The YouGov/Sunday Times poll this morning showed a four point Labour lead, interpreted in some quarters of the commentariat as showing an advance for Labour after the Paxman interviews. As ever, it was only one poll. Now we have a second post-Paxman poll, a ComRes telephone poll for the Daily Mail & ITV, and this one shows the complete opposite – CON 36%(+1), LAB 32%(-3), LDEM 9%(+1), UKIP 12%(+2), GRN 5%(-2) (tabs).

If the four point Labour lead in this morning’s YouGov poll equalled their best this year, this poll is the biggest Conservative lead ComRes have delivered since 2010. Where YouGov showed Miliband’s ratings improving, ComRes show Cameron widening his lead as best Prime Minister.

There is no great mystery here, I expect we’re just seeing normal sample variation. I said this morning we needed to wait for some more polling to have any idea whether the Paxman interviews had really had any effect, whether there was a consistent trend. With two polls now showing movement in opposition directions there certainly isn’t yet. It could still be that the rest of the week’s polls show a similar movement to YouGov and the ComRes poll was just a blip… or that the rest of the week’s polls show a similar movement to ComRes and the YouGov poll was a blip. I’ve a sneaky suspicion though that we’ve just happened to see two outliers in opposite directions, and we’re going to see lots of polls showing no clear movement. Time will tell.


442 Responses to “ComRes/Mail/ITV – CON 36, LAB 32, LD 9, UKIP 12, GRN 5”

1 4 5 6 7 8 9
  1. Here in Islington North the only communication we’ve received is a leaflet from the Tory PPC. Even he doesn’t look convinced of his chances.

  2. @Little Red Rock
    ” I don’t understand why STV has fallen out of fashion.”

    Under STV with 5 member seats the “quota” is 16.67%. Parties that can’t drum up that much support will only take seats sporadically, and will be underrepresented. Two things have changed in the political system over the last 5 years that have made some people less inclined to believe in STV. 1) the LD vote has slumped well below 16.67% 2) the UKIP vote has risen above 16.67%, especially when voters percieve no risk of their vote being wasted.

    The LDs still officially advocate it, but they’re shouting about it a lot less now.

  3. On the BES report, I should have said “proportion of non-2010 voters”. That must be confused to an extent by first-timers,

  4. @LRR & Hawthorn

    Exactly. Quit or do both is exactly my question.

    Jim has said he intends to take a second job (Holyrood) as soon as possible or 2016 at the latest:

    JM “I’ve given a commitment that I’ll be in the Scottish Parliament in 2016 and Labour’s candidate for first minister in 2016. I’d like to be there sooner than that and in terms of how we do that and what we do, of course I’ll let my constituents know first and my constituency party know first. But there’s a cast-iron guarantee to be a candidate in 2016 at those elections, if not before.”

    So I don’t think it’s unreasonable for me to ask whether he intends to be a part time MP for 4 years if he’ll be standing down and causing a by-election in 12 months.

    He’s the one who decided to run for Holyrood!

  5. UNICORN

    Another (rather morbid) thought crosses my mind too.

    Given the age profile of Conservative voters, presumably the Conservatives need to do better than have equal churn from the other parties as their 2010 VI would be disproportionately likely to snuff it in the following five years.

    The same applies to UKIP, but most of their current voters were not 2010 UKIP voters.

  6. I’m 47 and have lived through 10 GE’s and this is the first time that I ever been interested enough to visit sites like this and see what is really happening with the brass tacks of an election.

    I am so excited to see this Yougov poll tonight.

    Thanks to everyone for their opinions as it all helps me try to understand.

  7. Will the squeeze of the small parties unravel after Thursday?

    https://twitter.com/montie/status/582512661927202816/photo/1

    We saw the VI of the small parties went down in the post debate ICM poll when the public could only see the 2 main parties. Now they are introduced to 7.

    I think we will see a rise of the ‘others’ after this debate.

    Miliband does seem to have lucked out with prime position right in the middle though, it should make him look more like a PM.

  8. @Assiduosity

    What I found, if it is me, was that Brecon and Radnorshire was the 27th most marginal seat for LD, but that Lord A found when he reported out in his January 2015 poll that Con still need a 1.5% swing.

    The difficulty in assessing this now is the movement in the polls since Lord A reported out:

    LD 31% (-15%)
    Conseravtive 28% (-10%)
    UKIP 17% (+15%)
    Labour 15% (+4%)
    Plaid Cymru 8% (+3%)
    Green 2%

    Any “swingback” to Conservative from UKIP or movement to Labour from LD and this seat becomes a Conservative gain.

