Tuesday, March 31, 2015

Cracking ComRes poll of all Labour-held seats in Scotland shows gigantic 19% swing to the SNP

The headline pretty much contains all the information I have at the moment, but I'll update this post when more details are made available.  A 19% swing is (unbelievably) a touch lower than the Ashcroft constituency polls were generally showing, although that may simply be because the swing isn't quite as huge in affluent Labour-held areas that Ashcroft hasn't covered yet.  It would still be enough to cause absolute carnage.

If there was a uniform swing of 19% across the whole country, that would imply a national SNP lead of approximately 16% - exactly the same as suggested by the recent ICM poll.  However, you'd expect the Labour-SNP swing to be somewhat lower in seats that Labour aren't competitive in, so the fact that ComRes have only surveyed Labour-held seats would - on the face of it - indicate that they're picking up a slightly lower SNP lead than most other pollsters.

There's a big caveat here, though.  Most Scotland-wide polls are sensibly weighted by recalled vote from the last Holyrood election, and from the referendum.  One thing we'll have to look out for is whether ComRes have made the same mistake that Ashcroft did, and weighted their results to recalled vote from 2010 - a procedure that we know is wildly unreliable.  They may have felt they had no option but to do that, because they wouldn't have been able to match 2011 Holyrood vote recall to the correct boundaries.  But the reality is that if 2011 weighting isn't possible, it would be much better to simply dispense with past vote weighting altogether, because otherwise there's a severe risk of underestimating the SNP's lead.

UPDATE : According to the Daily Mail, these are the voting intention numbers in the ComRes poll.  The percentage changes are from the 2010 results in Labour-held seats.

SNP 43% (+24)
Labour 37% (-14)
Conservatives 13% (-1)
Greens 2% (+1)
UKIP 2% (+1)
Liberal Democrats 2% (-12)

Don't be startled by the seemingly 'narrow' lead - you'd see much the same thing if other polls were restricted to Labour-held seats.  The wonders of the first-past-the-post electoral system can easily translate these sorts of figures into something approaching a clean sweep of seats if the votes for the leading party are sufficiently evenly spread - and that's exactly what we suspect may be happening.  The recent ICM poll suggested that the swing was highest in the Labour heartland seats which are toughest for the SNP to win, and lower in No-voting seats where Labour are starting from a more vulnerable position, due to a split unionist vote.

ITV, who commissioned the poll, are suggesting that the SNP would take 28 Labour seats, while Labour would hold the remaining 12.  But that's based on a uniform swing, and almost certainly underestimates the SNP's potential gains.  Labour's hopes of achieving respectability in defeat depend on cutting the SNP's lead, not on getting their prayer-mat out and hoping for a uniform swing that simply isn't going to happen.

A crude look at the percentage change figures would suggest that the SNP are hoovering up almost all of the lost Labour votes, and almost all of the lost Liberal Democrat votes - ie. the SNP vote is up by far more than the Labour vote is down.  But it may not be quite as simple as that - the datasets will hopefully tell us if underlying movement from the Liberal Democrats to Labour is being disguised by the gargantuan swing from Labour to the SNP.  That's what appeared to happen in the 2011 election.

UPDATE II : I've now had a chance to look at the datasets, and they are truly bizarre.  ComRes have indeed committed the cardinal sin of weighting by recalled 2010 vote - but when they asked people how they voted in 2010, they didn't even give the SNP as an option!  The only options were Labour, the Conservatives, the Liberal Democrats, or "some other party".  That approach will take quite a bit of justifying, given that the SNP outpolled both the Tories and the Lib Dems in 2010.  (Even in 2010.)

I can think of two possible explanations for what ComRes have done - either a) stupidity, or b) it's a cunning plan to try to make people less confused over how they voted in 2010, ie. if you don't even offer the SNP as an option, people who voted Labour in 2010 but then switched to the SNP in 2011 might be more likely to answer the question accurately.  You'd have to be ultra-charitable to think that was the plan, but even if it was, it doesn't appear to have worked to any great extent.  Respondents who say they voted for "some other party" have been downweighted from 264 to 207, meaning that the SNP's position may well have been underestimated due to false recall.

