Wednesday, May 6, 2015

SNP move into election day with 20% lead, says spiffing, splendiferous Survation survey

Well, so much for the YouGov poll at the weekend being the last before polling day - we've had no fewer than three in the last few hours, although it does seem likely that this Survation poll will be the last one until the exit poll at 10pm tomorrow.  It contains something of a curveball, because Survation have decided to headline the results of an additional voting intention question asked using a replica ballot paper, rather than the one asked in the normal way.  Consequently, the headline numbers below are not directly comparable with anything that has gone before.

Scottish voting intentions for tomorrow's UK general election (Survation, 3rd-6th May) :

SNP 45.9%
Labour 25.8%
Conservatives 15.0%
Liberal Democrats 7.1%
Greens 2.6%
UKIP 2.4%

In line with recent Survation polls, the standard voting intention question produced a somewhat bigger SNP lead than that -

SNP 48.9% (-2.3)
Labour 24.8% (-0.8)
Conservatives 15.5% (+1.2)
Liberal Democrats 5.9% (+0.5)
Greens 2.4% (+1.0)
UKIP 2.0% (n/c)

In reality, only the results of the standard question are much use to us at the moment.  There's no point trying to estimate exactly what the true SNP lead is, because different firms are producing very different numbers.  What we can try to nail down is the trend.  Two of the three polls today have suggested a slight narrowing of the SNP lead (on the standard question), while the third has suggested a slight increase.  The odd one out is Panelbase, and it's noticeable that they started their fieldwork three days earlier than YouGov and two days earlier than Survation, so I suppose it's not impossible that they missed an ultra-late swing to Labour that the other pollsters managed to detect.

The most that can be said is that there's no firm evidence of any movement to Labour, but if it has happened it must be very small.

The Record, who commissioned the Survation poll, are making a song and dance about how the ballot paper version of the question points to a marginally less overwhelming landslide for the SNP.  They suggest that this is partly because of people who would otherwise be SNP voters drifting off to 'local heroes' standing for the unionist parties, and partly because of anti-SNP tactical voting.  The snag, though, is that the YouGov poll also asked an additional question in an attempt to see if local factors made any difference, and found that the SNP lead actually increased.  Doubtless Survation would claim that their approach is superior because they name actual candidates, but nevertheless the evidence is obviously contradictory at this stage.

What does look increasingly likely is that the Liberal Democrat vote has strengthened as the campaign has drawn to a close.  The YouGov poll at the weekend suggested that the Lib Dems were doing significantly better than at any point since the referendum, and today's poll from the same firm confirms that result wasn't a fluke.  The Panelbase poll has the Lib Dems equalling their post-referendum record high, while the 6% in the standard question from Survation is not a record, but is still above average.  This development may pose a problem for the SNP in one or two of the tougher Lib Dem-held seats.


  1. Again, that's with using the "ballot paper" method.

    Standard VI is SNP on 49%.

    1. My inkling is to say that in actuality the results if the poll was correct would be somewhere in between the normal VI and this "ballot paper" method, which probably has more impact when looking on your screen than in the polling booth, where I get the impression people only really look for the party logo, vote and leave.

    2. Survation article explaining the "ballot paper" method.

    3. I think you are correct there.

    4. From the Daily record article, the standard VI appears to be

      SNP 49%, LAB 25%, CON 15%, LIB 7%, () ,SNP Lead +24%

      which makes a big difference.

  2. Night before a GE and the worst we're doing is 20% ahead.

    Boab, if you're reading this: you were maybe - MAYBE - right about the Yes Alliance, mate. ;-)

    1. Hahaha.
      I'll be honest, at this stage, I couldnae gie two hoots, as long as we are winning.

  3. This comment has been removed by the author.

  4. Really only a fairly minor boost for Libs and Labour when comparing the 'ballot paper' method to standing VI question. Does suggest that the supposedly large reserves of incumbent personal votes and tactical votes which these parties are pinning their hopes on are being overstated. ScotlandVotes predictor has the Libs down to 1 MP on both sets of figures.

