Survation had a new Scottish poll out for the Daily Record this morning. It showed the same sort of surge in SNP support that we’ve seen in other recent Scottish polls from Ipsos MORI, YouGov and Panelbase – in this case Westminster voting intentions are CON 17%, LAB 24%, LDEM 6%, SNP 46%, UKIP 5% (tabs are here.) I don’t imagine uniform swing calculators are really any sort of guide to how things would work out in a re-alignment of this sort of huge scale, but on paper these figures would give the SNP 52 seats in Scotland and Labour just five, and in practice it would surely produce a huge number of SNP gains. The question remains whether Labour can mount a recovery in Scotland prior to the election once they have elected a new leader, or whether this SNP surge will be maintained.

This afternoon there was also some reporting of a new Opinium poll (tabs here). Opinium don’t seem to have officially released voting intention figures, but they are provided as crossbreaks on a new poll, so we can see that the VI figures would have been CON 34%, LAB 33%, LDEM 5%, UKIP 18%, GRN 4%. This would be the first Tory lead from Opinium since the Omnishambles budget, and the lowest any poll has shown the Lib Dems so far this Parliament.

291 Responses to “Survation Scottish poll and a Tory lead from Opinium”

1 2 3 4 5 6
  1. Phil Haines

    “it would take a lot of contradictory evidence to make me reassess the conclusion that I drew.”

    Of that, I have no doubt at all! :-)

  2. I take your point that there’s still a long time to go, and so the current SNP support might not translate into huge numbers of seats, but you can’t stop folk speculating about coalition combinations in the meantime!

    I’m not – I’m saying that it’s partisan to the point of preventing worthwhile discussion of polling to say that anyone who says that the SNP won’t get 55 seats shouldn’t be taken seriously.

    To take your moon analogy, there’s about as much chance of Labour ending up with four seats as there is of Jim Sillars becoming Scotland’s 60th MP by claiming the Moon to be part of Scotland and then being elected unopposed. That said, Labour losing a significant number of their 2010 total of 41 seats is not in doubt.

  3. SNP will get a conference bounce. Now Salmond is no longer First Minister people seem to love him and Nicola will get a honeymoon. Labour will also get a bounce when they announce their leader.

    Labour will run with the Vote SNP get the Tories slogan (in fact they have already started putting out leaflets with that on them).

    Murphy is a clever politician so I am sure he will find numerous ways to wrong foot the SNP. Findlay is more left wing but he will not stand up to hard interviews as shown on Marr (which was not a hard interview) when he made up Labour policy on the spot. Boysack might remove some of the toxicity. However, Murphy is likely to win so we can assume political games from him which will likely claw some support back.

    So my guess is SNP will end up between 25-30 seats which would be a great achievement and possibly make them 3rd largest party.

  4. Boyack NOT Boysack – sorry Sarah

  5. Re: predicting “swing-back” etc

    In the past calculating seats on current polls was worth doing, because the PM could call an election at any time and a seat calculation was a clue to whether that might happen. Now, after the fixed term Coalition fix, we know when the election will be, so I can’t see much point in turning vote share into seat.

    For what it’s worth, have been predicting a hung parliament with no two party coalition possible for some time now, based, as far as I can tell, on the idea of reversion to a mean. The SNP seats currently predicted (29) matches Allan Christies guess for 2015, which is down on a high of 35. On that evidence the SNP might be expected to play a lead role in negotiations. Ukip, on the other hand, will be too small to matter with 4 seats. This sounds a reasonable guess to me.

    On the other hand it has, like the other modellers’ predictions, the assumption of linearity built into it. I don’t feel that is where we are now. I feel politics is more chaotic than that, now moreso than ever. That is, a small event before the election could shift the election out of the twilight zone of coalition and into overall majority, or even landslide.

  6. I am labouring the point a bit because the polls were remarkably accurate in particular YouGov.and I want to give credit where credit is due

    That’s a different matter. The overall trend of the polls was accurate, in the same way that the overall picture in Scottish GE polling is accurate.

    What’s in question is the absolute values (i.e. the SNP margin over second place)), given that despite getting the direction of travel correct, when it came to the margin between outcomes Indyref polls were consistently wrong in the same direction every time.

  7. COUPER2802


  8. @OldNat

    Try playing the ball.

  9. Tax credits have soared to £30,000,000,000?
    That can’t be right surely.
    So we taxpayers are subsidising employers so they can pay low wages. Some economic turn-around.

    And Shevi, it’s fewer seats.
    Sorry sad I know; the girl can’t help it.

