I know it's a bit cheeky of me to say this, given that it's only a week since I wrote an entire blogpost about a subsample, but don't pay too much attention to the excitement over the Ipsos-Mori subsample putting the SNP on 55%, Labour on 15% and the Tories on 12%. Once undecideds and unlikely-to-votes were removed, the sample size was a tiny 60, and the results won't have been properly weighted anyway. Of course if there had been a sustained pattern of subsamples putting the SNP well over 50%, then it might be significant, but that simply isn't the case. High 30s and low 40s has been much more common.
What is more meaningful and interesting about the poll is the Britain-wide result, and in particular Ipsos-Mori's decision to follow the practice of Opinium and ComRes by including the Independent Group as an option on a standard voting intention question. As has been the case in the Opinium and ComRes polls, the Independent Group performed significantly worse than in polls from other firms asking hypothetical questions that make a song and dance of reminding respondents that there is a new kid on the block. But at least Opinium and ComRes have put the Independent Group somewhere between 4% and 8%. Ipsos-Mori, by contrast, have them barely troubling the scorer.
Britain-wide voting intentions (Ipsos-Mori):
Liberal Democrats 8%
Independent Group 2%
Plaid Cymru 1%
Brexit Party 1%
We don't have to look far for an explanation for the divergence - Ipsos-Mori polled by telephone, whereas Opinium and ComRes used an online panel. It's an open question as to whether telephone or online data collection produces more accurate results at present (online polls surprised everyone during the EU referendum by being closer to the final result), but it seems intuitively likely that telephone would at least be superior when it comes to testing support for the Independent Group. Online panels are likely to have a disproportionate number of politically-engaged respondents who are more aware of the new proto-party's existence than the average person on the street.
So if the Independent Group really are on 2%, it's not hard to see how a snap general election would pose an existential threat for them - and that could be exactly where we're headed. The theory seems to be that if parliament coalesces around a softer Brexit as a solution to the current crisis, the anti-Europeans on the Tory benches will sabotage it by bringing about an election - either by the direct means of voting against the government on a no confidence motion, or more likely by a show of strength that demonstrates the government will no longer be able to govern without their help. Which could leave Chuka and co making panic-stricken do-or-die overtures to potential defectors - but how many Labour or Tory MPs are really going to throw their careers away a few weeks before an election that could secure their positions for another five years if they just sit tight?
This may prove to have been one of the most ill-timed breakaways in political history.