A new ComRes poll for Tuesday’s Independent has headline voting intention figures, with changes from last month, of CON 40% (-1), LAB 27% (-6), LDEM 18%(+2) and Others 14%.

The 13 point Conservative lead is the largest recorded for almost twenty years; in the unlikely event that such a whopping swing occured in a uniform fashion at a general election it would produce a Tory majority of 58. It is worth remembering however that ComRes do tend to have the most favourable weighting for the Conservatives (and for ‘others’) so we should expect them to report larger leads than companies like YouGov and ICM.

The changes in the topline figures match the general trends we’ve seen elsewhere. Labour are down sharply, the Liberal Democrats are up, benefiting either from the publicity of their leadership election or through Labour’s misfortune, the Conservative vote is down very slightly – the same as we saw with BPIX, rather than the sharp drop we saw with ICM. My instinct is that the ICM poll may be a rogue, that the actual picture is that the Conservatives are steady-ish, Labour are down significantly but the Lib Dems (and others) are the beneficaries of their troubles – that is just my own personal judgement through. We’ve still got YouGov to come this week, Populus early next week and an Ipsos MORI poll at some unspecified point, so we shouldn’t have a lack of polls to make judgements on.

108 Responses to “Tories 13 points ahead in ComRes poll”

1 2 3
  1. RodCrosby

    No Rod NBeale is not talking tosh. Why do you seem to assume that there will be a uniform swing for a start? There are also more marginals than you think once you reverse past tactical voting or perhaps you can’t accept that it can work both ways?. And we had a hung parliament and a near hung parliament three times in the 1960’s and 1970’s but not since then so there is no trend in that sense towards another one. History is not in your favour. Whilst it is true that the number of non Lab-Con MP’s has greatly increased in the last 30 years a counter factor so far as the Conservatives’ task is concerned has to be the reduction in the number of unwinnable Scottish seats at the last GE. If Rod you believe that the Lib Dems will hold all their 60 odd seats at the next GE then yes you are right the chances of a hung parliament increase immeasurably but you did not say that.And although I am reluctant to make predictions- of one thing I am certain-the Lib Dems will lose a lot of seats in the south to the Conservatives at the next GE just as they did in the 20007 local elections.A hung parliamnet is certainly a realistic prospect but it is not odds on.

  2. I wonder how much lower Labour can go? 27% is only 3.5% higher than their worst ever rating (by Gallup) in December 1981.
    The Tories’ highest ever rating was 56% in May 1968, (again by Gallup).

  3. Nick Keene. More tosh, I’m afraid. I am not assuming a uniform swing. I am assuming a randomly-distributed swing with a standard deviation of about 3.0 about the mean, which for practical purposes a uniform swing is a good approximation. An even better approximation may be the probabilistic method used by John Curtice and David Firth to forecast the 2005 election, which I have also explored.

    I can accept in theory that “tactical voting” may reverse, but there has been little sign of it so far. Until I see evidence to the contrary I must assume it isn’t happening.

    History is bunk. The reason we haven’t had HPs is because the votes margin of victory since 1979 has been unusually large for both Labour and Con. The underlying increasing probability of a HP has been masked by the decisive vote leads Lab and Con have happened to enjoy over each other, and latterly by the bias in the system towards Labour.

    As for the LibDems, you can hypothesize whatever you like about their number of seats at the next GE. I happen to think they now have a baseline of about 45 seats, but what is more, any losses are likely to be partially offset by Nationalist gains, leading to only a modest reduction third-party seats, if at all. So a HP is still highly likely.

  4. John T
    The boudary arguments were done on the basis of parties making representations and the independent Commission deciding. It was quite gentlemanly until Campbell ensured an army of lawyers turned up for Labour catching everyone else on the hop.
    Niave? Yes, to expect New Labour not to squeeze ……STOP! Non partisan site….