At the very least, the use of 2010 weighting means that this poll is not directly comparable with any full-scale Scottish poll, and so the superficial appearance that it may be marginally less awful for Labour than the results we've seen from other firms is fairly meaningless.

The point I made above about where the Liberal Democrats' lost votes are going is borne out by the datasets - fractionally more Lib Dem voters from 2010 are planning to vote Labour than for the SNP, but that is being offset by a much bigger shift from Labour to the SNP than the headline numbers would suggest.

Perhaps the most devastating detail of the poll from Labour's point of view is that just 33% of respondents say that Labour is the party they most closely identify with, irrespective of their current voting intention.   The SNP are only just behind on that measure, on 31%.  Remember these numbers are only from seats that Labour won in 2010, in most cases by a massive margin.  It really does look as if a large number of ex-Labour voters are not merely "on holiday", but have undergone their own personal revolution, and as a result are not coming back any time soon.

50 comments:

  1. Here's a helpful graph:

    http://ukpollingreport.co.uk/wp-content/uploads/2015/02/Scotlandtargets.jpg

    A 19% uniform swing would imply Lab could hold about 10 seats, but as we know from Ashcroft the biggest swings are coming where the SNP need them most.

    How's that bounce coming along, Jim? Hearty lolz! :-)

    ReplyDelete
  2. I guess Jim Murphy's honeymoon is well and truly over LOL.

    ReplyDelete
  3. Wouldn't that cost a fortune to poll 40 seats individually? Or it a single poll covering the 40 seats with a small sample from each?

    ReplyDelete
    Replies
    1. I'm pretty sure it's the latter.

      Delete
    2. It will be. They've done that format before in a poll of English marginal seats.

      http://www.comres.co.uk/polls/itv-news-marginal-seats-poll/

      Strange thing is that a 19% swing is a relatively "good" score for Labour. Ashcroft generally found swings above 20%. Although this poll isn't as detailed (1,000 across 40 seats rather than 1,000 in each) and it will include some seats where you would expect the swing to be smaller, because there is a larger Tory / LD presence

      Delete
  4. James, I'm not sure there would be much lower swings in the non-Labour seats.

    Obviously where Labour have less vote share (i.e. rural seats) they will have less to lose, reducing the size of the swing. But this will be compensated for in most of those seats by the Liberal Democrats having more votes to lose (i.e. in the 11 they are defending) than they do in most of the Labour-held urban seats.

    ReplyDelete
  5. A quick calculation from me suggests this poll would look like national ones if extrapolated.

    ReplyDelete
  6. I wonder what would be the result if people were asked to recall both their 2010 and 2011 votes and if they had changed party in those two elections.

    ReplyDelete
  7. Didn't Ashcroft find the swing in the non-Labour seats he polled within a few points of the swing in the Labour seats? I believe that was the case.

    ReplyDelete
  8. Swings from Labour to the SNP, in those seats where the Tories and/or LibDems are traditionally stronger, may be dampened somewhat by tactical voting. In other words, Labour losses to the SNP in those seats may also be huge, but are compensated (partly) by ex Tory and LibDem voters intending to try and stop the SNP. Perhaps not, but it is a possibility.

    ReplyDelete
  9. We should also keep in mind that not all 2010 Labour seats had massive majorities. For some a 19 percent swing is not achievable. It could be as Ashcroft polling predicted, larger swings in the seats with massive majorities and smaller but still winning swings in those closer races. I am going to be most interested in reading any regional or seat by seat breakdowns if published.

    ReplyDelete
  10. That is a very good omen for 7th May if we are polling 19% ahead in Labour only seats. Some seats will have higher swings some lower but very encouraging indeed.

    ReplyDelete
    Replies
    1. Marcia : 19% is the swing, not the lead.

      Delete
    2. My enthusiasm is running away with me. :)

      Delete
  11. What does the seat distribution look like with these figures?

    ReplyDelete
  12. I see Ashcroft is releasing constituency polls tomorrow, though it seems only 8 and so I imagine none will be in Scotland unless anyone knows better? There are definitely at least a couple up here which would be of interest because no-one is sure how they are going, e.g. East Dumbarton.