  5. How does this new methodology deal with the scenario where a voter responds favourably to a ballot box prompt but actually doesn't vote have any intention of voting. It seems to me that it might encourage previous backers of a candidate to say they'll vote for him / here when perhaps they don't intend to vote and might have answered differently if they hadn't been faced with the choice. Parties face the problem mostly not of voters switching but not coming out. I wonder if this method doesn't encourage party supporters but non-voters to back a party which they won't leave the house to support.

  6. Hey James, think I've spotted either a typo or some sort of heinous sampling error in the datatabs - only 31 AB voters down out of over 1600:
    If it is a typo it's possible the lead should be higher as the SNP is on 52% on the first question among that group.

    1. I agree, this looks wrong. Their previous poll (link below) had 214 AB respondents out of a total sample of 1,015 (21%). This poll, as you say, only has 31 out of 1,660 (<2%). That doesn't make any sense.

      I was slightly surprised about you saying that the SNP do slightly better amongst this group than on average, but it was also true in the previous poll.

    2. Yep. SNP over 50% in a simple re-calc for straight VI. Looks like someone typed 31 instead of e.g. 310.

  7. Some interesting results on YouGov's predictions, the ones the BBC were using anyway. It seems like Labour are a whopping 3% ahead of the SNP in Edinburgh East:

    That seemed weird to me. It's one of the only "red" seats (i.e. leaning Labour), the others are Dunfermline, GNE, Rutherglen and Aberdeen South. Aberdeen South seems to be the only "leaning Labour" seat i.e. where Labour have a good chance of winning. I'm not surprised - isn't South Aberdeen the rich part?

    Edinburgh East is really disappointing though - it would be a shame if the other seats stayed Labour but Tommy Sheppard is our best candidate. I'd hate to see him lose, I've been looking forward to him doing well for months and months. How can it be, though? Considering Edinburgh East had such a high Yes vote...

    I'm greatly heartened by Berwick though - even though the Tories are ahead there's 0.2% difference between the Tories and the SNP there. It's incredible stuff. I suppose though in a Tory seat we needed to be a good bit ahead before we had a chance at winning - we have problems with turnout (as the referendum showed) but the Tories get out and vote come rain or shine.

    1. I also very much hope that Tommy wins; it's the Greens' target seat, though, and they have a relatively high-profile candidate in Peter McColl. They may be splitting the anti-Labour vote. I know they are perfectly entitled to stand, but I really wish they hadn't.

    2. I'm not convinced by the reliability of YouGov nowcast. To take Edinburgh East for example, the totals of the vote shares suggested do not add up to anything like 100%!

      Lab 28 - 36
      SNP 25 - 33
      Con 6 - 11
      Green 6 - 10
      LD 3 - 6
      Ukip 1 - 2

      Take the midpoint of those ranges and the total is only 84%. Even the top of the ranges given barely adds up to 100%.

    3. Different AnonymousMay 7, 2015 at 9:14 AM

      Well went and voted. As it happens I'm in Edinburgh East. As usual the parties put out placards (newspaper style) on the street in front of the entrance of the building that has the polling stations.

      Only Lab, Green, and SNP. Nay Tories or LibDem.

      Interestingly, the Labour one didn't seem to have any mention of the word 'Labour' at all. Lots of red, big picture of the candidate (Sheila Gilmore) dressed in red, candidates name, candidate's name next to box with X in it. The word Labour conspicuously absent. Maybe the small print?

      SNP was SNP/Tommy Sheppard, Green just green no candidate name.

      Hmmmm. Does anyone know whether 'don't mention the Labour party' is a general thing?

  8. 3 tremendous final polls for us.

    To think we'd be on average 20% clear of Labour is mindblowing really.

    Now it's up to the people. Get out and vote, folks and drag your pals, family and even enemies with you. Near every poll in the past, and figures have shown that older people do vote whilst younger folk don't (as much). We also can tell from tens of polls now that the SNP are popular, very popular with the younger populace of Scotland. I hope we encourage everyone to vote.

    I have no idea how tomorrow will go. I'm hoping for an excellent showing.

    55-59 - it'll be an earthquake if we poll that well, and take near on every seat. I'm expecting a bit of tactical voting in such as Edinburgh South and a few others to keep the SNP out.

    45-55 an excellent night, and a wonderful result.