  10. @Chrishornet

    I do not share your assumption that the SNP will win the popular vote next May.Suspect it will end up something like Lab 35 – SNP 30.


    “Scottish National Party activists on this very site are utterly desperate to say that the two things are completely unrelated (and post in tandem to give the impression that this is a consensus view). They then go on to proudly boast that according to GE VI polling, five in six Yes voters intend to vote SNP”

    “I’ll naively attempt to lower the temperature here, knowing that I will no doubt be subjected to thinly veiled bigotry for the above comments”

    And then he waffles on and on and on. You’re utter hatred towards the SNP is there for all to see.

    Your grandstanding is incredible. Insult fellow posters as bigots for commenting on current polling.

    Look I understand why you’re upset at the recent polling in Scotland, if i were you then I would be spitting venom out as well but please..please don’t ever label me as an SNP activist because I so happen to comment on Scottish Westminster Labour’s dire poll ratings.

    Ok I’m done with you, you’re a nasty piece of work.

  12. ChrisHornet

    I base my skepticism on the sheer extent to which the SNP will finish ahead of Labour in vote share (because there’s no doubt at all that they will) on the fact that, averaging a poll a day between debate between the Yes and No leaders and polling day, EVERY SINGLE ONE OF TWENTY GOT THE YES-NO MARGIN WRONG IN FAVOUR OF YES.

    But not by much – only a couple of points. No one is claiming that the percentages in May will be exactly this or that. But if you look at the average of the last four Scottish polls (SNP 47%, Lab 26%, Con 14%, Lib Dem 5%), knocking a couple of points off the SNP will give Labour back some
    seats, but the SNP will still be on 48. It’s not going to make much difference with those sort of numbers.

    In any case all polls only give public opinion as at the time. In referendum campaigns we know that those who are undecided till the last minute tend to opt for the status quo position. So the polls may have been pretty accurate and unable to pick up those last few percent of people.

    And you could just as easily argue that the all the polls in the Holyrood election of 2011 underestimated the SNP on the regional vote and so the SNP should be expected to get more than they currently poll. It’s just not a very strong argument.

    I do happen to agree that the SNP are on course for 20 seats or more. But whether we are talking in the UK-wide context of a likely hung parliament, or a Scotland-wide context of there being a total of 59 seats, the difference between 20-25 and 50-55 is enormous.

    On the contrary the difference is surprisingly small. If you play about with Anthony’s advanced swingometer, the SNP only get 20 seats on 37% (Lab 36%, others as above). Take the figures up to SNP 47%, Lab 26% and the SNP have 53 seats for an additional swing of just 5%. That’s the way First Past The Post works – it’s actually quite hard for the SNP to get 20-40 seats.

    Of course there may well be local variations. Personal or tactical votes may help MPs such as Kennedy, Moore and Mundell to hang on, though equally some of those remaining Labour seats might also go to the SNP as they were stronger Yes areas.


    @”Tax credits have soared to £30,000,000,000?
    That can’t be right surely.”

    Correct.-that isn’t right:-

    Tax Credits £ bn ( 2014/15 prices)
    2010/11 £31.1 bn
    2014/15 £28.4 bn

    These are DWP numbers , quoted in a recent report by IFS, who say that the above reduction is 2.7% in real terms-but compares with a 4.6% reduction in announced cuts.

    IFS attribute the difference to ” slower-than-expected earnings growth “


    I agree with you when looking at seat projections etc you need to look at a number of factors.

    My workings on the seat projections was just a generic example on how you need to factor in the Scottish VI when allocating seats over all in the UK to parties.

  15. @ChrisHornet

    The poll’s absolute values were VERY ACCURATE most polls were coming in at 52-48 in the week running up to Sept 18th apart from one outlier which had a very small sample and wasnt taken seriously. The final polls were picked up the late swing to No with the final Survation 53-47 and then the final YouGov 54-46.

    But if you are taking about polls in July\August where YouGov and Survation\Panelbase were very different. ( think YouGov were correct although we will never know).

    So I actually finally understand your point. A few months out some polls may have over-estimated Yes and therefore some polls may now be over-estimating SNP – Could be valid and the polls are quite different with a span of about 10% but because SNP is high people arent picking up the variation. But even if we take the lower end the SNP would still be on 40-50 seats.

  16. As far as indyref polls showing a bias for yes, do opinion polls for most referendums not overestimate the change option, as undecided voters usually back the status quo when the time comes? I doubt that Scottish polling for elections can be directly compared.