    ReplyDelete
    Replies
    1. More than a couple - I'd like to see all the Lib Dem-held seats to see if there's a consistent pattern (but particularly Berwickshire, Roxburgh & Selkirk and Orkney & Shetland), plus a selection of the more affluent Labour-held seats.

      Delete
    2. He said that he would be polling seats that had already been polled and shown close results. Most likely it will be the Con / Lab marginals in England (poss Wales).

      Delete
  13. would like to see an ashcroft poll in mings ne fife seat

    ReplyDelete
  14. (From previous thread)

    TNS have fixed their data tables. Scottish sub-sample is SNP 43, Lab 23, Con 19, Others <7

    ReplyDelete
  15. SNP 43% (+24)
    Labour 37% (-14)

    The way to look at this is that this means in these 40 Labour seats, Labour scored 51% and the SNP scored 19% in 2010. That means the SNP were just below (1%) their national performance, while Labour were well above (8%) their national performance.

    If you crudely translate that into 2015 shares (adding 1 to the SNP and taking 8 off Labour), it would give SNP 44, Lab 29 nationally.

    ReplyDelete
  16. Swingback is on! Crossover soon? lol

    Mildly encouraging for us. I always expected a tad of recovery as the election neared, hope these are the first signs.

    Though as I've said before part of me is nervous-excitedly looking forward to the progressive shaking of the kaleidoscope that this election is promising.

    Especially now you lot are under Sturgeon, who shows signs of being genuinely progressive left of centre, unlike the corporatist, instinctively-Right Salmond.

    ReplyDelete
    Replies
    1. Wait a minute... there aren't any previous ComRes polls of Scotland to compare with, except the tiny cross-breaks of their GB polls. It's impossible to tell if the "lower" SNP lead (+15) is due to a recent movement or their methodology. Remember there have been a couple of polls this year that showed lower SNP leads (the first TNS poll and Panelbase).

      Even if there has been a genuine movement, it's no more than a few points in total. It's the difference between being almost totally wiped out and saving maybe 10 or so seats.

      Delete
  17. Wouldn't it be karma for Labour to have 6 seats, Tories, and Liberals one between them and the SNP 46?

    Have I counted right?

    ReplyDelete
    Replies
    1. No. That adds up to 54.

      Professor Colin Rallings (Plymouth Uni) estimates that this poll would result in Labour losing 28 of the 40 seats polled. Presumably they would also lose Falkirk, which they won in 2010 but was "lost" when Eric Joyce left Labour..

      Delete
  18. Appears to be 2010 weighted. Lab weighted up, Lib weighted up. Con down. SNP must be down too but they don't show numbers. Even with weighting the sample is too Labour heavy.

    Quite an incredible result given this.

    ReplyDelete
  19. One interesting point from the poll is that the Tories are almost unchanged from 2010. No tactical voting there then.

    ReplyDelete
  20. It has been noted that the poll is still using 2010 weighting and has downgraded SNP support by about 20%..

    I noted that as well. I am having trouble with the polling companies insistence in sticking with 2010 weighting in Scotland. While it may make sense in England to follow the pattern on weighting, using 2010 weighting in Scotland ignores the sea change since 2011 and especially the impact of the 2014 referendum and its results. The companies that have abandoned 2010 weighting or using a combination of weightings including 2011 Scottish election or the European election are showing a larger SNP support.

    That said, it would certainly be interesting if one of our resident experts were to recalculate the numbers without the 2010 weighting or with the more current elections.

    Is anyone up to the job?

    Thanks in advance.

    ReplyDelete
  21. This must be one of the strangest polls I've seen. How can you predict the result of 1 seat with only approx 25 people per seat? At least with Ashcroft he was doing 1000 people per constituency

    I'd be interested to know from those more statistically minded on what the margin of error is for the comres poll compared to the Ashcroft polls

    ReplyDelete
    Replies
    1. They're not trying to predict the results of each seat. The only breakdown of seats offered is between the size of Labour lead in 2010 (less than 30%, between 30% and 40%, or above 40%). The estimate by Professor Rallings will be based on those breakdown figures and how they apply to the various seats using uniform swing between 2010 and this poll.

