Channel 4 has a new poll of marginal seats for Channel 4 which once again suggests a slightly higher swing to the Tories in Lab/Con marginals. YouGov has polled this same group of marginal seats four times for Channel 4 over the last two years (for my posts on the previous rounds, see here, here and here). Voting intention in these seats stands at CON 39%, LAB 37%, LDEM 15%.

The 60 seats YouGov poll are all Labour held Conservative targets that need swings of around 3% to 7% to fall (i.e. the Labour seats that the Conservatives would need to win to get an overall majority). At the last election they had a Labour lead of around 11%, so this poll reflects a swing to the Conservatives of about 6.5% in Labour held marginals. YouGov are currently showing a swing of about 4.5% in their national polls, suggesting the swing is 2 points higher in Labour held marginals.

While the Labour lead has gone up and down over the years, this differential is actually pretty consistent with YouGov’s last three rounds of polling:

In September 2008 it showed a Conservative lead of 13 points – a 12 point swing when YouGov’s national polls were showing an 11 point national swing
In October 2008 it showed a Conservative lead of 5 points – an 8 point swing when YouGov’s national poll were showing a 5.5 point national swing
In February 2009 it showed a Conservative lead of 7 points – a 9 point swing when YouGov’s national polls were showing 7 point national swing

Polling evidence from several companies has consistently suggested the Conservatives are outperforming in Labour marginals to some degree (though there are differences in the extent!). This latest poll confirms that… but it is a relatively small advantage. If this swing was repeated in all Labour marginals, it would not be enough for the Conservatives to get an overall majority. It’s also worth remembering that it doesn’t tell us anything about Lib Dem v Conservative marginals – at the risk of straying into a horrible fairground related metaphor what the Conservatives gain on the Labour held swings, they may be losing on the Lib Dem roundabout.

(On other matters, a TNS BMRB voting intention survey conducted in February has surfaced here. The fieldwork is a bit elderly now, from Feb 18th to the 24th, and other than it being face-to-face I don’t know anything about the methodology, but the topline figures are CON 36%, LAB 32%, LDEM 21%.)


128 Responses to “YouGov poll of Labour marginals”

1 2 3
  1. @TheGreatGonzo,

    Astro Empires. Very good game if you’re a complete geek like me. And let’s be honest, why would I be on UKPR if I wasn’t?!

  2. Effectively what this poll suggests is that compared to the same company and same methodology, the swing to the Tories in these marginals has fallen by 2.5% in exactly one month. we can still argue about whether YouGov or AR have the most accurate methodology, but everyone should be able to agree that this is a shockingly fast movement within the key battleground seats.

    For those who think Ashcroft hasn’t and won’t have an effect, the story has at the very least prevented the Tories from discussing real policy issues – they have in effect lost a week of campaign time, which is about 10% of the time they have left to secure a GE victory. i wonder whether it has had an insidious effect on the underlying numbers. If only 20% of voters think cameron will be a change for the better, that’s a real problem.

    I think this poll more than any other so far will be ringing alarm bells in CCHQ

  3. @Neil
    I’m a geek, I’ll make a note of it before this little off-topic exchange gets pulled ;-)

  4. If you sign up, TGG, do so on Epsilon server and start in the E64 region…. My nickname is “Light Aura”

  5. Hi there, first post on this site and I hope it’s not appalling etiquette to copy something I just wrote on my own blog, but as my blog has a readership of roughly zero (within a margin of error!) it’d be good to share some thoughts with other people! Anyhow…an amateur view…:

    Anything with less than 6% majority for the Labour, we can assume that the Tories will win – at any rate, they’ll have to if they are to win a majority in Parliament. They will also have to take the bulk of the 6-14%ers as well, presumably why YouGov chose these seats as their sample.

    It has been suggested that the Conservatives have been effectively getting their message across in key marginals better than nationwide across the country and some Tory commentators have been optimistic about their prospects; the topline figures this time around won’t make pleasing reading with just a 2% gap – comfortable for Labour. (The results: Con 39%/Lab 375/LD 15%/others 9%). Gordon Brown’s approval rating is pretty low (49% describing him as fair, poor or dreadful) but this is almost unchanged from 12 months ago when the same survey was taken. Cameron’s rating are marginally better – he will also be encouraged by his “DK” ratings falling; his presence is being felt, for better or for worse.