  5. That should be ‘boundary changes’ not boudary???

  6. The Tories did reach I think 55-30 in November 1976, and that could have been Gallup.

  7. I am fed up with statements on this blog such as Mike Richardson’s ‘there’s no way back for Labour now’ as I don’t think that recent events have been so momentous as to ‘fix’ public opinion for the next two years. This blog is best when people engage in serious analysis.

  8. Having said :-“If Cameron has any sense he will try to make people like NewsElephant more aware of real differences with Labour.By which I mean the role of central government, it’s limitations, and it’s alternatives.”….I am delighted to read that Cameron has begun to do just that in his CBI speech.

    This is fundamental political philosophy on which Cameron differs from Brown ( as distinct from Blair !). It needs to be discussed by both sides much more.

  9. Colin,

    ” the role of central government, it’s limitations, and it’s alternatives”

    The problem with this is that talk of restricting government can spook exactly the type of people that Cameron needs to win over.

    Redwood talking about to much regulation just weeks before the regulators failed over Northern Rock hurt the Tories ability to hit Labour when it was vulnerable.

    Praising private enterprise when large parts of the public think the tax status of Private Equity is a scandal wouldn’t be the smartest thing to do.

    Currently a better strategy for Cameron is just to concentrate on “Labour is Crap”, and “It’s time for Change”.

    It’s not smart to play to his own supporters when he needs Labour’s and if he emphasising ideological differences he risks pushing Labour supporters back towards Brown.

    So far Cameron isn’t hurting Labour as much as they are hurting themselves, a good reason for Tories not to get to excited and indeed one reason why they aren’t rising as fast as Labour are falling.

    A bit like IHT at the conference, you really only wheel out key policies when the time is right, whether that be if you are in trouble or when it will make a difference, like near a definite election.

    There may be a time for Clear Blue Water,and if Cameron is true to form might even be green, but this isn’t it. While Labour are shooting at their own feet Cameron can keep his gun holstered.


  10. RodC: people just don’t behave like that. If the country decides, with a substantial lead (say 5% or more) that they want Labour out then people will vote Labour out. There is enough collective understanding of how elections work in those who vote (and who canvass) for this to happen. If you want to be technical, you just have to assume that the “random” variation in the swing is strongly correlated with the labour majority that needs to be overturned.

    The Boundary Commission is independent but works within parameters which strongly favour Labour. This is partly because it has to use wildly out-of-date census data and because it is not supposed to make big changes in the number of seats in a region. Thus the North and esp Northern Cities are grossly over-represented. The Isle of Wight (C) has 110k electors. By contrast the average seat in Wales has 55k electors, it’s 65k in Scotland and 70k in England.

  11. NBeale,

    “The Isle of Wight has 110k electors”.

    “Na h-Eileanan an lar” has only 20k, electors, Why?, because it’s a isolated seat, islands.

    Part of the boundary commissions role is to create constituencies that make logical sense to the people in them.

    Like it or not the “Labour” safe seats reflect a reality on the ground, both demographic and geographic.

    Not liking it doesn’t make it rigged.


  12. NBeale: never happened and never will. Only about 20% of people even know who their MP is, never mind what the majority is, or what that “means”, or how they could exploit it, or what anyone else will do…. At best, tactical voting will add 1% to the swing in the marginals – worth about 10 seats – as in 1997.

  13. As a test of the bias, i put these results into the seat calculator, only switching the Labour and Lib Dem numbers. It really proves how pathetically biased the system is:
    Con 40% 395 seats
    Lab 18% 134 seats
    LD 27% 91 seats
    I then tested how much of a lead libdems would need to beat labour on seats, keeping the tories at 40% and others at 15%. The libdems needed 30%, double the labour number, to overtake labour on seats. pathetic.

    On a significantly more relevant note, its not surprising at all that the Tories aren’t rising. If for nothing else, its because Cameron isn’t doing anything, at least as far as the front page is concerned. As ive said before, hes waiting for the right time to strike, and when he does, then the Tories will continue to rise.

