Friday, August 4, 2017

Lament for the Colonel as Scottish Tories remain in third place in YouGov subsample

There's been such a drought of polling recently that I was beginning to think we weren't going to get anything more until the end of the English school holidays.  However, a Britain-wide YouGov poll suddenly appeared today, suggesting that Labour's very narrow lead over the Tories has stabilised after falling from a peak of 8 points in early July.  The Scottish subsample continues to show the now familiar tight three-way battle: Labour 33%, SNP 29%, Conservatives 28%, Liberal Democrats 7%, Greens 2%, UKIP 2%.

The most significant thing about those figures is that the Tories are in third place, which has consistently been the case in all four post-election YouGov subsamples.  The lead has been switching back and forth between the SNP and Labour, so it's anyone's guess which of those two parties would be in first place if YouGov conducted a full-scale Scottish poll of Westminster voting intentions right now.  An average of the four subsamples produces an exact dead heat: SNP 32.3%, Labour 32.3%, Conservatives 26.3%, Liberal Democrats 6.0%.

On a more positive note, the SNP have had the lead in the majority of subsamples conducted across all firms.  There have been fourteen subsamples since the election, with the SNP ahead in eight, Labour in five, and the Tories in only one.  The SNP have been in either first or second place in all fourteen subsamples, whereas both Labour and the Tories have been in third place in some - underscoring the general impression that the SNP are the party most likely to have a small overall lead.

51 comments:

  1. Frankly the Tories losing barely anything since June is nothing to celebrate. And those YouGov numbers on a uniform swing would mean an SNP collapse on seats, fourth place behind the Lib Dems. (Labour 34, Tories 17, Lib Dems 5, SNP 3). FPTP for you.

    But what we really need is a Holyrood poll. We have no idea if the Labour surge has spread there and if so to what extent. If the nationalist majority faces being wiped out in 2021, then I’m not sure what they should do re indyref2. Certainly the Tory gov at Westminster wouldn’t be in a rush to grant one if they think they can wait it out.

    ReplyDelete
    Replies
    1. I think we should be moving on from asking Westminster to "grant" a referendum. We should simply hold a referendum (subject to legal challenges) when the time is right.

      There isn't going to be a uniform swing, and in my opinion the idea that the SNP could fall from 35 seats to only 2 or 3 is pretty unlikely (unless things get a lot worse than they are now). Incidentally, on the YouGov average, the SNP have slipped 4.6% since the elecion, and the Tories have slipped 2.3%. Both parties seem to have taken a hit due to the Labour surge.

      Delete
    2. What if the legal challenges succeed or Westminster simply ignores the outcome of such a vote?

      As a non-independent country (our choice, remember), we cannot make these decisions unilaterally.

      Delete
    3. Self-determination counts for something. Even the Spanish government are being confronted with that reality.

      Delete
    4. Now weighted on 2017 voting. Those who voted in 2015 and didn't in 2017 might turn up next time but may not be included as likely to vote.

      Delete
    5. Absolutely. A poll putting the SNP in the 30s now means something different from a poll putting the SNP in the 30s in May.

      Delete
    6. You are assuming that a missing % of people who previously voted SNP and now don't will turn up again at some point in the future to help you out and that this will sufficiently outweigh any similar effect involving the other parties to allow you to win. Clutching at straws here, no?

      Delete
    7. The reason I use the words "grant" a referendum is because, realistically, holding one without Westminster's approval would be pointless. I'd be welcome to hear any suggestions as to why unionists would even participate and Westminster recognise the result if it wasn't an agreed vote.

      I also agree that falling so far is unthinkable, but Labour thought that before 2015. We have to be aware that the SNP's support is so ridiculously evenly spread geographically that it's not totally implausible to come 2nd in every single seat to different parties. I don't really share your optimism that a uniform swing is unlikely in any future election. If anything, the situation for the SNP could be even worse as the bulk of the SNP > Labour vote switching will be occurring in seats that the SNP still hold. Certainly the Tory vote has topped out and they've already taken most of their target seats.