    Change “These are DWP numbers , quoted in a recent report by IFS, who say that the above reduction is 2.7% in real terms-but compares with a 4.6% reduction in announced cuts.”

    to” “These are DWP numbers , quoted in a recent report by IFS, who say that the above reduction is £2.7bn in real terms-but compares with a £4.6bn reduction in announced cuts.”

    my typo.


    Your last post is spot on and explains much better the point I was trying to put across.

    “In any case all polls only give public opinion as at the time. In referendum campaigns we know that those who are undecided till the last minute tend to opt for the status quo position. So the polls may have been pretty accurate and unable to pick up those last few percent of people”

    Can I also add that we have never had a Scottish indy ref before so polling was always going to be fluid right up until polling day.

    Also I don’t think the current SNP rampage will hold up anything like the 55 seats the polls are suggesting come 2015 (22 I predict) but current polling is all we can go on now and for some reason ChrisHornet has taken a flakey to my comments on the latest Daily Record poll!!

    Who know’s what will happen in 6 months time but the polls today are what they are…snapshots of real time opinion and parties will have to act accordingly to them.

  19. Are these MP’s about to defect after tomorrow’s by-election?

    For Conservaitvies

    Mark Pritchard MP for The Wrekin

    Chris Kelly MP for Dudley South

    Brian Binley MP for Northampton South

    For Labour

    Graham Stringer MP for Blackley and Broughton

    Does anyone have more info?

  20. Roger Mexico

    “Allan it doesn’t work like that. You need to use a specialised seat calculator with the ability to enter separate Scottish and GB figures such as Anthony’s advance swingometer or Electoral Calculus’s regional model.”

    I agree its not an easy calculation which is why earlier in the thread I pointed out to SHEVII that the University Group do take due allowance for recent movements in Scotland and based on the most recent polls they forecast the Tories with a very small lead in both votes and seats.


    I don’t quite go along with your reasoning. On the basis of what happened in 2010 Labour needs to be 7 – 10% ahead in order to win by 7% next May – but present polls might suffice to produce a neck and neck result in terms of vote share.

  22. If Stringer does decide to go there will be a lot of people waving him off; the bunting will be out!

  23. TOH

    While academics can be as wrong as the rest of us, at least their modelling is something that they base on evidence, and constantly adjust to get as close as possible to the actual result.

    I wouldn’t dismiss their predictions, just because I would prefer that the Tories weren’t the largest party!

  24. So is this Labour’s new strategy…Never mind the polls in Scotland because the pollsters got it wrong with the indy ref (They didn’t actually towards the end)

    It’s a big gamble but when town halls across Scotland are being packed to the rafters every week by new SNP members and their membership not far off 100k then I think it’s pretty naive for some to put such a large emphasis on polling prior to a indy ref v current and projected VI for a UK 2015 GE.

    The dynamics have utterly changed in Scotland as we saw in 2011. The SNP won 55 FPTP seats on 45%, Labour won 15 on 30 something % and that was on the back of Labour winning 41 out of the 59 Scottish seats just months before!!

  25. OLDNAT

    I’m not suggesting that the academic forecasters are correct, they may or may not be, but I for one am very interested to see how correct they prove to be. They are at least trying to take full account of the current situation in Scotland.

  26. ToH

    Their modelling for Scotland seems a bit rough and ready at the moment. Without any constituency polls, they can’t be anything else.

    Roll on Ashcroft!

    I have been impressed by the way they have adjusted modelling in England to deal with the new phenomenon of UKIP – the area where the largest party will mainly be determined.

  27. OLDNAT

    Agree with both your comments, they are clearer than Fisher on how they are adjusting their model which is helpful.

  28. Okay latest Yougov CON 32% LAB 34% LIB 7% UKIP 15% GRN 6%

    Cross break fro Scotland CON 15% LAB 25% LIB 8% SNP 41%

    That gives us (Scotland ) seats SNP 43 LAB 12 LIB 3 TORY 1

    Just to make the damn thing work (has to add up to 90%) for prediction I gave Labour an extra 1%

    So taking into account the Scottish seat tally and the national VI for the UK how would this translate into Westminster seats UK wide?

  29. Actually I’m being too generous by giving Labour an extra 1% because it gives them two extra seats but what’s a couple of seats amongst friends. It’s coming up for the season of goodwill.

  30. The Consumer Council for Water today published stats on strange things found in sewers.

    It turns out fat and false teeth are the most commonly found odd items.

    It looks suspiciously like someone is dissolving elderly people and disposing of them in our sewers.