    The other possibility is that Cameron isn’t actually concerned about getting any higher. Attacking the government at times like this has it’s own perils, and this far from a likely election, the reward(better polling #’s) is certainly not worth the risk.

    Running the seat predictors is all very well, but as long as the Tories have any kind of lead (which is looking like an increasing long term probability) the next election is years away. Holding on 40% is very good during the interim. So long as he retains his poll lead, Cameron doesn’t have too much of a need to widen the gap until the election draws closer.

  14. I think the boundarties clearly favour labour- and instanvces like the Isle of Wight and the overrepresentation of Northern cities needs to stop. But that is only one of a pheotha of reasons why the system favours labour- some of them actually serve to confirm that things are not, in fact, as unfair as they appear. I believe there’s an excellent article about it somewhere on this site.

    Counting any chickens is madness, There was a poll the other day that put the parties level, now this with a 13% Con lead. This is a uniquely erratic period on the political weather of this country in which anythng could happen. I do think though – even though I am always very concerned with keeping my positings balanced and non-partisan (my favorite thing about this site) – that the Government is messing up its PR something truly terrible at the moment. They are almost spared the full impact of some of the sh*t simply because there is so much other sh*t that there isn’t room in the newspapers/ time for it on the news! The situation is recoverable- bu they need to get their act together on the PR front fast.

  15. Luke, I read the article on the bias a while ago, it really is a good one, and anyone who hasnt seen it should check it out. Just click election guide in the side bar, and scroll down. On the side bar there an election bias button right above the swingometer.

  16. I am officially never relying on a seat calculator every again. I gave each party 1% and other 97%. The result was
    Con 196 seats
    Lab 247
    LD 102
    Others 87

    obviously the calculators predictions are weighted on the basis that the supporters of minor parties are spread very thinly, but when it gets these results at such extremes, the formula must be flawed. While these results are less likely than the USA voting in a competent president, i do believe a similar flaw may exist with the Tory equation, affecting results when the Tories have a substantial lead

  17. Thanks oliver (& Luke) I had suspected the word “rigged” was inappropriate when applied to boundaries.

    As for “armies of lawyers” sent by Campbell to argue for Labour, that would compromise the indepedence of the Commission. Happy to be proved naive, Sally C, but I’d need to see the proof.

  18. Sally C – I’m not sure where Alistair Campbell comes into this. Labour organised a far more effective approach to the boundary review that came into force in 1997, with centralised responses to provisional recommendations. The Conservatives didn’t, and in some cases you had different local Conservative associations arguing against each other and safe conservative seats trying to bolster their own position rather than argue for an arrangement that was best for the Conservatives overall. The result was a set of boundary changes in 1997 that were far better for Labour than they would otherwise have been.

    In the recent set of boundary charges both of the main parties had a centralised, organised system of responding, with Roger Pratt working full time on it for the Conservative party (and by all accounts doing quite well at it) hence the Labour and Conservative representations probably pretty much cancelled each other out.

    NBeale – the lag on the population figures the boundary commission use does indeed help the Labour party. There is no sort of regional number of seats rule, the same quota is applied throughout England (Scotland and Wales have separate quotas, meaning they do indeed have more representation, though in Scotland this has mostly been dealt with). While there is inertia in the system – the Commission will try and minimize change – this does not outweigh the allocation of seats according to the quota. If cities in the North are still over-represented, its because their population is still shrinking, not because of any sort of regional quota.

