A ComRes poll for the Independent tomorrow has topline voting intentions, with changes from their last poll, of CON 41%(+7), LAB 33%(-4), LDEM 16%(+1). In amongst the flurry of polls we’ve had over the last two months ComRes haven’t polled since mid-September, back before conference season began, so missed out on all of the Labour surge after their conference and the Conservative recover after theirs. The changes in this poll are across the whole of the party conference season and the non-election annoucement (one might well think this gives us a far better picture than pouncing on all the ups and downs in the last six weeks’ of polls. Maybe it does, but given the speculation over a possible general election polls during conference were inevitable this year).

The eight point lead is the largest enjoyed by the Conservatives in any poll since April, though since they started weighting by past vote ComRes have tended to produce some of the better poll findings for the Tories. It is also a good rating for the Lib Dems compared to their recent poor showings. It’s tempting to take these findings as a suggestion that a Lib Dem recovery would hurt Labour more than the Conservatives, I think it’s a bit early to conclude that yet though, ComRes tend to be nice to the Tories anyway and, not having seen any of the really poor showings for the Lib Dems in ComRes polls, we can’t really conclude this is a recovery from an slump that happened ‘offstage’ in the gap between ComRes polls. I don’t think we’ll know for sure till there is a new Lib Dem leader whose had a chance to make their mark.

56 Responses to “ComRes gives the Conservatives an 8 point lead”

1 2
  1. Anthony,

    I’d be interested in your take on the breakdown of the ComRes figures. As always the sample size is low but i am surprised that so many of our supporters are saying they will definitely vote.

    The feel good factor for the SNP is strong, but at present we seem more likely to vote at Westminster than even the Tories, which seems remarkable.

    I think this goes a long way to the results showing us welll ahead of Labour if there was an election, a result I just can’t bring myself to believe.

    To put it in persepctive a result like that would se the SNP getting more seats than Labour.

    We really do need a decent Scottish poll ( any chance Mr Kelner could point out to the Scottish Media that they might be missing something significant) so we can tie this down one way or the other.


  2. I hope I’m right in thinking that ComRes’s silence over the last weeks has been simply because no media outlet has asked them to carry out a poll?

    Peter – Could it be an SNP “conference bounce”? Alex did come across well in the media in the last week – “an announcement for every day in office” etc. Most English taxpayers now think prescriptions and school meals are free for all up there as of last May!

    Do polling organisations actively tout for business? I’d have thought that that could bring them into disrepute! Might I suggest you use your proximity to the “Scottish Media” to get a poll done.(presumably there’s a “Black Isle Echo?)

    I get the feeling that Cameron is content for the time being to try to make Brown’s life as uncomfortable as possible, and maybe provoke Labour into making rash decisions. He knows the election is a while off yet – any “commitments” he talks about now can easily be swept away before manifesto time.

    The polls might well be about to revert to the traditional “mid-term” effect, where the voters are disappointed that their hopes haven’t been met, excited by the prospect of realisable change, and thoroughly vulnerable to a pre-election bribe or two in eighteen months time.

  3. Who else had the first thoughts of “Lib Dem 16% – Ooh, they’re doing well here”?

    Just shows how far they’ve fallen in expectations that 16% seems good in comparison.

    The big questions are of course will a new leader get a boost – and if so from whom? And how permanent will it be?

  4. in the election guide section of this site, the uniform swing result for this ComRes poll should be checked i think, it suggests Labor maintaining a substantial majority, and doesn’t make sense compared to the other polls shown.

  5. It seems that the drop in Labour is mirrored in Scotland, but instead of going to the Conservatives, the vote is heading to the SNP.

    With the Liberals under 15 and the Conservatives under 20, the results of Labour being under 30 will cause a seismic shift in seats in Scotland.

    Electoral Calculator shows these figures resulting in a massive loss for Labour and a massive jump for the SNP.

