Tuesday, November 19, 2019

Latest YouGov average suggests the SNP are on course for twelve seat gains

It's truly astonishing - we're three weeks into this campaign and we still haven't had a full-scale Scottish poll (apart from an out-of-date one from YouGov).  Until that changes, I'm going to keep posting averages from the five most recent Scottish subsamples conducted by YouGov, because that's the best available substitute.  Unlike Scottish subsamples from other firms, YouGov's appear to be correctly structured and weighted - which means the only problem with them is the large margin of error caused by the small sample size.

YouGov subsample average:

SNP 43.2%
Conservatives 25.6%
Labour 12.6%
Liberal Democrats 11.2%
Brexit Party 3.8%
Greens 3.4%

The most discernible trend over the course of the campaign has been a Tory recovery, but that's been at the expense of the Brexit Party and possibly the Liberal Democrats, rather than the SNP.  It's actually quite impressive how well the SNP vote is holding up at a time when their main opponent is gaining ground.  But the big concern is that the rigged TV debates could lead to a repeat of the 2017 scenario where the Tory surge didn't seem to be the end of the world until the SNP suddenly had to fight on a second front due to the late Labour comeback.

Note that the Tory recovery isn't completely factored in to the average yet, because two of the five subsamples were conducted before Farage decided not to contest Tory seats.  In today's subsample, the Tories are on 28% of the vote, which almost takes them back to where they were in 2017.  But because the SNP vote is higher than it was two years ago, a uniform swing would still result in the Tories losing seats to the SNP.  The seats projection from the average, based on the Electoral Calculus model, is -

SNP 47 (+12), Conservatives 7 (-6), Liberal Democrats 4 (n/c), Labour 1 (-6)

Incidentally, today's YouGov poll bucks the recent trend across the polling industry by showing Labour eating into the GB-wide Tory lead a little.  But that may be just a reversion to the mean, because the last poll was particularly favourable for the Tories, putting them on 45% - which would be the highest share of the vote for any party in a UK general election since the 1970s.

*  *  *

Click here for a handy list of SNP election crowdfunders.


  1. Unionists are more likely to vote down Yes/No lines and lend their vote to the Tories. Nationalists are more likely to vote down left/right lines and lend their vote to Labour.

    The results of this were seen in 2017. They will likely be seen again in 2019.

    1. It's a bit early to jump to that conclusion. There may at some point in the campaign be a Labour recovery fuelled by Yes voters, but it certainly hasn't happened yet.

    2. I dont think that's true for loads of reasons.

    3. I'm afraid that's a heap of wishful thinking there Labour's chance of recovery in Scotland is to say the least non existent

    4. It's not non-existent at all. Today's subsample mirrors ones from the 2017 campaign. Both Tories and SNP are at similar levels. All it needs is for Corbyn to dominate the agenda and he could easily drain the SNP suddenly as he did in that election.

  2. Before the campaign kicks off properly, my gut feeling at this stage is the SNP will make seat gains but probably just to the low 40s.

    It's looking likely now the Tories will have a similar vote share to last time. Whether there will be swings against them in more remain seats and towards them in more leave ones underneath it though is something we don't know. If that's the case, I suspect they will hold their seats in the borders and north east easily, although Ayr, East Renfrewshire, Ochil and Stirling could be SNP gains. None of them are anywhere near certain though and I wouldn't be surprised if they held onto 12 of their 13 seats to be honest.

    As for Labour, the SNP seem to have built up a big enough lead than could still see the party gain those 6 seats back lost in 2017, even if Labour do make a recovery. Just a 3% swing is all that is needed, especially with the Lib Dems up a little.

    I don't believe Swinson is any danger whatsoever, she'll get half the vote in her seat, the Lib Dems will hold their other 3 and gain North East Fife with little difficulty. Unionists are getting excited about the possibility that Ian Blackford could lose his seat, however given the Lib Dems are 19% behind and in 3rd place it would be some feat, especially if the SNP vote goes up just a little. I think he'll be alright.

    1. Calum a couple of points
      First of all, if you look at Swinson's vote in East Dunbartonshire, then over the four elections she has stood in there, she has never got anywhere near half the vote. In fact her vote is quite stable - a low of 18551 (38.7%) in 2010 and a high of 21023 in 2017 (40.3%). That vote might reflect a measure of tactical voting, but the fact is that it's less than John Nicholson's vote in 2015, which suggests there is an SNP vote there. Whether it was a one-off/ particular to 2015 is hard to say, but I do tend to agree with Kezia Dugdale (Politics Show on BBC2 last week) that the SNP wont make the mistake of not getting the vote out as they did in 2017 (most SNP candidates had a lower vote).
      The other problem for Swinson is that a WASPI woman is standing. She wont get elected, but she might just take enough votes off Swinson - who was up to her neck in this - to make it closer than the Lib Dems would like.
      As for the North East, I rather think on balance you are right. However, it is becoming clearer that the Tories would sell out the fishing industry in order to secure a future trade deal with the EU - indeed the EU have been clear this is, for them, a "red line".
      Re NE Fife - Gethins problem (as above re Dugdale) was a failure to get the vote out. His vote declined by almost 5000 votes, but the Lib Dem vote actually fell on 2015 by about 400. So as long as his problem was that the vote stayed home in 2017 and greater efforts are made to get it out this time, I reckon he'll be ok.
      But I do hope Gethins can hang on, as he is one of the most admirable SNP MPs, imo.

  3. What would be particularly useful would be a record of these averages (if they arent going to produce regular Scotland only polls). The sub-samples if they are the best we can get might tell us more if as well as looking at the numbers we could see the trend over a period of time. Just a thought

  4. My own view is that there are no full Scottish samples to date as the polling companies appear to be taking the 'unitary British state' view and that only the overall UK result matters.

    Given the Tory view (Scotland doesn't exist) and the Labour view (Scotland doesn't need to exist) is being reflected and endorsed by the broadcasters exclusion of the (massively) largest party by popular vote, seats and membership in this country (Scotland) from the UK-wide debates I am not surprised.

  5. My guess is that the result will depend a lot on whether SNP voters will actually vote or stay at home. Especially if the SNP's manifesto turns out to be rather lukewarm and underwhelming with regard to getting a second independence referendum (if the UK won't play ball), I expect it'll be hard to get out the vote.

  6. That makes Scotland the least Socialist part of the UK. Buckinghamshire is more left-wing by orders of magnitude. The most popular party here seems to be run by corporate lobbyists.

    1. The UK should break up then, allowing English socialism to flourish, free from the hard right anti-migrant, pro-brexit, Tory voting Scots.

  7. If we had the same turnout as last time, then the Tories have very little chance of doing better than in 2017. Everything suggests they maxed out the pro-UK + pro-brexit vote then. They lost part of that to Farage, and have recently regained most of it. Like a mini England in Scotland.

    So if the SNP got exactly the same share as last time, they might even pick up a seat or so from e.g. Lab losses to Lib Dem; the former having lost to the latter somewhat net (and to the SNP).

    However, polling is consistent with the SNP up at least 5 points (mainly from Lab net we assume), and Yes is up the same. The two are likely related.

    Each SNP point increase sees additional unionist seats falling. The key will be turnout. A decent turnout of SNP voters and the seat gains will be big, even if the share is only into the low 40's.

  8. Not Ayr but Ayr,Carrick and Cumnock