Ipsos-Mori poll : SNP take double-digit leads on both ballots

Tonight's poll certainly lived up to the earlier billing, with the SNP moving into an 11-point lead on the consituency ballot, and a 10-point lead on the regional list. Here are the full figures for the four main parties -

Constituency vote :

SNP 45% (+8)
Labour 34% (-2)
Conservatives 10% (-3)
Liberal Democrats 9% (-1)

Regional list vote :

SNP 42% (+7)
Labour 32% (-1)
Conservatives 10% (-3)
Liberal Democrats 8% (-2)

By my reckoning this is now the eighth Holyrood poll in a row to show the SNP up on its winning 2007 share of the vote, and for the first time the seat projections point to a hypothetical parliamentary majority in favour of holding an independence referendum.

But now for the reality-check. Ezio suggested on the previous thread that these figures could be a bit too good to be true, and my gut feeling is that might just turn out to be right. For starters, the fieldwork dates partly overlap with the weekend YouGov poll that showed a much narrower lead for the SNP. Admittedly that can be partly explained by the fact that a greater percentage of the YouGov sampling took place before the party's manifesto launch and Alex Salmond's well-received appearance on Question Time, but I wonder if there are methodology issues at play as well. Ipsos-Mori of course famously showed an SNP lead in February when in retrospect it seems likely that Labour still had the upper hand, albeit possibly not by much. Rather more troublingly, in the run-up to last year's Westminster general election the company also produced figures that don't seem even remotely credible with the benefit of hindsight.

Certainly the combined evidence of the two polls points overwhelmingly to some kind of SNP lead (at least as of the weekend) but just how significant a lead is much more open to question at this point. Also noticeable in Ipsos-Mori's figures is that the Conservatives seem to be struggling almost as much as the Liberal Democrats, which is somewhat at variance with what other pollsters have been suggesting.

*

This is an issue I've raised before, but I was slightly troubled to see Alex Salmond put all his eggs in one basket on Newsnight Scotland and more or less assert that the party that wins most seats will form the Scottish government. OK, that's probably a very welcome sign of confidence at this stage in proceedings, but it still seems like a totally needless hostage to fortune, not least because it's perfectly conceivable that the SNP could win a mandate in terms of the popular vote, but fall just short in terms of seats due to a quirk of the electoral system.