  6. Oliver: I think you’re making a mistake.

    I get results from this poll = Con 323 (+125, just short of absolute majority), Lab 270, LD 28, Others 11, NI 18.

  7. Hmm, you’re going on the text rather than entering the numbers? Just noticed that, its I guess the results from the previous poll, not this one.

  8. Having just checked the calculated seat allocations (and in particular the ComRes poll) I find that the analysis is meaningless.

    Using the online calculator I produce an outcome of:

    Conservative: 323 [41%]
    Labour: 270 [33%]
    LibDems: 28 [16%]
    Others: 18 [10%]

    Obviously this is not the message YouGov’s corporate sponsors “required” [Source: http://ukpollingreport.co.uk/guide/%5D:

    Conservative: 228 [41%]
    Labour: 352 [33%]
    LibDems: 41 [16%]

    So much for your reputation….

  9. Now that the Tories are hitting over 40% percent, it would be interesting to see the reginal breakdown. Are the tories making advances in the north of England or are they still flatlining at 29%;which would mean increasing their votes in safe southern seats.

  10. John T,

    The poll was conducted during the conference so there could have been a boost, but there are two things that make me think it is more than that, or at least it’s not the whole story.

    Firstly the actaul numbers voting for us isn’t due to switching but rather what appears to be an end to the reluctance of SNP voters to come out for a westminster election. In the past there has been a marked difference between our turn out at Holyrood and Westminster.

    Secondly if you check back you see something similar in September under ComRes.

    Now it could be that when people give there answer they are not really thinking about Westminster as such and are just showing support for what the SNP is currently doing.

    The other thing to bare in mind is that where as in a UK election the LibDems tend to get a boost from the publicity the SNP get squeezed as we don’t get equal coverage on UK news with the other three parties.

    Again it’s yet another Blogg that Anthony might consider, mapping over time the turn out per party and the numbers who will stick with it.

    In the past for westminster the turnout was Tories top then libDems, followed by Labour and in Scotland the SNP last. We are now seeing the SNP top and ahead of the tories and that is a big change.

    For a while the SNP have had the most loyal vote followed by the Tories then Labour and finally the LibDems although I ssupect a lot of the reason for the apparently low LibDem loyalty is tactical voting.

    I certainly think there is a change, but whether it is sustainable until the next Uk election or represents a real shift in voter behaviour I just don’t, know but it is interesting.


  11. Oliver/Philip – the projections on the frontpage of the Guide are automatically updated whenever I stick a new poll in the database. Unfortunately the frontpage looks for the latest Communicate projection in the database, rather than the latest ComRes one, so it was putting out old figures. Corrected now.

    (*Alternatively of course, it could be that incorrect swing projections are all part of my evil YouGov paymasters’ plot for world domination, probably involving lizards. There you’ve found me out, and I would have gotten away with it if it wasn’t for those meddling kids.)

  12. Anthony,

    Welcome to the club, the Scooby Doo school of politics.

    As an SNP member I am regularly called a tartan tory, a secret trotskite, and on occasions a Nazi (because we’re nationalists and partly socialist).

    At least once each I’ve been a “Orange Mason” and indeed a “Agent of Rome”.

    I’d like to assure you all that other than the nationalist bit,and as you can probably tell being left of centre in inclination and liberal in outlook, I am none of the above.


  13. Well it’s a very good poll for the Tories and the first one that shows Labour’s share of the vote dipping to a level where they might not win. I will follow my own advice and not seek to debunk the poll just because I don’t like what it says. My only comment on ComRes is that they have been AWOL during the most important period for polling in the past 10 years and it is therefore hard for us to know how they would have recorded the sharp fluctuations that occurred over the conference season. Regardless, they now record a 8 point Tory lead which (according to the Independent who commissioned the poll) has arisen from a surge of support for the Tories in London and the South East. Given the very small sample it is probably wisest not to focus on regional shifts but to take the national figures at face value.