    30-44 we'll have won our first ever UK election in Scotland. No doubt a few of the usual suspects in the media and the Unionist parties will paint this up as some form of disappointment. Will it heck! It won't be the massive numbers of polling, but it'll be an absolutely historic night for the SNP - who even could be the 3rd biggest party (terms of numbers) with this.

    20-29 again, we'll have tripled our MPs, whilst a good score, I reckon it'll be a bit disappointing.

    10-20 how can gains be a poor night? I think there is next to no chance of this happening, but still, a mass of tactical voting? I can't see it.

    I don't want to be over-confident either. I was a rollercoaster during the last months of the Indyref, and right up the 17th I thought we were going to do it - something changed on the 17th, I felt we just didn't have the numbers, we would be close, but would somehow just fall short and predicted 45-55 to my close circle. I am blowng my trumpet FAR TOO MUCH in this post, but , we will do very well tomorrow.

    All the best, and hope to see some posters through the night tomorrow. May even have a few drinks.

    1. I don't see how you guys could have a 'bad' night, at the end of the day you're going to make almost unheard of gains for a minor party, so it's a breakthrough result, no matter what.

      I suppose the only disappointment would be if the macro result doesn't work out for you. A better than expected LD result combined with a small breakout by either LAB or CON could mean a LD coalition with one of the big parties.

      But overall it's going to be a good night for you lot, so I hope you enjoy yourselves :)

  9. If someone ever needs cheering up, Foulkes is always there for the job.

    Today he tweeted this:
    "Pollsters are making even bigger fools of themselves this time and giving me plenty ammunition for my Private Members Bill"

    "@AngusMacNeilSNP Back tomorrow to celebrate our victory in UK & Western Isles"

    "The secret wish that Nicola confided in the French Ambassador could come true unless we get a big Labour vote in whole of UK tomorrow."

    "@AngusMacNeilSNP @NicolaSturgeon Very quick & very sensitive response from @AngusMacNeilSNP . @Alasdair4MP has him on the run"

    "The only Scottish constituency not visited by @NicolaSturgeon is the Western Isles. Seems they have given up on @AngusMacNeilSNP"

    Basically, he seems to be suggesting something is up with the polls, he also is still talking about frenchgate (even though the rest of Labour gave up on it). He also thinks Labour are going to make a gain on the Western Isles.

    1. Foulkes doesn't know shit. He's just trolling.

  10. Right now GOTV is a four letter word. I've spent several hours reorganising the shambolic dog's breakfast the Activate system makes of voters lists in a very rural constituency. If they gave us the sheets even a day earlier it would help.

    I've emailed all the rural leafleting teams with the addresses in their patch. I've sorted the 60 or so addresses in the village and wondered vaguely how we can get round half of them.

    I need sleep.

    But my mind is boggled. The very biggest poshest mansion in the area. A house whose drive I've been walking down with leaflets for years, feeling like a worm who is about to be escorted off the premises. A house the real old timer said was bound to be Tory, might as well drive on past. And I nearly did.

    On the list.

    1. Aha. Did some discreet checking. Mansion owner is indeed a big Tory. Biggest Tory donor in the county. Donated £100,000 recently. May have dispensed with his gardener's services because he discovered the gardener was SNP. (Yes, I asked the gardener!)

      Someone has been pranking the telephone canvassers. I wonder how often that happens. Local branches really have to scrutinise these lists.

  11. You know, if the LibDems are recovering a bit, it could let Michael Moore hang on. If we can't have BR&S, and I'm desperate for Calum to win, I'd far rather Moore held than the Tories got the seat!

  12. Assuming that a) the boo-boo re: lack of ABs doesn't ruin the poll and b) the "ballot paper"method is more reliable, the regional cross-breaks are interesting.

    Glasgow looks like a total SNP landslide - SNP 54, Lab 28, Others <8.

    As does NE Scotland (which was a better SNP area anyway) - SNP 54, Con 18, Lab 17, Lib 6. I suppose the Labour vote might be concentrated in Dundee / Aberdeen, but still.