  19. Well Rod you sure have a novel way of looking at things! I will confine myself to the following observations
    1 In my constituency for example anti Labour tactical voting took place in 2005 which resulted in the Labour majority being reduced to just 400 votes.
    2 You obviously follow these matters more closely than I and so you will be aware that Curtice and Firth recently revised their thinking following the patterns and trends they indentified in the 2007 local elections.
    3 History to you may be bunk but it is often said that those who do not learn from history are doomed to repeat it.
    4 Given that the Tories are almost unrepresented in Scotland it matters not one jot to the prospect of any possible Tory majority if the Nationalists manage to win any seats from either Labour or the Lib Dems.
    5 The fact that there is a bias towards Labour in the electoral system and long has been is a gimme. In 1974 they got more seats than the Tories despite polling half a million fewer votes but that did not stop the latter winning handsomely in 1979 with ‘only’ a 7% lead-just about where they are now.
    Nobody on this site -save good old Mark (Richardson)-is deluding themselves that it is a straight run from here to a Tory majority but to rule one out as you appear to do is well…tosh!!

  20. Peter Cairns

    Yes I agree that Cameron can bide his time just

    I don’t agree that exposure of the flaw in believing that The State can both regulate and manage everything in our lives is counterproductive.
    Cameron’s CBI speech contained a litany of the wasted cost & uselessness of the Quangotocracy we have now.The reaction from CBI has been very positive.
    The policy announcement on schools-including opening up the supplyside-has met with howls from Balls about taking control away from Local Authorities.
    IDS’ work on social breakdown and it’s solutions is predicated on the voluntary sector.

    These approaches go to the heart of difference in political philosophy between Labour and Conservative.They will ( I hope!) be inherant in policy as it is developed.They cannot be hidden-and I don’t think they should be.

    Yes “smaller government” has in the past left sections of society exposed to the rougher edges of the market. But Cameron understands that more clearly than anyone-his CBI speech specifically told his audience-who he avoided last year) that he would be looking for their participation in matters of social & community welfare.IDS-(traditionaly portayed as a right winger) has produced work on social deprivation which any (traditional) socialist would be hard pressed to disown.

    The biggest problem which I would concede is taking this philosophy in to those areas of the UK where large parts of the population are reliant on The State for their income.But it can be done & I am sure Cameron has the confidence now to do it.That is where the battle ground will be.

    Ultimately the Conservatives have to make people believe that there is a better way.
    I abandoned them in 1997 -firstly because I was sick of the pantomime apology for a government they had become-but ultimately because I believed Blair offered a better way. ( my wife never lets me forget it !!)

  21. I find it odd that You Gov do have not had a panel survey since 1st November. Where are the questions about immigration figures, HMRC data files and party funding?

  22. Mike,
    We must be watching different channels. From what you say he has been out and about possilbly in your neck of the woods?… but not in mine. It is not a critisism. As I said before, you have got to vary the bowling. He has been on speaking engagements and is on his way to the US soon. Since “Parki” much of it is not mainstream national tea-time viewing [or I must be out a lot].

    Re:boundary changes .
    I am afraid I believe Alistair Campbell had a terribly malign effect on British politics and one for which the Labour Party may be paying a price. I do not think he did anything to enhance the view of politicians. However, you are quite correct that whilst he may personify in my mind a cynicism I detest, he was not directly responsible in this case and therefore, I apologise. It only wish he would do the same.
    With regard to the Commission themselves, they are a professional body who is supposed to get it right whatever the standard of representation put before them. Good Judges know exactly what points should be made and when they have been badly made because they have studied the case. The Court of Appeal does not say, “Well you got it wrong Judge but it is not your fault, the barrister was pants.” They are there to get the decision right in the interests of the voter, not be buffeted about by partisan concerns and I am afraid that when you look at the country at large, there is plenty of evidence to suggest that the boundary commission have not succeeded as well as they might.

  23. Sally C, you’re making a coherent case for dis-allowing any representations by the parties – I hadn’t known that such representations were taken so seriously, so I’m glad you raised that.

    Appeal Court judges – do they hear appeals if Boundary commission decisions are challenged? If so, I’m sure they’d consider the merits of the case, as put by the legal representatives/barristers, so you wouldn’t get any reduction in partisan buffetting. Appeal judges quite often find that the original judge “mis-directed himself”, presumably quite often as a result of being swayed by the skill of one or other barrister.