    Going forward, we will need of course to see how ICM and Populus show the parties. In the past week Yougov have given the Tories a 3 point lead and Mori show a 1 point Labour lead. Whether ComRes will stand out as overstating Tory support or prove to be the harbinger of a Tory surge is yet to be seen.

    Putting my floating voter hat on for a moment, I detect a mild disinterest in politics at the moment. We’ve had some excitement but we’re back to the daily prattle of the Diana enquiry and the Tapas 9. My prediction is that we will continue to see apparently conflicting polls for a while yet depending on which side of the bed the floating voter gets out of in the morning.

  14. You would say that though, wouldn’t you Peter? You secret Trot you!

    Arnie – you are entirely right to pay little attention to the regional shifts within GB polls, with small sample sizes for the individual splits it really is impossible to discern any real movement there amongst the noise of random sample error. ComRes have tended to produce the highest Tory leads of late (they had the parties neck and neck when all around showed Labour leads and were the final pollster to record a Tory lead during the Blair->Brown handover), so I wouldn’t be surprised if ICM found a lower lead.

  15. Arnie is right.I think it is a bit rash of everybody to read so much into this poll or the recent contradictory one from MORI. Wait and see which way the ICM poll bounces and just perhaps it might be possible to reach some hesitant conclusions about which way the prevailing wind is actually blowing.

  16. Arnie – i think it’s a little unfair to say ComRes have been AWOL – they seem to me to do monthly polls for The Independent, and can’t be expected to carry out polls when they haven’t been commissioned.

    Peter – you’re suspiciously quiet about my previous allusions to witchcraft another thread. I’d have expect that to feature on your list of labels wrongly attached to you – unless of course I was nearer the knuckle than I’d thought?

  17. I still fear (as a Tory) that our vote may be quite a concentrated vote, building up in areas which feel heavily taxed, and somewhat hitting the buffers in marginals where there has been a lot of public sector agency regeneration.

  18. 601, 9.27am:

    Here is the regional breakdown :

    The Conservatives are 24 points ahead of Labour in the South East (50 : 26), 10 points ahead in the Midlands (46 : 36) and 6 points ahead in Wales and the South West (38 :

    There has been a remarkable recovery for the Tories in Northern England where they are only 3 points behind Labour (38 : 41)

    In Scotland, the Tories are on 17% while Labour are on 24%, with the other parties (mostly the SNP) on 44%.

  19. Joe James B –
    I’m not a Tory, but i disagree – I think the Tory lead is because of the success of the “message delivery”. Cameron’s message, that “life under Labour isn’t fair” is hitting home – in the allegations of denied referendum, denied election, IHT, discrepancies re the West Lothian Question, immigration, etc.

    Areas of high public sector employment may well be worried about the prospects of job losses (aka “waste-cutting”), but the appeal is to the voters’ sense of fairness, rather than to their wallets.

  20. I actually think that as far as the Lib’s goes, ComRes are the most realistic about the Liberal position. Those polls that were constantly showing the Lib’s on 11, 12, 13 points were completely unrealistic if you ask me, and theres no way the Lib’s would have polled that low in the GE. 16% looks a much more sensible position.

    Whether that means the Labour/Tory ratio is right, I’m not sure. I rather suspect Labour would be a bit up on that, say 35-36% but I think they have the Tories about right on 41%.

  21. John T, that breakdown would be enough to give the Tories a comfortable majority, wouldn’t it? Thete are enough steats in the south and the Midlands to deliver the Conservatives a victory, yes?

  22. Sorry, that last post should have been directed to Mountjoy.

  23. GIN – I don’t know how the map, or the House of Commons would look, but I’m sure things will shift around a few times over the next months.

  24. GIN – Yes, hese results suggest that the Tories would pick up enough marginals in the midlands and north that would give them a majority in the Commons.