    Could this explain weak Tory VI?

  31. ALEC

    Do you think you should get out more ?


  32. @Valerie

    Colin is correct about the figures from the IFS. I posted the report on here


    George O is £15bn out in his social security cuts.

    The problems according to the IFS are due to

    more spending on pensions
    faster rising rents
    low wage growth
    delays in changing the disability living allowance

    Allister Heath tends to be cavalier with his figures, as befits a supporter of high finance, but he was making a larger point about subsidising jobs. his solutions would be to end all in work benefits.

    I dread to think what the consequences would be for unemployment and the economy, but then subsidising jobs causes problems too.

  33. For those planning to visit our northern climes after 5 December might want to note that from then, the Scottish Parliament has decided that the drink driving limit will reduce to European levels.

  34. @chrishornet

    You may want to note that our friend Allan Christie is that brave and rare beast a self confessed Tory supporter in Scotland!

  35. ‘Allan Christie

    It gives UK wide figures of

    Con 255
    Lab 312
    Lib 18
    SNP 43
    Oth 22

    Lab/Lib majority of 14 allowing for SF and Speaker. Slightly improved by PC and SDLP support to about 25.

  36. @Colin – “Do you think you should get out more ?”

    It’s funny you should say that, as it’s almost word for word what I asked my parole officer the other day.

  37. Alec’s on good form today. :-)

  38. What is the purpose of these “qualities” questions that the Sun gets YG to ask?

    Tories think Miliband and Clegg have no good qualities

    Labs think Cameron and Clegg have no good qualities


  39. GRAHAM
    I am simply saying IMO, things will not jog along like this for the next 6 months. However, if they do Labour, for widely known reasons will be largest party, (probably).

    You may want to note that our friend Allan Christie is that brave and rare beast a self confessed Tory supporter in Scotland!

    Actually think I might become a kipper.

  41. The apparent increase in Tax Credits from 2000 to 2010 is mostly due to the fact that the system changed in 2003 and so the totals involved aren’t comparable.

    Prior to April 2003 Working Families Tax Credit was a continuation of Family Credit which had run from 1986 to 1999. Working Families Tax Credit was just a rebrand of this with a bit of tweaking to allowances.

    From April 2013 Child Tax Credit and Working Tax Credit were introduced. The Working Tax Credit took on elements of the old Working Families Tax Credit but also made it available to low earnings individuals without families.

    Child Tax Credit took on elements of Income Support to support Children. Prior to Child Tax Credit low income parents would claim income support and an element of this was calculated to include allowances for their children.

    Once Child Tax Credit was introduced these elements of Income Support which had been allocated to support children were reallocated as Child Tax Credit to continue to support children.

    So any comparison of figures should take into account this change. Taking the amount of Income Support formerly paid for children into account pre 2003 would most likely account for a large proportion of the difference.

  42. JOHNB160
    ‘Allan Christie
    It gives UK wide figures of
    Con 255
    Lab 312
    Lib 18
    SNP 43
    Oth 22
    Lab/Lib majority of 14 allowing for SF and Speaker. Slightly improved by PC and SDLP support to about 25

    Thanks for that. Gives a clearer picture on the overall seat projection on the basis of today’s Yougov including the Scottish cross break seat projection.

  43. Sorry typo – should read 2003 not 2013 above.

  44. Allan Christie

    “Actually think I might become a kipper.”

    With your English heritage, shouldn’t you become a bloater instead?

  45. @Gazprom

    Thanks for that, always nice to have someone with more knowledge contributing. I like that

    also the £28.5bn is a forecast for GB, the forecast for the UK seems to be £29.5bn

    LINK to outturn and forecast budget 2014

  46. ALEC


  47. SNP Activist

    Ok I’m done with you, you’re a nasty piece of work.

    Way to take the moral high ground.

    Oldnat, I thought that in the previous Holyrood parliament the Scottish government position was that there should be a variable allowance to take into account the isolated nature of pubs in the (geographical) majority of the country? Or was that Labour?

  48. I think when they lose to tomorrow, the Cons are going to blame the candidate.

    some evidence of that is starting here.

  49. ChrisHornet

    I hadn’t heard of anyone suggesting a variation in the alcohol limit.

  50. Survation4 minutes ago

    Stockton South Current VI (with chg since 2010 general election): CON 39% (0), LAB 37% (-1), LD 3% (-12), UKIP 18% (+15), OTHERS 3% (-2)

    Something happening in Stockton?

1 2 3 4 5 6