    A sense of “fairness” is a recurrent theme in Cameron’s approach – I wonder if more light will be shed on perceived discrepancies between constituencies.

  24. Colin – don’t let your wife give you a hard time – Blair DID offer a better way. Whether he provided it is hotly disputed!

  25. John
    I wound’t want to advocate no representations. They should not live in an ivory tower where no one can shout at them.
    Judges, when appealled on the law have usually ‘misunderstood’ the law. Often it arises where the law is in a state of flux and the interpretation of the law has not been settled upon, or where a direction/law has been badly drafted has had unforseen consequences.
    Contrary to what is often reported in the press, Judges are usually good quality barristers who will have read the papers in the case and will be able to anticipate anything that another good quality barrister might say.
    And therein lies the rub.
    Any ‘democratic’ decision making body relies on the skill and independence of its decision makers. Without it, poor decisions are made which undermines the publics faith in the system. That is what happened re.1997

  26. Sorry to disagree on such a fine point, but top barristers are paid very well to swing a judge to the view of their client. You seem to be suggesting that judges don’t need barristers, or at least don’t need “special” ones to set the case out. Former solicitors sit as district court judges, by the way, and give poorly prepared barristers a hard time.

    It might well be true that the decision makers of the Boundary Commission make poor decisions, but if they do so because of undue interference by representatives of the parties, then those partisan representations should stop, in the interests of democracy, and the only influence should come from “experts” in the study of the relevant subjects.

  27. Peter,

    Na h-Eileanan an lar could easily be combined with some of the other Scottish islands and still have a ‘logical sense to the people in them’. How York outer fits that bill escapes me.

    John T,

    If, as our host suggests, ‘Out of the ten smallest seats, 9 are held by Labour and 1 by the Liberal Democrats’ whilst ‘out of the 10 largest English seats, 7 are Conservative, 3 are Lib Dem’ it is not odd that people smell a rat.

  28. Ralph – I entirely agree that we need perhaps more publicity for the machinations of the B.C., but I can’t quite believe that those outcomes happen because of a corrupt process.

  29. The boundary commission do a solid and even handed job (though they make useless decisions on seat names sometimes). It’s just that some of the details within the rules they follow do, over time, lead to biases towards Labour. The utilisation of historic population figures rather than projected ones and the presumption in favour of minimum change are both rational and defensible provisions. If all else were equal (and quite possibly anyway) they would almost certainly be better than the alternative. Given the pattern of population movement though they build systemic advantage to Labour into the system.

  30. Ralph,

    The combined population of all the Scottish Islands is less than the Isle of Wight, and that’s from the shetlands down to Aran.

    From Barra to Unst is (according to Goggle maps) 490 miles and 18 hours travel. London to Dundee is only 470 and 8 hours.

    Thus in trying to fit in with natural communities effectively putting people before politicains is a thankless task for the BC.


  31. John T
    “don’t let your wife give you a hard time – Blair DID offer a better way. Whether he provided it is hotly disputed!”

    I’ll get her to speak to you some time!
    I certainly didn’t expect it all to morph into Our Friends in the North. (why is it always the North East?)
    I will not be surprised if Gordon starts wearing a Gannex coat and smoking a pipe!

  32. Peter,

    When I lived on Benbecula the people of Barra saw those who lived on Lewis as alien as those who lived on the Mainland.

    A Hebrides seat including the present Na h-Eileanan an Iar and the Inner Hebrides would produce a seat with a population of well over forty thousand (as compared to 26,000 odd).

    Your leader in 2005 got more people voting for him than all those who voted in Na h-Eileanan an Iar, but got just over 50% of the vote.