    Given that these seats are all Labour-Conservative battlegrounds, the crucial fight will be for the squeeze on the Liberal Democrat vote, so an analysis of the opinions those claiming to be currently intending to vote Lib Dem on the Labour and Tory credentials respectively may give insight into how those Liberals who will decide to relinquish their true colours and vote tactically will go. Unsurprisingly, they can’t get excited by either Brown or Cameron but Brown is the more popular. 39% of Lib Dems thought that the country would take a turn for the worse under a Conservative administration, with only 8% thinking life would get better – worrying for the Tories when it comes to the squeeze.

    The headline for the Lib Dems is that their prize asset, Vince Cable, is considered the best potential Chancellor by quite some distance: 10 percentage points ahead of Alastair Darling. It’s not a surprise but will make a good soundbite for Cowley Street. This question (and the subsequent one asking respondents to compare the economic credentials of Cameron-and-Osborne vs Brown-and-Darling) put in lights something that was becoming more and more obvious: George Osborne is the Tories’ Achilles Heel. Given that the economy was by far the most imporant issue to the electorate, this should worry Conservative HQ. Health was another important issue, and with the Labour-led “Save the NHS” campaign bubbling along nicely, the Conservatives may look to the immigration issue to push in the marginal seats. Immigration was the second most commonly cited issue and with the Tories strongly out in front in their handling of it (although as I pointed out yesterday, I think the question was badly worded), it might be in their interests to let it rear its ugly head.

    Asking whether David Cameron is “heavyweight” or “lightweight” is a cheap question but the results were pretty comprehensive: 62% describing him as lightweight. Even amongst Tory voters, only 44% thought he was a heavyweight. When the personalities of the leaders were analysed, Labour may worry that the trait that was most prominent about Brown was that he “can’t work effectively with his colleagues”; having said that, you could flip that on its head by noting that only 30% thought that Cameron “can work effectively with his colleagues” (when I was responding I didn’t spot the subtle difference in wording!). Also encouraging for Labour is that post-Bullygate, there is only a small difference in the numbers who think of him as a bully versus those who’d describe him as a strong leader. As for Cameron, the “too posh” jibe was also a bit cheap – although it’s interesting to note that there was only a small difference in sentiment between ABC1s and C2DEs on this.

    Nick Clegg’s problem is one of visibility. His rating are consistently poor for individual issues, and comparitively lower than you might expect from his party’s ratings. Voters are clearly not stupid – it must be a long time since a Liberal shadow chancellor was considered the economic heavyweight – but Clegg has an identity crisis. The upcoming TV debates, and other election materials where the three main parties will get equal billing, will come not a moment too soon and allow Clegg to show the electorate who he is and where he stands on the main issues. I would expect Clegg’s ratings to rockets once the campaigns begin in earnest; it’s a shame that the PM+Chancellor combo question didn’t include Clagg + Cable as an option as I think that duo could be very popular indeed at this time.

    As I surmised from the wording of the questions, incumbent Brown was considered less honest than Cameron on the state of the economy (hardly surprising – it’s not new for a politician to downplay his own mess) whilst on the other hand Brown is significantly ahead when it comes to “understanding the problems ordinary people face in difficult economic times”. A subtle question loaded with class warfare, but one that may inspire some Labour campaig literature.

    One interesting question with an unbalanced response scale regarded the economy and the way forward. respondents were offered three options on how to get the country out of the mire: (1) cut the deficit, (2) not cut the deficit as it would risk recovery, (3) not cut the deficit but protect jobs and services (I paraphrase here!). The results here, I feel, are rather meaningless – with two “don’t cut the deficit” options versus one “cut it”, bias is introduced. 48% plumped for a “don’t cut the deficit” option. Interestingly, there was a very low (13%) “Don’t Know” response rate for such a technical question; this bodes well for any politician who actually wants to talk about policy, and may explain Vince Cable’s high ratings as he has a knack for presenting technical arguements convincingly.

    Of course, these results only refer to Labour-Conservative battles. The Tories will need to pinch some Liberal Democrat seats too – and that’s without mentioning numerous Tory-held seats that the Lib Dems could quietly snaffle. YouGov’s Peter Kellner, in his analysis, reckons that the Conservatives will struggle to make a net gain of more than ten seats from the Liberals, and that these figures applied to the marginal seats would not be enough to give Cameron’s party a majority.