      MOE will be almost exactly the same. Once you get up to around 1,000 in a sample it makes very little difference what size of population is surveyed. There's little or no difference between the polling MOE of a constituency (100,000 or less) or Great Britain (60,000,000+).

      Delete
    2. Presumably the Ashcroft constituency polling, with 1000 per seat is still a better indicator - unless things have moved since then.
      Also his results are extremely useful for SNP tactics.

      Delete
  22. The problem with tactical voting, is that Labour is having to throw a lot of insults at the Tory Party, so that won't be going down to well with their voters.

    We have also had the laughable fall-outs where Labour accused the guys who were organising the tactical voting site, of being secret Tories, because they were telling people to vote Tory in a Labour held seat.

    It seems to me that the Labour Party are all for tactical voting, as long as it benefits them, but any suggestion that they might lose seats in a ward, so the electorate should vote for another party, has them frothing. lol.

    It's also looking like the Tories will hold their vote, (as far as numbers go) while the Libs & Labour will see a huge drop in voter numbers, I would imagine that for Ruth Davidson, this will be seen as an endorsement of her time as leader, so she will be working hard to get the Tories to back her party, in whatever ward.

    I can see a similar thing happening with the Libs Dems, but for different reasons, ie, a loss of seats can be blamed on the Nick Clegg/Danny Alexander factor, but a crash on voting numbers, will be seen as his own failings.

    All in all, I can only see slight changes to some seats due to tactical voting by small numbers of voters, but not the orchestrated campaign that unionists were trying to create, when the polls were showing the first signs of an SNP surge.

    Labours biggest problem, is that they put all their eggs in one basket (pardon the pun) when they decided to make Jim Murphy the focal point of almost every Labour announcement/ TV appearance/ MSM press release.

    The fact that they did this, along with the repeated line from him and Kesia that 'the Labour Party just haven't been good enough' suggests strongly, that they thought the Labour Parties main problem, was the poor leadership of Johann Lamont, and that if they could get shot of her and put the 'smooth operator' Jim Murphy in her place, Labour would soon bounce back.

    Margaret Curran was so convinced that this would happen, that she was prepared to stab her life-long friend Johann in the back, to save her own seat.

    That's why it has been so important watching Jim Murphy's personal poll ratings plummet, as it is a clear indication that as people are listening to him, they are not liking what they hear, and more importantly they are not believing him.

    Keep it up Jim, you are doing a great job... For the SNP! :-)


    ReplyDelete
  23. This comment has been removed by the author.

    ReplyDelete
  24. I’m coming towards the conclusion that as Tory voters see SLab in a tail-spin they have to choose between the lesser of their 2 evils:

    1. Vote tactically to try to prevent an SNP landslide,
    2. Vote with their party of choice, suck it up and laugh wildly at Labour on May 8.

    I’m more and more inclined to believe that the latter option is becoming a real option for many Tory voters – they could potentially end up within a few percentage points of Labour across all Scotland.

    Ruth Davidson is a monumental pain sometimes but she has galvanised the Scottish Tories good and proper. Her Question Time performance the other week was a good example of her attitude; she had a couple of opportunities to really go on the offensive but she chose to be non-partisan and look at the problem as opposed to trying to apportion causality.

    The Scottish Tories might be more inclined to revel in Labour’s pain than to assist them. The rationale is fairly straightforward. Regardless of an SNP or a Labour MP being elected it will be an anti-Tory MP. Why sacrifice the party’s vote for no reward?

    ReplyDelete
    Replies
    1. Implicitly (if not explicitly) encouraging your supporters to vote tactically is the kind of thing you would do in a by-election where you have little to lose and may wish to embarrass a rival. Not in a general election.

      Delete
    2. I believe that Murphy and Rennie could learn a lot by referring to the conduct of Ruth Davidson. I would never vote for her in a million years but she actually comes across as rather honourable. I cannot deduce yet if she is actually so but she comes across as such. For the Scottish Tories she is just the ticket.