  33. I do not believe for one moment that the Boundary Commission is corrupt but I do wonder-and perhaps Anthony Wells can answer this one-if they are underfunded and poorly resourced. If not why is it that they appear-but Anthony I stand corrected by you-to be about 10 years behind the times? That is to say that at the next election it is claimed in some quarters that the BC will only be remedying anomalies evident at the 1997/2001 elections but not later. Surely in this computer age they should have the kit to enable them to keep the boundaries up to date by changing them in between elections as they deem fit. And so long as they have the right of appeal the political parties should not be able to drag out the representation process beyond a prescribed timescale.
    There is no excuse however for the BC tolerating such blatant differences as exist between the number of voters in some seats and contributors are right to single out the Western Isles and the Isle of Wight especially as the former has had its own MSP at Holyrood since 1999 and thus retains its own electoral indentity.
    Clearly for its own sake the next Tory government will want to review the whole questionn of electoral boundaries but can any political party be trusted not to feather its own nest? In 1979 for example Jim Callaghan disgracefully ordered his parliamentary troops to vote down the new boundaries already approved by the BC because the changes favoured the opposition. This is a thorny issue.

  34. Ralph,

    40,000 population isn’t 40,000 electorate, and it still doesn’t get over the fact that it would be an enormous areas with nothing linking people other than the fact that they live on Islands.

    Again do you divide up the poulation to give national balance, or do you build it up from existing communities.


  35. John T
    Expensive barristers are usually paid to sway juries, not judges.
    And yes solicitors do sit as District Judges and Crown Court Judges for that matter but they can only spot a bad barrister or solicitor because they know what they are doing wrong.

  36. Boundary Committee rules can be found on their web site(s) – there is more than one committee.
    I primise not to give chapter and verse, but there is an interesting difference in principle between parliamentaries and council, that for parliamentaries they committee works on existing electorate only, whereas for councils, they rely on projections. Population drift from tending-to-labour urban areas to tending-to-tory shires is why parliamentary boundaries tend to favour labour.

  37. Sally C
    I don’t agree. very expensive barristers appear in all manner of appeal cases and are indeed paid to convince the judge of the merits of their clients’ case. The whole purpose of an adversarial system is to achieve justice by reasoned argument, and judges cannot do so without the highest quality representation of all sides of the case before them. There are relatively few “bad barristers”, but I don’t understand why uyou seem to be blaming them anyway

  38. >

    Whomever said this is being unfair. The boundary committees have a heirarchy of rules to follow. Having seen the process through three times, I can say that on each occasion they have followed the heirarchy. One of the results was pants for my party, but “thems the rules.”.
    As regards party representation – well there is always scope for arguing the case with a barrister, but at the same time, the barrister cannot only win that which he/she can argue to be within the rules.

  39. Oh Grief, I have missed out the thing that I felt was unfair. For absence of doubt, it is the following I think to be unfair:

    “there is plenty of evidence to suggest that the boundary commission have not succeeded as well as they might.”

  40. Peter,

    Na h-Eileanan an lar is far too small a seat so the options are either to add it in with something else without much thought or try and create a seat with at least some linking features.

  41. Ralph,

    Your right in the sense that like the Isle of Wight, it works as a unit, but is out of line with the norm. I am not a big supporter of it as such but it highlights that you need to balance between natural communities and uniform size.

    What we tend to get is a mix of historic and geographic. There is the odd situation with Fife, which has it’s own historic constituencies. If you were to look it in terms of travel to work zones, north and south have more in common and could be joined in to Dundee and Edinburgh.

    That might make electoral sense as that’s where people work, but the fifers would hate it.

    One of the key points made in the recent Gould report about the fiasco at the Holyrood elections was that “the electorate were treated as an afterthought”. I think that should be a warning for us all.

    The BC tries to minimise change and to build around natural constituencies even if that means the sizes vary. I wouldn’t want it to give an inch to parties trying to alter the boundaries for electoral gain.


