Sunday, March 8, 2015

SNP lead by 15% in new Poll of Polls

It's been a while since I last updated the Poll of Polls, and I was interested to see how this one would work out, because the Scottish subsamples over recent days have been all over the place.  There were a couple of narrower gaps reported by YouGov on Thursday and Friday, but any slight worries we might have had over those were allayed by huge SNP leads in Friday's Populus subsample, and in today's YouGov poll for the Sunday Times.

This update is based on seven Scottish subsamples from GB-wide polls - five from YouGov, one from Populus and one from Opinium.  As a number of us have been pointing out for a few weeks, the new Opinium methodology is utterly bonkers, with their Scottish subsample figures practically constituting works of fiction - so the fact that Opinium accounts for as much as one-seventh of today's sample effectively means that the SNP lead is being artificially adjusted downwards.

Scottish voting intentions for the May 2015 UK general election :

SNP 42.1% (-1.7)
Labour 26.9% (+3.0)
Conservatives 18.7% (+1.3)
Liberal Democrats 5.4% (-1.5)
Greens 3.6% (+0.3)
UKIP 3.0% (-1.5)

(The Poll of Polls uses the Scottish subsamples from all GB-wide polls that have been conducted entirely within the last seven days and for which datasets have been provided, and also all full-scale Scottish polls that have been conducted at least partly within the last seven days. Full-scale polls are given ten times the weighting of subsamples.)

*  *  *

Kenny "Devo or Death" Farquharson live-tweeted Jim Murphy's speech at the Scottish Labour conference yesterday, and there was one line in particular that made me laugh -

"Reminds me of John Smith's platform speeches, some of which were awful."

It took a good twenty seconds for the implication to fully dawn on me.  It was a bit like saying : "Jim isn't always a good public speaker, but then Jesus had his off days as well, which means...oh my God...wow."

21 comments:

  1. had a look at the Opinium tables, and I for one ,cannot make heads nor tails of it ..Are they saying labour is ahead in Scotland ? by a large margin too ...Seriously deluded or maybe not ?
    Can some-one explain it to me ?

    ReplyDelete
  2. If the Opinium poll is downweighting the average, is it likely we're still talking a 16-18% lead for the SNP?

    ReplyDelete
  3. Although I think John Smith is hugely overrated, Kenny Farq's assessment of Murph's speech reminds me a little of this Calvin and Hobbes.

    ReplyDelete
  4. I'd advise avoiding betting off the back of that Opinium subsample ;)

    ReplyDelete
  5. "SNP 42.1% (-1.7)
    Labour 26.9% (+3.0)
    Conservatives 18.7% (+1.3)
    Liberal Democrats 5.4% (-1.5)
    Greens 3.6% (+0.3)
    UKIP 3.0% (-1.5)"

    Good news for the Scottish Conservatives. That percentage is in line with their aggregate polling average from Electoral Calculus. They're up, which isn't surprising given Ruth Davidson is the only credible opposition leader left standing against the SNP.

    Labour are a shower, the LDems don't exist anymore and the Greens are signed up to the SNPs YES revivalism band.

    ReplyDelete
  6. The sub-sample results over-rate the Tories by a few points. They are down compared to 2010 in every single one of the 24 constituencies polled by Ashcroft, albeit their decrease is normally less than that suffered by Labour or the Lib Dems. The Tories are also significantly lower in the Scottish national polling than in the results implied from the GB polling.

    In fairness to the Tories and Davidson, this may be due to some Tories in hopeless seats voting tactically rather than a further erosion in their support, but there is no credible sign of a Scottish Tory revival. They may gain a seat in May compared to 2010, but that would be due to the collapse of others rather than anything done by them.

    ReplyDelete
    Replies
    1. Dave just scored his worst sat rating in Scotland in UK Yougov polls.

      Maybe a sign his 'Don't let the Jocks get a say in UK governance' approach is bearing fruit?