      Delete
    8. Argyll and Bute and Perthshire North are 2 seats that could easily be won by the conservatives next time. There quite possibly are others out in the sticks / Edinburgh direction.

      Delete
  2. Scotland Votes, Westminster:

    Labour: 30 MPs
    SNP: 14 MPs
    Tory: 8 MPs
    LibDems: 7 MPs

    Based on the averages given above.

    ReplyDelete
    Replies
    1. I am a card carrying Conservative however, should this come to pass, I will welcome Scottish Labour as liberators.

      Delete
    2. You must chuck flowers at Corbyn and McDonnell as they do their victory march down the Royal Mile (to the tune of The Internationale).

      Delete
    3. Wrong James, the Soldier Song more like. Those two are as socialist as the Nat sis.

      Delete
    4. As a conservative it is disappointing that, once again, Scotland thinks socialism is the answer - be it socialism in Scotland alone or socialism in the whole of the UK - it always seems to be the answer to everything up here. Who speaks for liberty?

      But if I absolutely must choose between the two then Labour it is. Like the Red Army sweeping through Eastern Europe in 1944-45, it's bad - but better than what is being replaced.

      Delete
    5. Comrade Tommy was the answer! Where is he!! No not Tommy Steele and the Little White Bullshit.

      Delete
    6. Tommy Sheridan - last seen trying to convince rural farming communities and middle class suburbanites that socialism is the way forward, brothers and sisters, and you should vote Scottish independence to get it.

      Delete
    7. The unionists are putting the band back together...

      Delete
  3. James where was the Labour surge in Scotland at the *actual* GE17 ballot? There is a reason Britain Elect does not advocate use of sub-samples as useful indicators. I fully respect your knowledge on the science (dark arts?!) but there's a real risk of misdirection related to actual current opinion\voting intent on the ground in Scotland by giving these sub-samples repeated credibility they do not merit (even if they are being judged over time). If sub-samples were sufficient indicators we wouldn't need full blown polls but that is not the case.

    ReplyDelete
    Replies
    1. All I can say is that I think you're incorrect. The mantra about subsamples not being credible is taken so literally that people refuse to believe the evidence of their own eyes, even after a large number of subsamples over a long period of time. It was blindingly obvious from the subsamples in October 2014 that a massive SNP surge was underway, but most polling sites didn't report it (and didn't even mention it in passing) because they were "only subsamples".

      YouGov do apparently weight their Scottish subsamples separately, so the likelihood is that a full-scale Scottish poll would show something reasonably similar to the average I mentioned in the blogpost.

      Delete
    2. James - it doesn't detract from the mathematical reality that a UK wide poll of 1000 with 100 Scots will have a margin of error around +\-10%. I don't recall all the 'subsample evidence' re the SNP's post-Indyref surge and whether that potential example is the exception rather rule; however, I think I'd like to see a spread of full scale Scottish polls re Westminster and Holyrood voting intentions before concluding Labour had reeled in the Tories and the SNP.

      Delete
    3. "it doesn't detract from the mathematical reality that a UK wide poll of 1000 with 100 Scots will have a margin of error around +\-10%"

      Which becomes much less relevant once you have lots of them. Nobody is saying an individual subsample tells us very much - but that's not the point.

      Delete
    4. James - it doesn't detract from the mathematical reality that a UK wide poll of 1000 with 100 Scots will have a margin of error around +\-10%.

      Strictly correct, but when you have multiple samples showing in the same range, the aggregate margin for error reduces as well.

      It's not a linear relationship, because you may double-sample. So 5 subsamples of 200 is not quite as accurate as 1 sample of 1000, but it's still a significant modifier.

      Delete
  4. Unionists well ahead of the nat si. A good Brexit from the EU criminal mafia will see the Nat sis in the dustbin of history...
    The Jock parliament will have to halve its waster MSP'S eventually...They should do it voluntarily instead of moaning as Jocks do.

    ReplyDelete
    Replies
    1. What the Nat sis have never explained is why they want to be subservient to the EU criminal fraternity and turn their back on our neighbours. It can only be hatred of the English.

      Delete
    2. Could also be hatred of the Welsh. Why's no one ever suggest that?