  42. Looking at the ComRes figures in more detail a few things struck me on page 34.

    Firstly Labour only seems to be ahead of the Tories in two areas. 18-24 voting, and in Scotland (although in Scotland they trail the SNP so it’s hardly good news). The 18-24 vote is more than 2 to 1 which would be great if you believe that even after filtering they would all come out and vote. It is slightly odd given Cameron’s big push at portraying himself s “the Future”.

    The over 65’s figure is more than two to one the other way which must be more worrying. Regionally out with Scotland they appear to be doing worse than the national average everywhere but the North of England, where they are still trailing.

    The LibDems are above their average from 18-54 year olds, and seem to be doing particularly well in the midlands and to a lesser extent the North.

    Of some concern must be the fact that they are behind in the South East and in the South West where they have been historically strong. This is it continues might make them somewhat defensive in focusing more on holding existing seats that winning new ones.

    It would be good if the election calculator could be set for each regional page so that you could put in different scores for the regions;

    North-East, North-West, Yorkshire, East Midlands, West Midlands, South-West, South-East, Eastern, London, Wales, Scotland. IF it could tie in with the YouGov or other regional breaks it would be ideal.


  43. John T
    I don’t think I blamed Barristers for anything and frankly don’t see how you can interpret my comments that way! The adversorial system is, as you say, based on reasoned argument. However good judges are perfectly able to take into account the quality of the representation they have in front of them and understand when a point is being made particularly well or unusually badly. Clearly the advocates have to be up to the job, especially where those cases are comlpicated. I am not suggesting my cat could do it. A representative has to be competent enough to present the case before them [where the case is not self evident from the papers] and that bar may be high. I merely make the point that most judges are not to be “flanneled” and are able to distinguish between the merit of the argument and the skills of those debating it and will not ignore a valid point just because it is not well made.
    Incidently, if you visit the Court Of Appeal[!!] or speak to barristers who do, I am sure you will be aware that in the vast majority of cases the Judges give coded messages, eg.”We have read the papers.” Translated, “We know what we are doing…so keep it brief, very brief”.
    I think we are splitting hairs here as well as getting off the point of this blog. Anyone would think you just like an argument? Like Barristers DO??? Oh! Have I offended you by suggesting that Judges have often got the measure of flashy counsel? Are you in fact, flashy counsel? Flashy solicitor perhaps!
    I bear you no ill will if you are!

  44. Nor me you, and I’m not telling!
    I agree we’re off the point, and possibly even agree that the systems in our democracy should be seen to be fair.

  45. Peter , I have commented elsewhere on the fact that the LibDems best area for both Comres and ICM latest polls was the Midlands . I am sure this is just a quirk of non weighted sub samples , the Midlands was at the last GE and in most polls since then the LibDems weakest area and logically it should be so . I seriously think that a regional swing analysis of the sort you suggesy will give misleading results which are statistically invalid .

  46. At what point did the voting system start to penalise the Tories? Not in 1983.
    Which seats would disappear in order to give a closer balance between votes and seats.?

  47. Can’t wait for the next POLLS after the latest revelations on Labour Party funding . I wonder how corrupt individuals or a government has to be in this country in the last 10 years – previously by telling a small lie in court 14 years prior got you into jail as well as accepting a free meal in France – Oh how times have changed – or is it the judiciary is pro Labour ??

  48. Perhaps i’m being unkind – maybe what we are seeing is in fact Brown’s election cancellation “VISION” for Britain ?

  49. Both in your report and that in “The Independent”, “Others” were given as 14%, but not further analysed.

    I clicked on your ComRes link and see that they put UKIP on 2%, Green on 3%, SNP on 3% (30% in Scotland), Plaid on 1% (7% in Wales) and BNP on 1%. I make this 12%; but there are presumably rounding errors.

    If “Others” figures are going to be so high, reports of polls are going to have to be more accurate about them.