    The other areas are more competitive. Willie Rennie is very excited about the Lib Dems being on 32.5% in the Highlands & Islands cross-break, but there are two problems with this. 1) They are defending every seat in this region except Western Isles, which has by far the smallest population. 2) The SNP are on 42.6%. If you were being optimistic about the Lib Dems you would say this could mean them holding onto O&S and one mainland seat, but there's no way (as Rennie claims) that they could hold all of them. It's not mathematically possible based on those percentages.

    Other competitive are Lothians, West Scotland and South of Scotland. Lothians has SNP 37, Lab 29, Con 18, Others <7. Bear in mind though that the SNP are coming from a very low 2010 base in pretty much all of these seats. Ashcroft gave the SNP leads in three Edinburgh seats with around (or just under) 40% of the vote, so a 37% cross-break is not particularly inconsistent with this. I think though that there is a pretty good chance that Lab will defend some of their seats here (East Lothian and Edin South being the likeliest).

    West Scotland has SNP 43, Lab 32, Con 14, Others <4.

    South (much smaller sample) has SNP 37, Con 30, Lab 23, Others <5. Again, this is more or less what you would expect - the Lib Dems have melted away completely in the south outside of Moore's seat. Then the other seats are some sort of combination of SNP, Con and/or Lab. Probably the most unpredictable region as a whole.

  13. Good luck everyone, and thanks to James Kelly for his fantastic work over the past months. What will happen will now happen. There will be some disappointments today, and also some great surprises. Such is the nature of the beast. The important things is that this is another battle, another big step on the way. The war goes on, and if we stay the course, eventually, we will win.

  14. James

    Western Isles and Moray

  15. Well, shall be leaving for work shortly via the polling station. Being in Moore's seat, it's a tough gig locally, but you never know.

    Have a good day all and see you on the other side.

  16. James "I think though that there is a pretty good chance that Lab will defend some of their seats here (East Lothian and Edin South being the likeliest)"

    East Lothian will be close, which is unreal given a 12k majority. The East Lothian SNP have put in a fantastic amount of fingers crossed.

    Ps, big thanks to James Kelly for this blog.


  17. Yeah just wanna add my voice to those thankin James for this fab blog....I bet I'm not the only one who looks at it first before moving on to Wings, Bella, the National etc (won't bother listing sundry msm sources that I look at to see what rubbish they're spewing...with a couple of honourable exceptions).
    So thanks tonnes Mr Kelly, you make a difference

  18. I to would like to thank James, it fantastic to see the many sites run by individuals giving us a different news stream. Thats evolution for you.

  19. Thank you very much James for all your effor. Don't be thinking that this is time to retire though!

  20. Another big thank you from me James for all your effort. Your blog has been a first port of call for me throughout the campaign. ;)

  21. Adding my thanks for your very informative posts

  22. What is the SgP final-final poll of polls?

  23. Really interesting read at on what it's like to be at the other end of the phone, targets and so on.

  24. I went to the polling station in my sleepy little NE Fife village this morning (what a gorgeous day!) to find an enterprising SNP person had whacked a massive great "Vote Gethins" sign on the lamp post in prime position. Lib Dems and Tories just managed to squeeze theirs underneath (they well much smaller...). Labour never bothered and in fact they are the only party that has not canvassed here at all, which says it all really. They have given up. The Lib Dem candidate comes from my village and is a very congenial (if rather dull) chap but I'd be gutted if he managed to bring it home here, though they have been trying very hard in fairness.

    But would it be a bad thing if some Labour/Tory people vote/drift to Lib Dems? I am relaxed about that as I feel the independence issue will be resolved ultimately by a change in course of the Labour party. It saddens me to say it but in the short to medium term we cannot have independence or certainly at least a very adequate federal arrangement without them.

    Fun times today though. I will literally be cheering the roof off if Mhairi Black outs wee Douglas Alexander. In an ideal world i'd have went down to Paisley and volunteered all week but too much stuff to do here. The champagne however is very much on ice. Will J F Murphy survive? I suspect maybe just but I don't mind because the more we have of him the better for the SNP.

