ComRes have a poll of marginal seats out tonight covering the forty most marginal seats with Labour and Conservative in first and second place (so 25 with Tory incumbents, 15 with Labour incumbents). Collectively the vote in these seats was CON 37%, LAB 37%, LDEM 18%, UKIP 3% at the last general election. In today’s poll ComRes found current support of CON 33%(-4), LAB 35%(-2), LDEM 8%(-10), UKIP 17%(+14). That’s a swing from Conservative to Labour of just one point, far lower than the swing shown in ComRes’s GB polls (it would be the equivalent of a national poll showing a Conservative lead of five points) suggesting Labour are doing worse in key marginals than in the country as a whole.

ComRes also break down the figures for the Conservative held and Labour held seats (though given the sample was only 1000 to begin with, caveats about sample size obviously kick in here). In the Tory held seats the Conservatives have a lead of 2 points (no change since the election), in the Labour held seats Labour have an 8 point lead (up 6 since the election). What that means is if these figures were repeated at the general election none of these Con/Lab seats would change hands at all – the Tories would hold theirs, Labour would hold theirs. In practice it wouldn’t work like that of course, there isn’t a uniform swing and some seats would probably switch in both directions, but it’s a suggestion that there isn’t really a swing to either Lab or Con in the key marginals.

A word about polls of marginal seats. I’m always a bit wary of reading too much into them. In theory they *should* be far more useful in predicting elections, in practice… perhaps not, probably because of how uncommon they are. They tend to be rare one-offs and show contradictory things: the last Ashcroft poll of marginals showed Labour doing better in marginals, the previous Ashcroft marginal poll has show the Tories doing a tiny bit better, this one shows the Tories doing much better. That might be different methods, or change over time (there are years between those polls!) or just normal margin of error. With so few such polls it’s impossible to tell – yet because they are so rare there’s a temptation to read a lot into them. You shouldn’t. Rarity of a poll doesn’t decrease its confidence interval. The good news is that this ComRes poll is apparently the start of a series, so assuming they are relatively frequent we will have a better chance to at least be able to look at trends and averages over time and how they relate to ComRes’s national figures.

204 Responses to “ComRes poll of marginal seats”

1 2 3 5
  1. Polling mania!!…

  2. It’s daft for Comes to carry out a marginals poll during a Euro election campaign. Of course UKIP will skew the results.

  3. Quite a big jump for Ukip there…

  4. Yes I think 17% for UKIP is unrealistic, and no seats changing hands is boring. Waiting on Ashcroft’s now!

    Headed for a tight election I think.

  5. I know it’s one poll in isolation etc etc, but that is a terrible poll for Labour.

    As an aside, 1000 in total, over 40 constituencies, is about 25 people per seat? I think in truth it’s a terrible poll for ComRes.

  6. RAF

    Presumably it’s also daft for Ashcroft to have conducted his larger marginal poll during the Euro campaign?

  7. That’s 3 final Euro polls today, and one yesterday calling UKIP by the way. Always thought Labour would turn it around, and they still might, but it’s probably 40/60.

    Mind you, still a couple of days, and news reports for people to hear, campaigners for people to meet etc.

  8. The only marginals that matter are the ones where Labour are in second place. Why on earth did Comres bother with the ones that Labour already hold? A larger sample from the ‘real’ marginals would have been more reliable.

  9. Although the turnout is likely to be pitiful on Thursday, at least we will have millions of people casting real votes as opposed to a few thousand responding to pollsters, either in person, on line or on the phone. No debates about methodology, no reallocated Don’t Knows and no likelihood to vote weightings, just a simple cross in a box.

    These exercises in democracy don’t come around too often and, as anaemic as they’ve become, we should treasure them nonetheless. They should tell us a little more about the current state of political play as well, certainly the council elections if not the slightly surreal Euros.

    In many ways, when we debate opinion polls, we’re discussing virtual politics; statistical substitutes for the real thing. I’m all opinion polled out for now. I can’t take any more. Bring out the ballot boxes!! :-)

  10. Problem for Labour if they don’t pick up 2010 LDs in the marginals… theoretically they knew they were in a marginal when they voted at the last election and didn’t care whether Labour or the Conservatives won, so Ukip is an equally palatable option.