    There are two points that need to be addressed with “Others”. Firstly, a margin of error of about 3% may be alright when repeatedly polling intentions to vote for the major parties. But in relation to minor parties this degree of uncertainty ranges from finding nobody voting for the party at all to the party saving its deposits. This is particular trued for a poll like this ComRes one which seems to have a rather small sample size and to be collected by a comparatively cheap method which the pollsters hope to correct by adjustment.

    Secondly, the standard question of voting intentions does not consider the seat in which the respondent expects to vote. We are now faced with a number of parties which will stand in only some seats. And people tend to know that in a particular seat UKIP are likely to stand, not not BNP, or whatever. This is likely to affect their responses as, again for instance, a disaffected left-winger may state an intention to vote Green where there is a candidate in place, but say Labour, Liberal Democrat or “Don’t Know” where this is not the case.

    The situation is further complicated because:-
    1. Minor party candidates may only be selected shortly before an election, causing issues with the logitudinal comparison of polls.
    2. Minor parties are likely to stand differentially in setas held by one major party rather than the other two.
    3. Minor parties may be less likely to stand in marginal seats where it is difficult to pick up a protest vote.
    4. Voters may be less likely to vote for minor parties in marginal seats.
    5. The voting strengths of minor parties, and not just SNP and Plaid, varies considerably between regions.

    It is not my main point, but an implication of the previous paragraph is that I suggest that Others would, in the notional election tomorrow, get considerably more than 14% in those seats in which they stand.

    If the Greens are on 3%, if voters know which seats they are likely to be standing in, it looks likely there is going to be a widespread issue (as indeed I have found myself commenting for several individual seats like Canterbury) as to whether the Green deposit will be saved. And in some safe seats (safe from the point of view of one of the largest three parties) where the Greens are building strongly, such as York Central, they may be seriously challenging the third placed major party candidate. This could cause tactical voting on a considerable scale. I have made this point in relation to the Greens, but similar points could be made in relation to UKIP.

    I think that if the Other vote goes on looking likely to be large at the next General Election, which seems likely for several reasons other than polling figures, it would be advisable to do three things in relation to use of polling results. Firstly, poll users, including newspapers and TV, should require larger sampling sizes, even if to offset costs this means commissioning polls less frequently. Secondly, the question about voting intentions should be followed by a question (or questions) as to what parties the respondent expects to be standing where they will be voting. Thirdly, and particularly, I think that there should be much more use of polls of particular (presumablably mostly marginal) constituencies rather than nationally and of regionally aggregated data.

    Potential voters were a far more homogenous population from which to sample in the 1990s, let alone the 1950s. This may be bad news for polling orgnaizations and their cutomers, but it is how the world is.

  50. Cllr Peter Cairns
    I have put together a swingometer based on Anthony’s data
    [You will need WinRar to unpack the Excel spreadsheet

    It models several adjustments to the “Uniform Swing”…
    1. Regional swing patterns based on Elections since 1992. Also included is a notional “1997 neg” pattern, which is the Tories doing well exactly where Labour did well in 1997.
    2. LibDem incumbency. Based on elections where the LibDems fell back, they tended to perform significantly better in seats they held, especially first-time incumbents. However they also suffer where an incumbent stands down. An incumbency factor of “0” is this average historical pattern, “+” is more, “-” less. I’d stick with “0”
    3. A probabilistic forecast can be made, which takes into consideration the variation in the distribution of the marginals, similar to the method employed to forecast the ’66’ majority in 2005.
    4. Tactical voting between the three main parties. “0” means the TV stays the same, “+” is more, “-” is less. I’d be cautious about anything more than +/-1.
    5. Minor adjustments for Sinn Fein abstention, SDLP-take-Labour-whip, etc..
    6. Nationalist rise. Effectively increments the SNP vote from its 2005 baseline. I am working on a way to show this more effectively. The SNP scored 17.7% in 2005, so if they score 30% next time, that would be a Nationalist Rise of about 12-13%, and so forth.

    Please enjoy, and keep an eye out for updates….

1 2 3