  25. This is interesting, because in the latest ComRes poll the Conservatives are ahead by 10 points in AB (42 : 32) and DE (41 : 31) social class and 12 points in C1 (43 : 31) and neck and neck amongst C2s (38 : 39). Similarly, there has been a remarkable recovery for the Tories in Northern England where they are only 3 points behind Labour (38 : 41).

    Direct shifting from Labour to Conservative will topple a lot more Labour MPs in Marginal Seats.

  26. Interesting how the media are not really giving any real coverage to the Liberals during their leadership changeover as they would to the 2 main parties . During this week i have noticed that the Liberals were pushed into a tiny slot 30 minutes after the headlines on news 24 and even on sunday only one political debate show talked to the Liberals .

    I think that the British public and the media are now accepting the fact that the Liberals have been and continue to be a protest vote party and also a tactical vote party with little significance in todays politics – we’re getting where we always should have been for a healthy democracy with or without Liberal meddling – a 2 party state . The POLLS are showing that more and more .

    The Liberals for too long have decided the choice of government in this country because of their tactical voting in marginal seats and have derailed true democracy .

    They are now getting their payback from the electorate for this meddling for the last 30 years + . They have lost control in Scotland and Wales & credibility with their pacts with Labour to gain power since the 70’s.

    At least with the Tories they have always said they would never join in a pact with any party to gain power or hold onto power (the Ulster Unionists as an exception – because they were Conservatives). This cannot be said of either Labour or the Liberals who will join with anyone to hold power !! At what cost – this can now be seen in the shambles in Scotland and Wales and the mess left in 1979 when the LibLab pact finally finished .

    [Mike – I haven’t had to moderate any comments from anybody here for a week or two now, which is good. Try not to slip back into being partisan – AW]

  27. I think there is a case for saying the Lib Dems get too much coverage.
    I’m not sure why they are allowed to have equal footing with the other 2 parties at General Elections.
    Maybe it was something that was changed after they reached 25% in 1983.

    But even on the basis of the 2005 election result, it is logical that their coverage should be cut by about a third.

  28. Mike What a very curious view of democracy you hold .

  29. When did the rule that the LDs get the same coverage come into effect? It does seem odd, given that in PMQs their leader get 2 questions v the opposition leaders 6, while other minor parties like UKIP/BNP/SNP/PC etc don’t get equal coverage.

  30. There are examples of Tories going in to partnership with others at County and local level to keep hold of power. That is what we call reflecting the wishes of the people!

  31. Much as I may think ComRes is what people ought to be voting, I do need to point out that the Weighted Moving Average is 40:37:13 and that ComRes has a Standard Deviation error of 3.3 (vs WMA, 3.6 vs interpolated) so an error in the lead of 4.7 is not atypical. It’s now pretty clear that the Conservatives are over 40% (6 polls in a row) which is (I think) what I predicted for October. The ComRes “error” is a 3 point switch between Lab and LibDem. I think we’ll see the WMA move to 43:33:14 or thereabouts within 6 months, and quite possibly by Christmas.

    BTW: doing the back-interpolation (average of 5 polls: 2 before and 2 after) it is clear that the Ipsos/Mori Observer poll was indeed a rogue, it is 5.2 points out on that basis (as opposed to 2.2 on WMA) and overall Ipsos/Mori have a StD of about 2.8 and underestimate C leads by about 1.5

  32. NBeale, how far back do you go to compute the WMA? If the last 6 polls in a row have put the Tories 40+, then how come the WMA is 40 dead.

  33. I find it fascinating how much the polls have swung in a month; it’s a cliche but they might equally reverse in the next 4 weeks (except in Scotland). If I was a politician I would not sleep easily as the swingers seem to swing too rapidly.

  34. Australia

    And not strictly relevant but interesting, here is the news from Australia—

    ‘The closely watched Newspoll in the Australian newspaper showed Labor leading Prime Minister John Howard’s conservative coalition government by a comfortable 58 percent to 42 percent on preferences, a gap of 16 points.’