    I need to stop lurking here and start posting, this place is a goldmine.

  6. Some newspapers are now reporting that the poll in and of itself implies that the Conservatives would be short by 11 seats, which is completely wrong.

    As we all know, the poll itself only deals with Lab/Con marginals, not LD/Con marginals, so no prediction of how close the Tories would be to winning a majority can possibly be made simply on the basis of this poll.

    Yet that is how the media is reporting it. Not the greatest day as far as the media reporting polls is concerned.

  7. William – still nonsense, you’ve just shifted the goalposts to try and make it true. Except, for what it’s worth, even with the shifted goalposts it’s STILL nonsense.

    The Conservative leads in final polls in 1979 were 2 points (Gallup), 5 pts (MORI/Express), 6.5 pts (Marplan), 7 pts (NOP) and 8 pts (MORI/standard). Their actual lead was 7.2 so all except MORI’s poll in the Evening standard underestimated the Conservative lead. A winning opposition out performed the polls.

    In 1970 the polls famously got it wrong, with all but ORC predicting Labour ahead, when in fact the Conservatives won. Once again, a winning opposition out performing the polls.

  8. Alec – I don’t know if I’ve mis-read something from your comment, but the last YouGov poll in these marginals was Feb 2009, not Feb 2010 – so it’s a drop over a year, not a month.

  9. Like a previous poster has said, why not a poll on all marginals, say those needing less than 8% to fall. There are too many variables to accurately assess that all 60 would fall. 37% suggests a reasonable amount of labour support when i imagine therange in 2005 would have been 35-45%. I couldnt imagine that with 37% support Labour will lose all 60 seats.

    On another note if the torys won the popular vote but the ld’s backed labour, wouldnt that be political suicide as the torys woud have a mandate to at least form a government.

  10. Historically, how do the polls tend to swing once the campaigns enter their “official” stage – towards the opposition or government?

  11. @Andy Stidwill

    In your first post on this thread you said

    “This swing is big enough for the Tories to win all the seats they need from Labour to win an overall majority.”

    In your most recent post on this thread you said

    “no prediction of how close the Tories would be to winning a majority can possibly be made simply on the basis of this poll”

    A mea culpa required perhaps ?

  12. @EOGHAN

    I have always felt that support does increase for the incumbant during an election campaign by a couple of percent but I am sure there are plenty here who disagree.

  13. I’ve tried looking for a list of the 60 constituencies, but without success so far. Can anyone point me to it, please?

  14. Epochery/Eoghan – I answered someone asked the same question the other day. Basically there’s no hard and fast rule. Sometimes the lead narrows, sometimes it widens, sometimes the incumbent does better, sometimes they don’t. It all depends on the campaign.

    If there is any consistent pattern, it’s that the Lib Dems do tend to improve thanks to the extra media coverage (though even that is exaggerated sometimes – many of the bigger campaign jumps for them come from the days before the Lib Dems were prompted for in pollster’s voting intention questions)

  15. I must admit I am confused. It has been said that the Tories are doing better in marginals than nationally. According to this they are doing much worse. Any explanations?

  16. David P – look at the swing, not the shares. The Conservatives were 11 points behind in these seats in 2005, so going to 2 points ahead is a 6.5% swing.

    In GB as a whole they were 3 points behind in 2005, so going to 6 points ahead is only a 4.5% swing.

  17. “If there is any consistent pattern, it’s that the Lib Dems do tend to improve thanks to the extra media coverage ”

    True – the Lib dems are also arguably the best of the parties at on-th-ground campaigning which hasn’t got going in earnest yet.

  18. @Epochery,

    If the LibDems backed Labour, their combined vote share would be (probably) over 50% so the coalition government would have a perfectly good mandate to govern, even if the Tories had the largest single party vote. Its a completely normal situation that wouldn’t be remarked on in countries more used to coalitions. We’d get used to it too after a few months I think.

    @Rob Sheffield,

    I think there is a subtle interpretation needed in what Andy S is saying. I read it thus;

    To get a majority the Tories need to achieve a combination of results that include getting a certain number of gains from Labour. This poll suggests that with regard to those gains from Labour they are (just barely) on course. However they also need to get a certain number of gains from the LibDems (and/or the nationalists) and this poll tells us nothing about that. Therefore both statements made by Andy are technically correct, albeit his wording was a bit unspecific.

    @Andy S,

    Sorry to jump in as your unsolicited defender there….