      Delete
    2. Sure there's noise due to being subsamples, but a sudden drop for all three amigos and notably for DC who's actually been seen as the least worst of them in terms of party leadership?

      http://www.statgeek.co.uk/charts/scotland-leadership-ratings-six-month.png

      From Statgeek of UKPR

      Delete
    3. What stupendous popularity!

      LOL

      Another triumph for the cowardly fop, calamity Clegg and little Ed. :-D

      Delete
    4. It's mind-boggling that Miliband is polling lower than Clegg in Scotland.

      Delete
  7. Extrapolating long term trend to May 7th

    SNP: 43.5%
    Labour: 26%
    Conservatives: 18%
    Lib Dems: 5.5%
    Greens: 3.5%
    UKIP: 3.5%

    ReplyDelete
    Replies
    1. That gives a 44/11/3/1 seat split on Electoral Calculus. I'd put that at the upper limit of what the SNP could expect.

      Delete
    2. I just don't know any more. Remember this? http://wingsoverscotland.com/how-many-chickens/

      That was mid-April of course, not March. But it was a near-incredulous response to a poll showing the SNP on course for 63 Holyrood seats. The reality was 69.

      I'm really frustrated today. There are stacks of leaflets to be delivered and I'm laid up with the flu. But on the other hand these leaflets are not languishing in my garage, as they would have been even a year ago. A friend has comandeered my book of route maps and is busy doling them out to a small army of eager volunteers that has materialised almost from nowhere. The village is nearly done, she says, and she can maybe get some of the rural routes organised that need a car. "They're really enthusiastic," she says.

      It's a whole new world out there.

      Delete
  8. Populus sub-sample: SNP 49, Lab 23, Con 12, LD 12.

    http://www.populus.co.uk/wp-content/uploads/OmOnline_Vote_09-03-2015_BPC.pdf

    Populus are now giving higher sub-sample scores for the SNP than YouGov because there is practically no down-weighting in their polls.

    ReplyDelete
  9. Just had a look at the Daily Redcoat on-line (yes I know)

    Wanted to see how they would spin and hype the Weekend Conference, but all I could see was an article about how Kezia would be helping the poorest kids, subject to approval.

    Ok, I added the 'subject to approval' part, but only because I know Labour Double-Speak.

    If the Redcoat doesn't even try to spin it, it must have been as dire as other reports tell us.

    ReplyDelete
  10. Just for fun I played around with the numbers on Electoral Calculus and tried to see what the lowest level of SNP support which would return all 59 seats for the SNP. I managed it with 50% for SNP, 20% for Labour, 15% for Cons and 5% for each of the rest. Anyone manage it with any smaller number for SNP?

    It might be a bit of wishful thinking but it doesn't seem too far away from this Populus sub-sample.

    Eppy

    ReplyDelete
  11. I made my own predictor using James's poll of polls, the 2010 results, and the Ashcroft polls and get the following seat numbers:-

    SNP 54
    LAB 2
    Lib Dem 2
    Con 1

    That's assuming poll weightings are valid. If they're downweighting too much then it could be almost a clean sweep with only the northern isles holding out.

    ReplyDelete
  12. Ashcroft subsample:
    54% SNP
    21% Lab
    16% Con
    5% Lib
    3% UKIP
    1% Green

    ReplyDelete
    Replies
    1. Those figures (if applied to all Scotland) give me:
      SNP 57
      Lib Dem 1
      Berwick Rox Selkirk is a 3 way split, Tories just pipping SNP with those numbers.

      Delete
    2. I'd grudge the Tories a seat further north less than I grudge them any of the three in the extreme south. Not only are they geographically big, so spoil the map, they make it look as if Scotland doesn't start until about Lanarkshire. Oh yes and I live in one of them.

      All these years, and I still might not get to live in an SNP WM constituency. Even though the one I voted in for nearly 55 years will probably fall. (Bye bye Frank Roy.)

      Delete