    What does the latest Welsh poll show? UKIP static, LD down a tad, Labour up a tad – but does that translate in this particular marginal?

    Acording to Ashcroft LD have lost 33% of their support in this marginal, whereas the latest Welsh national poll put the loss on 75.1% Wales wide since 2010.

    @ Chris in Cardiff

    I think it was Panelbase that found 50% of Welsh Greens were thinking of voting PC, and I observe that PC has just gone up to 11% and Green down to 5% in the latest national poll from 10% and 6% respectively.

    Some might call it a Green “squeeze”, but I think given Michelle Wood issuing a call to PC members to vote Green in England and that Green candidacies are more about getting seats in the regional list elections in 2016 – that 4% to 5%, up from .7% in 2010, would make the Green very happy in 2015.

    So my projection is expect PC to hang onto their three seats and possibly take Ceregidion, with support from Welsh Green swing voters.

    Green are obviously picking up support from LD and I cannot see that swing to either Conservative or Labour to knock off either a PC or LD MP.

    And since there is only one Con-Labour mariginal in Wales a Labour-Green squeeze in Wales simply does not apply.

    What I will observe is that Carmarthen West and South Pembrokeshire could come into play if the UKIP vote holds up in Wales:

    Conservative 33% (-8%)
    Labour 29% (-4%)
    Plaid Cymru 17% (+6%)
    UKIP 11% (+8%)
    LD 4% (3%)
    Green 3%

    A slight swingback from UKIP to Labour and this seat becomes a Labour gain. But this seat is the 69th most marginal seat for Con and there would need to be a swing of 4.3% without factoring in UKIP.

  9. A perfect example from the BBC of how the betting markets work:

    Bookmakers William Hill say someone has staked the first five-figure sum on the political make-up of the next government. A gambler in Wiltshire has put £10,000 on the Tories to win an overall majority at odds of 11/2. He would win £55,000 if the Conservatives won outright.

    The bookie has cuts the odds on a Conservative majority government to 9/2 – making it third favourite, after a Labour-led minority administration (2/1) or a Tory-led one (3/1).

    So there you have it – bookies odds are influenced by punters parting with their cash.

    It’s a mugs game.

  10. Any system that doesn’t have a constituency runs the risk of way too much power to the political party machines and removes the direct link of ‘being represented’.

    But a democratic mandate should insist that individual representatives can recite majority support.

    The French second ballot system does this very well. Take a look at the voting and results that have just taken place. You can vote on principle, and if need be vote tactically afterwards. And it gives a more proportional result.

    But can we really see the UK public being willing to vote twice in a week? Maybe not, hard enough to convince them to vote once, give that a miss then…

  11. Ashcroft National Poll, 27-29 March: CON 36%, LAB 34%, LDEM 6%, UKIP 10%, GRN 7%.

  12. Ashcroft National Poll, 27-29 March:

    CON 36%,
    LAB 34%,
    LDEM 6%,
    UKIP 10%,
    GRN 7%.

    Full details on @ConHome, 4pm.

  13. 36/34 Con/Lab latest ashcroft

  14. ‘BRAMLEY
    So there you have it – bookies odds are influenced by punters parting with their cash.
    It’s a mugs game.’

    Yes and no, for the astute gambler, aware of the latest polling and possible effects then it could be a good thing. Bookies seldom lose money, punters do. The clever gambler, if there is such a thing, will seek to capitalise on the stupidity of their fellow gamblers.
    For example I think the odds for Miliband being the next PM are very tempting at the moment

  15. That’s another very good poll for Con. Are UKIP being squeezed? It is beginning to seem like it.

  16. I suspect the new Ashcroft poll puts paid to the excited headlines of a ‘Labour surge’. It was nothing but sample variation. Seriously, people (including journalists and newspaper editors) know better than to base assumptions on a single poll. It just annoys me a bit to see people being misled.

  17. I will repost with the changes. It’s quite odd

    CON 36 (+3)
    LAB 34 (+1)
    LIB 6 (-2)
    UKIP 10 (-2)
    GRN 7 (+2)

    Dates 27th-29th
    N=1,004

  18. Tory lead halved in 24 hours!

    All nonsense of course. They are neck and neck and have been for months

  19. @JONATHAN

    I too think the French system has a great deal of merit but would be hard to sell in the UK. The French public are far more politically savvy than the British I am sorry to say. For those who say it is just really STV split over 2 weeks I have to say it works differently in practice. Knowing exactly who is in the run off can easily change your mind and it seems the French like the idea of making a positive choice.