    We’ve seen a slight national trend for the LD>Lab switchers drifting off, possibly to Ukip, but disproportionately so in the marginals?

  11. @Oldnat

    You think either was wise?

    There’s a clear UKIP dimension to the Euro polls in England that is moment to be he case (at least not as significantly) next year.

  12. Mr Nameless

    ‘Headed for a tight election I think.’

    Headed for Conservatives as at least the biggest party I think!

  13. RAF

    Since neither poll sets out to predict what will happen in these marginals in 2015, I think it is useful to have polling conducted during the Euro campaign. It will provide valuable data to compare with polling in subsequent months.

    That will help to illuminate if the UKIP phenomenon in England has a deeper base that could affect the governance of English domestic policy and (r?)UK issues as well.

  14. Tables are here:

    but one thing strikes me as a bit odd. This appears to be an online survey. Now normally marginal polls are telephone polls because that is the only way in which you can get sufficient respondents who live in a particular constituency or set of them. This appears to be a continuing survey of “the 40 most marginal Labour-Conservative constituencies”. So it is extremely unlikely that a panel would include enough people. Are they recruiting 26 people[1] from each constituency and asking them each time?

    [1] From the tables it looks like it is exactly 26

  15. Don’t Ashcroft’s polls usually have juicy, enormous sample sizes?

  16. That’s a swing from Conservative to Labour of just one point, far lower than the swing shown in ComRes’s GB polls (it would be the equivalent of a national poll showing a Conservative lead of five points)

    Those are the key words – if this is representative, this is a massively good poll for CON, massively bad for LAB and it suggests that the “skew” has unwound… Maybe far less anti-tory tactical voting that in previous years? Who knows… But Ashcroft’s poll will be very interesting

  17. @Mr N
    In september ’13 Ashcroft “polled nearly 13,000 voters in the 40 Conservative seats with the smallest majorities: 32 of which the party is defending against Labour, and 8 where the Liberal Democrats came second in 2010”.

    So not only a huge sample but a very different group of marginals as well.

  18. Sorry, second paragraph = my words… Bad html by me!

  19. Billy Bob
    As CB has said, enough already as far as polls are concerned especially this ComRes ‘marginal’, we have Ashcroft’s real deal along soon, then we’ll see. Howard’s right to ask why they bothered polling the Labour held seats.

  20. Ewen Lightfoot

    ” Howard’s right to ask why they bothered polling the Labour held seats.”

    Wouldn’t it be useful to see whether their is a differential shift in the UKIP vote between Tory & Lab held marginals between now & post-Euro polling?

  21. Interesting, but I wish we could have more of these polls.

    This confirms the last 3-4 years of polling, which shows very little direct swing between Lab and Con. When something happens, it’s another party siphoning off support or handing it over to one or both of the major parties.

    Labour are going to be depending on whatever they can get/keep from disgruntled 2010 LD’s and hope not to lose too much to UKIP. Conservatives have only the latter problem, but it would seem that its a worse problem for them. Let’s just see if the UKIP vote unwinds and stays that way or not.

    I think we might see a winner next time with an embarrassingly low vote share – worse than 2005. If it’s Labour, then they might not even win the popular vote.

    As for the EU vote on Thursday, looks like a tossup between Labour and UKIP for the winner on popular vote.

  22. Or even “there”.

  23. @NewForestRadical

    This constant negativity is doing our side of politics no favours at all.

    This embarrassing bedwetting that has been taking place among some Labour supporters in the last couple of weeks reminds me of the pathetic Chicken Little-ish freakouts on various liberal sites in the US after Obama had a terrible first debate against Romney.

    Teh skiez! Teh Skiez! Dey iz forllin!

    CTFO people and wait for the Duke of Belize to unveil his rather more gargantuan poll.

  24. @ Drunkenscouser

    I totally agree

  25. @AnthonyWells

    We don’t thank you enough. What is that. 5 blog posts in 24hrs? Thank you for your efforts, and keep up the very good work

  26. DrunkenScouser

    Quoting the CatBible? Why are you so nasty to R&D?

    Disgusting I call it. I will write a stiff letter to the Times!

