I've been mulling over how I can best keep track of Scottish voting intentions for next year's Westminster general election. Full-scale Scottish polls are likely to be thin on the ground, so any method is inevitably going to rely heavily on Scottish subsamples from GB-wide polls. That's far from ideal, but an even bigger problem is the distortion caused by the fact that YouGov produce far more polls than anyone else (five per week), and that they tend to be much less favourable to the SNP than any other firm. However, I don't really see any way round that if I want to keep the figures as up-to-date as possible. So this is what I've come up with for the new version of the Scot Goes Pop Poll of Polls -
1) All polls entirely conducted within the last seven days will be included, as long as the datasets have been published.
2) Any newly-published full-scale Scottish poll will be included even if the fieldwork falls partly outside the seven-day period, and will be given ten times the weighting of a subsample.
Putting that into practice for the first time, this is what it produces...
Scottish voting intentions for the May 2015 UK general election :
Liberal Democrats 6.8%
That's based on eight subsamples - four from YouGov, two from Populus, one from Ashcroft and one from ComRes. Incredibly, every single one of those subsamples has the SNP in the lead. However, it's worth bearing in mind that the only full-scale Scottish poll to have been conducted since the referendum (from Survation) actually had the SNP behind in Westminster voting intentions, albeit narrowly and within the margin of error. That poll isn't included in the above figures because the fieldwork ended more than seven days ago.
There should also be an Opinium poll in the mix, but unfortunately I can't include that because the datasets haven't been published yet.
The million dollar question is whether the SNP can possibly maintain this extraordinary level of support as we move further away from the referendum. But for the time being at least, the state of play is causing a rather amusing degree of concern among the Nat-bashing usual suspects in London. This is perhaps the most intensely satisfying tweet I've read over the last two weeks -
John Rentoul : "Alarming analysis by Peter Kellner suggesting SNP might win 26 of 59 Scottish seats, up from 6"
It's incredible, isn't it? Before the referendum, all Rentoul cared about was keeping Scotland within his beloved country, come hell or high water. Well, he got what he wanted, but he's still scared witless, because he's suddenly realised that the Scotland he "kept" is Scotland as it actually is (a "region" that votes in large numbers for the SNP and for self-government), rather than the "British" Scotland of his imagination.
Alex Salmond always used to say that independence would lead to England losing a surly lodger and gaining a good neighbour. Well, it became abundantly clear during the campaign that the London establishment wanted to keep their surly lodger at all costs - they lied, bribed and bullied to achieve that objective, and now they're going to get exactly what they asked for. If the SNP end up holding the balance of power at Westminster for the next five years as a direct result of the anti-independence terror campaign, it would be one of the most delicious ironies in human history. More pertinently, it would also take us a big step closer to Devo Max.
That's the prize, if we can all keep our focus for the next seven months.