    Oh the polls. I predict the polls are way out on SNP vote. I expect them to hit just over 50% minimum with not less than 50 seats. Tactical voting is a chimera. I had a scan through the leaflets by the door I have been collecting and every party is telling me to vote for them to keep the SNP out. I dare say in constituencies from Stranraer to Shetland it is the same. It is disorganised chaos and will probably help the SNP more than hinder them.

    Whatever happens, however, the Scottish press will somehow spin the result as some kind of Labour success, a little like the way we Brits have been fond of re-writing the absolute military catastrophe and humiliating retreat from Dunkirk into some kind of victory.

  25. Ashcroft UK Scottish subsample (bigger than normal at 180 unw base):

    54% SNP
    24% SNP
    2% Green
    1% UKIP

    1. What a mess that was. Ahem:
      54% SNP
      24% Lab
      11% Con
      6% Lib
      2% Green
      1% UKIP

    2. ICM and Mori sub-samples were also good (SNP >50%)

      Populus had a lower SNP score and lead 42-30

      After incorporating last night's and this morning's polls, Election Forecast bumped up the SNP projection from 51 to 53, so they must think they are good polls (or at least okay).

    3. Populus (n=353): SNP 42, Lab 30, Con 13, LD 7

      Mori (n=98): SNP 51, Lab 20, Con 20, LD 6

      ICM (n=177): SNP 53, Lab 22, Con 13, LD 5

  26. Could a small swing to Moore in Berwick, Roxburgh and Selkirk not possibly help Calum Kerr for the SNP. The last polling I saw had Toryboy Lamont just ahead of Calum, with Moore some way behind. If some votes go from Toryboy to Moore could that not help Calum edge it??

    1. That's what I wondered, but in reality Moore is very close behind Calum and I don't think a swing to Moore would actually help him.

      Here in DC&T the reason we're doing comparatively well is that our LibDem vote collapsed after Catriona got skelped by the voters in 2010, and it's collapsed SNP-wards. Calum needs the same thing to happen in BR&S to pull properly ahead, and the opposite may be happening.

      I'm desperate for Calum to win, but to be honesst if he doesn't I'd rather Moore held than the Tory got in. Here in DC&T we're trying to create Tory-free Scotland and we don't need one sneaking in under the wire in the next constituency.

    2. Rather Moore than Toryboy, but everything's crossed for Calum right now!!

  27. James - there's definitely an issue with Survation's final Scotland poll. Does hurt SNP from my re-calcs. Has anyone contacted Survation?

    Not that it really matters now!

  28. My final prediction on percentages:

    SNP - 49%

    SLAB - 28%

    SCON - 14%

    SLIB - 7%

    Minor parties - ~3%

    SNP will pocket the 45% from the referendum and get a few others from squeeezed Greens, SSP, unaligned lefties.

    SLAB seem to have a bedrock of about 24% and they will gain a couple of points due to incumbency, differential turnout and sheer force of habit.

    SCON may suffer from tactical voting from their own supporters switching to SLIB where they think it may make a difference, it is also Ms Davidson's first real test as leader (she was merely one of the choir during the referendum.) She may find that the great praise she gets in the Tory press in England does not translate too well across the border.

    SLIB will claw back a few points from the absolute tanking they were taking in the polls a few weeks ago, due to tactical voting and grassroots loyalty.

    Minor parties will be utterly squeezed out this time around - a shame, but they have next year to make a real mark.

    Ps. How would my guesswork translate into seats?

  29. MORI Subset
    50% SNP
    20% Con
    20% Lab
    5% Lib
    1% Green

    ICM final subset:
    53% SNP
    22% Lab
    13% Con
    3% UKIP
    1% Green

  30. Getting a bit worried to be honest - social media is positive but remember last time, if Twitter decided things Scotland would be independent.
    Hearing that Labour is more likely to get 20-25 seats, fighting SNP for first party. Labour pulling out all the stops. I think SNP might get 1 seat in Edinburgh but that's it.

    1. where did you hear that?

    2. No offense, but your really going to have to expand significantly on your post to get any kind of credibility with me. You say you are 'hearing' things, can you tell us what and from whom?

    3. I'm not saying Anon is necessarily a concern troll, but given the circumstances I think we'd be forgiven for having a suspicion. There were a number of posts of that sort on referendum day.

    4. James, sounds like garbage to me.