MORI Scottish poll shows 29 point lead for the SNP

There has been comparatively little Scottish polling since the referendum (partly I expect because newspapers had spent their budgets on pre-referendum polling). I've seen some people spending rather too much time collating and looking at Scottish crossbreaks in GB polls. Personally I wouldn't recommend putting too much weight on crossbreaks - aggregating them up gets round the sample size issue, but GB polls are still weighted at the GB level. If you think back to how het up people got about whether Scottish polls were weighted by Holyrood or Westminster voting intention, factored in place of birth, things like that - the Scottish sub-sample in a GB poll have no such controls, it's just how the Scottish respondents in a poll weighted to GB targets happen to fall out.

Nevertheless, they are a straw in the wind, and they'd been suggesting a strong showing for the SNP since the referendum. Today we have a proper, bespoke Scottish poll by Ipsos MORI and if anything it shows the SNP doing even better than the crossbreaks suggested. Topline voting intentions in Westminster with changes since the general election are CON 10%(-7), LAB 23%(-19), LDEM 6%(-13), SNP 52%(+32), GRN 6%(+5). Full results are here

This would, to say the least, be rather a radical turnaround from the last general election. I don't think swingometers offer much guidance in the case of really extreme results (a uniform swing would be mathematically impossible on this results - for example, there are about 9 seats in Scotland where Labour got less than 19% in 2010, so couldn't lose 19% this time round. The same applies in many seats for the Liberal Democrats) but for the record on a uniform swing these figures would result in the SNP winning all but two seats in Scotland.