      Delete
    3. The Welsh are very very nice people who could hate them! You Nat sis hate your own people thus handing their fate over to the EU. You are scumbags and you know it...

      Delete
    4. The EU aims to be a superstate and will increase centralisation vastly once Britain - a long term antagonist to EU federalism - is gone. There can be no independence in Europe. But there can be independence as part of a free United Kingdom, doing trade deals with our friends and allies that don't involve free movement of people or loss of sovereignty.

      Delete
    5. Aldo: the Remain voter who affects happiness that his side lost. You sound like Theresa May.

      Delete
    6. You Nat sis will never find happiness but you could sing along with Ken Dodd.

      Delete
    7. I voted for brexit in the end. Couldn't do the remain thing - it felt like treason.

      Delete
    8. Best not accuse independence supporters of being guided by feelings over reason, then.

      Delete
    9. Scotland is an independent nation in a Union with her neighbour. You fascists want to arrange Scotland to be a local authority in the former Third Reich.

      Delete
    10. There was plenty of reasoning in there as well. The UK out of Europe is a viable entity and can do extremely well, provided we don't panic at the first sign of short term turbulence.

      Delete
    11. The EU is German dominated, no doubt about it. Even France has to bend the knee to them now. They've finally done it - domination of Europe, with no shots being fired, this time.

      To take Scotland into that and say that we will be independent is laughable. We'll do what the German chancellor wants, relayed to us by a layer of eurocrats in between us and him/her - so it doesn't look as bad - and adopt a currency that will destroy our economy while further boosting the German economy. We will do all of this, presumably, whilst continuing to 'honour' the war dead. Haha!

      Delete
    12. And the Irish PM was in NI bumping his gums about Brexit. The Sinn Fein IRA slaughterd innocents for independence then sold themselves and Ireland to the EU. What a tragic waste of life down to a bunch of wasters.

      Delete
    13. Better together, GSTQ FTP WATP

      Delete
    14. When the Republic of Ireland became independent from us, we also became independent from them. Whatever we choose to do with our own country is not their concern.

      I would be interested to see who will enforce a hard border if British troops and police refuse. It would have to be the Republic and / or forces from continental Europe. Good luck guys - you'll need it!

      Delete
    15. Dont be stupid aldo how are we going to stop Turkish immigrants entering England through the back door. Thats why we stopped the Nat sis getting their separatist nirvana. We need to build a wall a really big wall to keep the papish terrorist lovers out

      Delete
    16. The UK government has already said it doesn't want a hard border though. So we should just say to Eire and the EU, "over to you". If they want to police their side of the border, they can. When the return to 1970s violence comes, it wont be our fault and it wont be our guys in the firing line.

      Delete
    17. Certainly not. I just can't stand the EU and so anything that hurts them or makes them look bad, we should do it.

      Delete
    18. Wee tiff between unionist bedfellows there...

      Delete
    19. Aldo is a perfect example of why pollsters are having problems being accurate. NOW he insists he was for brexit. Really? Like REALLY.

      Delete
    20. I flip flopped over it for months. But in the end I voted leave. The EU is a shitshow. We need to get out, even if it means some short term economic trouble.

      Delete
  5. Aldo and GWC? Right pair of headbenders you've got there, James

    ReplyDelete
    Replies
    1. We should fit right in then!

      Delete
    2. Kangaroo says
      @anonymous 1:22am
      Absolutely correct. Both of them are total card carrying British nutters.
      Scotland's wealth should benefit the people who live there, not some Westminster thief.
      As for GWC's misleading references to Germany, all you need to do is look at the list of wastemonster reserved powers as opposed to what the EU looks after and you will very quickly realise the vacuity of their arguments.

      Delete
  6. Ach! Young James when did you collect this bunch of ignorant , idiotic and unthinking commentators?

    They definitely require a bit of education, manners and maturity.

    ReplyDelete
  7. The Scottish sub-sample consists of 151 people and it is hence pretty meaningless to read much into it.

    ReplyDelete
    Replies
    1. Yes, which is exactly the point of looking at the average of the four post-election YouGov subsamples. That gives a more meaningful picture.

      Delete