      Delete
    3. I live in a constituency where (broadly, taking into account slight geographical differences) the Tories were a decent second to Labour in 2010 and a decent second to the SNP in 2011. The Tory vote is rock solid in this constituency and I can't see them voting tactically when they may well be hoping to come through the middle with LD votes moving to them. Could be interesting.

      Delete
    4. I think polling breakdowns tend to show Scottish Conservative voters as being mainly elderly (ISTR a poll where they had <5% in the 16-24 subsample and >20% in the +65 subsample). This probably means their vote is more static in nature - not willing to change or vote tactically.

      Delete
  25. Replies
    1. Laziness, rather than duplicity. ComRes, who have never polled Scotland specifically before, have just applied their normal methodology for GB polls. Wouldn't normally make much if any difference to a GB poll but possibly does here.

      Delete
  26. If Labour were truly devoted to their core ideals and had just lost their way I would feel differently.
    They aren't, in fact the only devotion they show is to personal self enrichment and using their power to help enrich friends and family. All the while looking down on ordinary people with an air of entitlement.
    Remind me how eactly are they meant to be different from the Conservative Party?
    That is why these figures are just soo pleasing.

    ReplyDelete
  27. How can ComRes be sure they have a representative sample of Labour v SNP voters?

    The 2010 other group which is used for weighting includes SNP, Green, UKIP and is down weighted considerably. How are the voters assigned back to the various parties in that group.

    ReplyDelete
  28. I am perplexed why some people put much store in the 2010 general election results. As far as I can remember back the SNP only really put in a token effort. They saved their efforts for 2011, I think some SNP supporters were actually disappointed at the time. However, the landslide win the following year meant a referendum on independence. Given the massive campaign and the 45 per cent Yes vote, with Glasgow, Dundee and other SLAB areas supporting independence, not to mention the SNP surge in membership afterwards, I really find it difficult to take the 2010 results seriously.

    ReplyDelete
  29. Desperate measures from the MSM as they try to project the polling results from the 40 previously Labour seats (SNP 43, Labour 37, Con 13, LD 2, UKIP2, Green 2) to all of Scotland. Some thick pundit punched these numbers into the Scotland Votes and projected SNP 30, Labour 27, Con 1, LD 1.

    Murphy and Labour are now using that calculation as a sign of fightback.

    It is going to be a long 5 weeks as the MSM will use any trick to talk down the SNP and pump up Labour.

    ReplyDelete
    Replies
    1. "Some thick pundit punched these numbers into the Scotland Votes"

      Do you want to name and shame? For the 40 seats surveyed by ComRes, the UNS projection is SNP 28, Labour 12. Of the other 19 not covered by the poll, the only one Labour have a prayer of winning is East Dunbartonshire. And the SNP already hold six seats, so adding that onto to 28 gives them 34 straight away.

      Even making VERY conservatives assumptions about the SNP failing to make maximum gains in Lib Dem and Tory seats, this poll would point to something like - SNP 42, Labour 13, Liberal Democrats 2, Conservatives 2.

      Delete
    2. I think it was someone at the Press Association as both the Herald and the Scotsman have the exact same error in the body of their stories. They were either too stupid, or too careless or simply too Labour to correct the error.

      Delete
  30. For the record (pun not intended), Magnus Gardham wrote the story in the Herald and David Maddox wrote the story in the Scotsman. This probably explains everything.

    ReplyDelete
  31. It's not bad though is it? It is good for that the SNP activists do not get complacent. And remember those pundits and on May 8th tweet them saying 'I thought you said Labour would be on 27 seats'

    But if extrpolating from the 40 Labour seat VI still means that SNP wins the election that's not bad either/

    ReplyDelete
  32. I see they boycotted Falkirk in producing their dodgy stats. I guess that would have made things look even better for the SNP though. Missing out the SNP in the recall question as well - tut tut, naughty. The MSM is afflicting not just the BBC but the other side as well, STV has joined them in the rabid zombie attack on the SNP and its efforts to conjure up a Brit Nat Lab revival. Sleekit Unionists.

    ReplyDelete