  27. What’s that old saying about polls and buses….and this being the year of the bus as I am told in London. It must mean many polls will block the roads to power….of course UKIP have been going on about that for some time….

    I was safely inside the Opera House tonight watching another doomed tragedy La Traviata….now I can wait for some pundit to opine about it not being over…until…

    Thanks to Anthony for his diligence, forbearance and always interesting analysis. Goodnight….Thursday to Sunday will be one long one…for someone somewhere….

  28. @OldNat

    Basement Kitteh haz no pitteh!

  29. ON
    If RM is right re the curious methodology involved in this poll, and the sample size, then very little would be gleaned from looking at the differentials regarding UKIP voters, especially if they are a novel 26 each time.

    Drunken S
    Music to my ears, as I said some time back it should be “Steady the Buffs” as far as Labour is concerned until we get some real results, against which we can calibrate ALL the polls between now and the GE.
    Do you think the Bedwetting Tendancy has replaced the Militant Tendancy in Labour ?

  30. The irony in all these polls is that, after the Euro Votes have been cast, but before they have been counted, we will be treated to a mammoth opinion poll result on Friday.

    Even if we allow for substantial split voting between the locals and Euros, we will have some clear guidance as to what we should expect on Sunday evening.

    One key indicator to watch is the actual number of votes for each party. Given that this set of local elections is concentrated in urban areas of England, if UKIP come even close to Lab’s vote in these areas, then they are very likely to be well ahead in the Euros.

  31. @AnthonyWells

    We don’t thank you enough. What is that. 5 blog posts in 24hrs? Thank you for your efforts, and keep up the very good work

    Hear, hear. A busy period, very well covered.

  32. Lol, ComRes. I think given the timing and the sample size we shouldn’t read too much into it… buuuuut Statty’s regional crossbreaks have been showing the Labour vote going to Ukip in the Midlands over the past few months, so it is possible it’s confirming a real trend.

    I believe the fieldwork for Ashcroft’s marginal poll was done much earlier and it should have a larger sample size. We should probably wait for that one before we get too excited.

    Re. Labour bedwetting, it is possible to simultaneously believe that they’re on course for a victory in 2015 and that 80% is lousy retention for an opposition one year out from an election. As far as I can see, Ed Miliband’s campaigning style is to look like he’s heading for disaster until the last moment and then win decisively (see 2010 leadership contest, 2013 party conference speech, 2014 electoral college/trade union reforms). So I think we’re in for a night of damp sheets but a bright dawn.

  33. Ewen Lightfoot

    It is a sine qua non that Roger is always right – having no axe to grind in UK issues – bloody foreigner!

    But, if you want to exclude every pollster with constantly varying methodologies, and only hope that they have guessed how to measure public opinion in a changing environment then you may search for one.

    (Actually, I agree that without clearer idea of its methods, the ComRes “26 in every marginal” looks somewhat dodgy. That, however, is a different thing from criticising the collection of marginal data during the Euro election campaign. If ComRes is crap now, that method will have equivalent excretory characteristics in the future.)

  34. And I’d like to third Martyn’s comments.

    Posting all these polls and moderating our nonsense must get pretty exhausting.

  35. Martyn et al

    It’s quite clear that the work produced by the so-called “Anthony” cannot be the work of one person.

    “He” is simply the cover name of a network of agents providing disinformation to subvert British Democracy.

    In due course, it will be discovered that he is none other than Karla (with a remarkable similarity to Patrick Stewart).

  36. @ Old Nat

    Nah, he can’t be Karla because he only had one daughter; Anthony has two children & IIRC at least one little Wells is a boy.

  37. Amber

    You prove my point! I’ve read Le Carre and you are clearly one of the members of the Labour Party ready to believe any Russian propaganda about their children production quotas.

    I will write a stiff letter to the Times about this!