  19. Neil A – thank you for excellent postings.
    people need to look at how the stats were in 2005, and the swing reqd from then to GE not the 39/37 figure. Also the poll did not bother with seats that are closet marginals as they assumed (rightly or wrongly) Cons will win.

  20. @Anthony – thnaks for pulling me up. I comletely misread the date of the last time this poll was conducted, so my last post was largely rubbish.

  21. Alec – I hadn’t read through the whole thread, so I wasn’t sure that you weren’t talking about something else I’d missed myself!

  22. Does the Labour poll underperformance at the GE have anything to do with tactical voting? If people respond Labour to a pollster, but vote LibDem to keep the Tories out, that would reduce Labour’s GE percentage, but not affect seats.

  23. Tories need a triple digit turnover on labour/tory marginals to have any chance of taking power. 95 seats changing power from labour to tories won’t be good enough as this poll would project. This would leave tories the biggest party in a hung parliament but labour in power in a minority government with Brown as PM. Tories better hope polling firms are underestimating their vote. I think the tories need at least over a 7.5 swing in marginals and the trend is in the wrong direction. Maybe by the election their swing in marginals with labour is 5 percent.

    This goes with another poll of labour support in northern england marginals firming up. Tories had a big lead last summer in a poll of northern marginals which came way down a poll a few months back.

    Labour mp’s were quoted in background in a recent article saying the best they can hope for going into election day would be a two percent tory lead in the recent daily yougov poll.

    With all the bad yougov news tories have dumped comras from last election for private polling and hired yougov for their private polling. Wonder if the former CEO of yougov winning a safe tory seat had something to do with this.

    This would leave tories starting on boundaries at 210 and would leave them at 305 seats.

    Getting 95 labour seats on the target seats on this page would get tories to target seat 125. This isn’t really that accurate because target seats leave out trends from previous elections and geographics.

    A recent poll had a swing of 1.5 in tories/lib dem marginals. This would only allow 9 lib dem seats.

    Out of the 27 lib dem target seats from this site in the tories top 125 this 1.5 swing would leave them with only 9 lib dem seats. Add in one SNP seat at that gets tories to 315. This is assuming they hold onto all their seats and the seats they don’t have now but are favored by boundaries. I think it’s possible some lib dems could keep seats tories are favored in with new boundaries but lib dem mp’s hold.

    This wouldn’t give cameron any good options. He would need DUP seats to get to 323 with Sinn Fein not seating. Tories don’t want to have to rely on DUP.

    Brown will make Cameron get a majority to present to the queen. I think Brown thinks he is the the only man for the job and he doesn’t think highly of cameron.

    The times this morning had an article and buried in it there was a line about labour mp’s predicting brown will stay on even if tories have more seats than labour but less than a majority.

  24. Sorry

    PhilBllu r u MLB Did you go throgh deed poll? lol

  25. While it’s true the poll shows the swing to the Tories is higher in the Lab-Tory marginals than nationally, it also shows support for Labour is much higher in these marginals than nationally.

    Labour is doing 5% better in the marginals than the latest national poll. The Tories are doing 1% worse.

    Although the Ashcroft affair might not be affecting national polls, given where he’s invested his money, the impact on marginals might be greater. Enough swing voters might just be appalled at attempts to ‘buy them off’ and so become less likely to vote Tory.

    Crucially, it’s the trend that counts. Support for Labour is rising in the marginals and nationally compared to even a month ago. If the trend continues till a May 6th polling day, Labour will win an outright majority, though probably only in single figures.

  26. Ian,

    It’s no surprise that Labour are doing better in these marginals than in the country as a whole – remember these are all Labour held seats….

  27. Quick point on the impartiality of the media.
    When I read Conservatives claiming that the British media are biased against them I splutter into my wine glass.
    Then, an hour or so later I can be heard yelling at the TV that they’re biased against Labour.
    I guess we just have to take the rough with the smooth and accept that just cos we don’t like it, that doesn’t mean it’s not true!!

  28. Anthony you are the one talking nonsense.
    The reason is that you want it to be nonsense. You are in denial about the scale of the mountain that the tories have to climb.

    You also talk about “final polls”

    But these arent final polls. There are 2 months to go yet. The final polls are not due until May.

    You spin every poll in favour of the tories but your precious tories are in trouble and deep down you know it

1 2 3