    Which, if transfered to the election in November would be a total landslide to Labor. Issues include; troops out of Iraq, Kyoto treaty, republic and it’s worth noting the Labour leader is also comfortably ahead as the preferred PM; he is a former diplomat and fluent Mandarin Speaker.

  35. Anthony,do any polls increase their sample to give a better regional picture or is this too expensive?

  36. Many thanks to Mountjoy for answering my question. :)

  37. Philip
    It’s a weighted moving average and I give 25% weight to the current poll, so in principle it “remembers” all the polls but obviously the weighting of old polls is very small (eg 2.5% after 8 polls, 0.25% after 16). This is the best (simple) way I know to get a measure which smooths out sampling errors sensibly. I also take a 5-poll average but use that to estimate the “true” state in the middle poll, so it lags the current poll by 2.

  38. Collin –

    Regional breaks still wouldn’t be particularly solid with bigger samples, since the polls are only weighted to be representative of the country as a whole, not to the representative within regions, even so, there wouldn’t be much interest in doing so. It would be a lot more expensive, yet its unlikely to sell any extra newspapers.

    If papers do want to do regional figures they just do aggregated figures from several months polls shoved together, which while out of date is probably just as accurate and doesn’t cost them any extra money.

  39. -MORI has Labour moving back ahead.
    -YouGov shows no change.
    -ComRes shows the Tories storming into their biggest lead for months.


    Again, support for “others” has dropped by 4%. Perhaps UKIP/BNP/Veritas has lost some support to the Conservatives?
    I despair at people thinking this poll result would translate into an overall majority of 2 for the Tories. It would be more like 50 once the “marginal seat factor” is given just an ounce of consideration.

  40. Excellent comments from Mike Richardson.
    Well said.

  41. Jack, the latest Newspoll gives Labour a 54/46% lead, which is still pretty comfortable.

  42. What’s that – a forced choice part of which poll?

  43. JJB – It isn’t a forced choice poll, Australian polls reallocate votes for minority parties to come to the 2PP (2 party preferred) figures.

    Of course, since voting is compulsory Australian elections are a forced choice :)

  44. Anthony,

    Any view on my point about the trends in likelyhood to vote, not just the SNP but in general trends on the parties ability to motivate their supporters to come out.

    If we are close to 40/40 then getting the vote out could make all the difference.

    You can have the most supporters and still lose, ask Al Gore…


  45. “If I was a politician I would not sleep easily as the swingers seem to swing too rapidly.”

    I think all that has happened in 2007 so far is that the Conservatives had a modest but steady lead, then we had a Brown Bounce, then we’re back to a modest but steady Tory lead again.

    It is very hard to envisage Labour notching up double-figure leads for some considerable time now that the Brown Bounce is over.
    That would probably only occur in the highly unlikely event of Gordon Brown being replaced as PM during the course of this Parliament – or, more likely, an eventual revival after some years in Opposition.

  46. The swing among the specific demographics from the previous ComRes poll to this one seem:

    Male: 11% Lab to Con / Female: 2% Lab to Con

    18-24: 13% Lab to Con / 25-34: 13.5% Lab to Con / 35-44: 2.5% Lab to Con / 45-54: 6.5% Con to Lab / 55-64: 7% Lab to Con / 65+: 12% Lab to Con

    AB: 3% Lab to Con / C1: 9% Lab to Con / C2: n/c / DE: 12% Lab to Con

    South East: 14.5% Lab to Con / Midlands: 6.5% Lab to Con / North: 2.5% Con to Lab / Wales & South West: 5% Lab to Con / Scotland: 7% Lab to SNP

  47. … seem remarkable (that should read)

  48. Thank-you Mountjoy.

  49. Sky News reports that the ICM/Guardian poll is Con 40, Lab 35, LD 18

  50. Athony.Many thanks-understood. So the regional breakdown published by Comres has no validity. Wonder why they issue the numbers?

1 2