  38. Anthony says we should be wary of reading too much into polls of marginals. That looks like understatement to me. This poll would suggest that the Tories are about 5% ahead of Labour nationally, which even the worst polls from Labour’s point of view simply are not showing to be the case. There is no convincing reason that Labour would do far worse in marginals as a whole (though obviously there will be some exceptions to this) than in safe seats for either themselves or the Tories. If anything it’s likely that Labour’s weakest performances will be in (usually safe) Tory seats in the South outside London. Nor do I get the feeling that Labour will pile up uselessly huge majorities in their safest seats, and fail in most of the marginals, though again there’s bound to be the odd exceptional performance even in an ultra-safe Labour seat somewhere (perhaps in London or Liverpool?). The results in the marginals is certain to bear at least some relation to the result in Britain as a whole, but this poll suggests a picture so far removed from what national polls are saying (the very worst of them in the last 3 years have shown a swing of 2.5% to Labour) that I don’t find it credible. I would still think that if Ashcroft’s forthcoming poll says the same sort of thing, unless & until there is some convincing evidence from national polls that the Tories have regained a clear lead over Labour. That clearly isn’t the case at the moment with Labour’s lead in the last week averaging at over 3% (a nationwide swing of about 5 & a bit per cent). Yes there is bound, in what is still expected to be a close contest nationally, to be the odd under-performance by Labour in a marginal seat, but this level of under-performance simply isn’t credible.

  39. I’m not convinced that the nuanced view of the Daily Record on the best way to stop U-Kip getting a Scottish seat will be clear to their readers, but it certainly isn’t a “Vote Labour” message!

    On Thursday, I wouldn’t be at all surprised to find my pal, who is a Labour polling agent, voting SNP, while I (as an SNP polling agent) votes Green.

    No wonder Scottish politics seems to confuse party loyalists dahn south! :-)

  40. Barnaby Marder

    I’ve been in or around politics for the last 54 years. During that time I’ve seen party loyalists (on all sides) responding that ” this level of under-performance simply isn’t credible.”

    Sometimes they have been right, but …..

  41. Oldnat

    It’s actually ‘dahn sarf’

    As a not quite as Northern as you Northerner.

  42. Jack R

    Yeah. It didn’t look right, but I’m no good at foreign languages. :-(

  43. @OldNat

    It’s Comres. If they told me today was Wednesday, I would check the calendar for confirmation.

  44. We really need to know more about ComRes’s methodology in these polls, but one thing that does seem to apply is that ComRes’s online polls are very UKIP-friendly and seem to over-represent them.

    There is also the matter of how the questions are asked. ComRes just seem to have asked their normal VI questions, but the whole point of marginal polling is that you want people to respond with how they would vote in their constituency so tactical voting, the candidates involved and so on are considered.

    You also need to be careful about how the marginals are representative. If Labour has lost votes to UKIP in the Midlands and 12 of these 40 marginals are in the Midlands, the overall result may not be useful because the selection is only representative of itself. It may tell us about the 40 closest marginals but not the 100 closest for example

  45. Re the Euros, I think that UKIP have lost some support in the last few days due to continuing gaffes by their candidates (including NF) and attacks by the media and other political parties, and feel that this has not been fully reflected in the final pre-election opinion polls. In addition, UKIP’s support is more evenly distributed across England & Wales and this election is still constituency-based (even if the constituencies are large), so while they will do better than in 2009, the main winner will be Lab. LD will be the main loser, with their support haemorrhaging outside the South of England.

    My prediction in terms of % of the vote (seats in brackets) for GB (70 seats) is therefore as follows:
    Lab 29% (23), UKIP 25% (20), Con 23% (20), LD 8% (3), Green 5% (1), SNP 4% (3), PC 1% (0), Others 5% (0).

  46. The problem with this exercise is that it bases marginality on the previous election. Of course some of these may be uber-marginals again but this election may be decided as much by the seats where the LDs were in the lead by a moderate margin as by these LAB/CON seats. Any LAB/SNP swings may also turn out to be a major factor.

    It would be very interesting if somebody did a poll of LD held seats actually.

  47. “Presumably it’s also daft for Ashcroft to have conducted his larger marginal poll during the Euro campaign?”

    The fieldwork has been finished for some time now, not sure of the dates though or whether some was during the Euro campaign.

  48. “The only marginals that matter are the ones where Labour are in second place. Why on earth did Comres bother with the ones that Labour already hold?”

    Are you serious??? What a hilarious view of the next election if you think it’s impossible for Labour to lose any seats.


    But, some of us do expect Labour to lose some seats in 2015.

  50. Far Easterner

    Sorry, misread your post, as you can see I am agreeing with you.

1 2 3 5