start to get properly gripping for anoraks from September, by my calculation. One last factor which might make a difference is Brighton Pavilion. I'm heading down on Friday to help Caroline Lucas get elected. It should be fun, as well as virtuous, and, if she wins, the extra profile for Greens nationally could help us out in 2011 too.
James,
One other question to ask in terms of extrapolating poll numbers is this: what boundaries are they using? AFAIK, Scotland Votes are still using the old boundaries. The figures I have for the new ones may be different to what they're getting. So I still get seven Green MSPs (one in every region except Central), 48 Labour (up on 2007, down on 2003), 35 SNP (greater than SNP + SSP in 2003 and the same tally as in 1999), 21 Tories (a party best), 18 LibDems (another party best) and Margo, in the event that she stands again (most likely an extra SNP seat if she doesn't).
So although the boundaries aren't totally finalised, the chances of any major changes at this stage are now extremely limited and we can start using the new boundaries to extrapolate polls for 2011.
Good questions. Scotland Votes also assumes Margo gets elected every time even if you have others/independents at zero. My guess is that SoS are using a Curtice analysis based on new boundaries, given he commentated on it, but it's not clear. I'll ask Kenny.
I wouldn't be too surprised if they were still using the old ones: although the Constituency boundaries haven't been finalised, the Regional ones haven't yet (though will be soon) and nothing has been made legal as yet. So I suppose, really, extrapolation depends on the question. If they're asking, "How would you vote if the election were held tomorrow?", then the old boundaries would hold, they're right and I'm wrong. If it's "How would you vote in 2011?", then they should be using the new ones...