  20. Postie

    The 16.7% quota is one of the features I like. A party doesn’t require 16% first preferences it just needs to creep over that line eventually. And a party with 10% plus vote share nationally would normally expect to pick up several seats unless either a) It is exceptionally unlucky and its vote is spread very evenly or b) whilst it enjoys a certain level of support it is seen as toxic by everybody else. Which, again, I like.

  21. LD and UKIP on 16 combined!

  22. @bluebob:

    “Any activists knocking on the door yet?”

    I’m in SW Surrey (Jeremy Hunt’s seat), currently ranked by Election Forecast as the most Con leading of all.

    I’ve had two Con leaflet drops: one from JH, one from candidates for local council (Waverley). Both focussed on on local issues – not a word about NHS, economy, education, benefits or immigration.

    I’ve also had a Labour leaflet – my partner, who has lived here much longer than I have, tells me this is the first time ever, that he can remember.

  23. Liked this (Stephen Bush on Ashcroft’s poll):

    https://twitter.com/stephenkb/status/582559300888973314

  24. Among the erudite statisticians / psephologists on this site, given the current stasis in the polls, what criteria in terms of number of consecutive polls, or percentage of recent polls, with what margin between the parties would, prompt them to announce:

    1) A clear lead, outside the M of E for one or other of the two major parties?
    2) A clear indication that the UKIP vote is on a decline?

  25. Mike Smithson posting on Ashcroft’s poll “England Only” (with change from 2010 by me):

    CON 40 (+0.4)
    LAB 34 (+5.9)
    LD 7 (-17.2)
    UKIP 11 (+7.5)
    GN 7 (+6.0)

    Very interesting!

  26. I wouldn’t say there has been stasis in the polls exactly. The Tories have clearly been ticking up over the last month or so and Labour have been ticking up a little less pronouncedly. Hence the picture is now looking more like roughly 34-34, from 32-33 at the end of Feb.

    On this basis I would make the Tories slight favourites to win a plurality in vote share but perhaps not by more than half a point. But who knows how the campaign will play out.

  27. Interesting little piece in the Guardian entitled “can you trust the polls”. For anyone who has been on here for a while it will be stuff they already know (MOE’s-not taking one poll in isolation etc etc) but what I didn’t entirely know was:

    a) the polls at the same distance form the election as we are now saw very minor movement to that on election day (or the result). So no more swingback last time at least.

    b) the polls on election day were all within a point of the actual result for the main two parties (just overestimated LD).

    The Guardian only chose 4 polling companies from each of the two dates and I’m assuming there were a lot more so maybe some of the others weren’t as neatly inline but seems to me if polls followed the pattern of last time then it would be neck and neck on election day.

    The difference of course this time is the rise of smaller parties and whether they fall again and that the polls we have had in the last 24 hours are very different to the almost unanimity of the pollsters 5 weeks out in 2010 (although looking at AW’s polls from 2010 it might be the Guardian just had a day where the polls were in sync but overall you wouldn’t have been far wrong on the poll averages).

  28. Unicorn

    Thanks for your link to the BES page. This was its URL.

    http://www.britishelectionstudy.com/uncategorized/the-ebb-and-flow-how-the-flow-of-the-vote-adds-up-to-a-liberal-democrat-collapse-in-may/#.VRljn_nF_l9

    In the eyeball of 2010 voters (and those who didn’t) the top two rows, ‘don’t know’ or ‘will now vote “other”‘ and the bottom one ‘did not vote’, it seems to me to scotch the notion that these categories are in any way significant in last minute decision making. In other words the chances that they could be in any way of influence on the outcome is remote.

    In simple terms, if they do vote, their choices will spread out insignificantly. We know that a good 30 – 35% of voters will not vote. I suspect that these who have recorded a VI now, would equally well be insignificant in terms of their impact in not voting.

    Common understanding is that the no-shows are likely to disadvantage Labour, but some evidence for this would be interesting.

  29. I think this is worth highlighting as well

    “Will definitely vote that way” = 61%
    “Might end up voting differently” = 39%

    Gosh

  30. @Statgeek

    Would love to know who leads these focus groups and what ‘refreshments’ are offered to the participants!

    On a more serious note, if the essentially updated views of Labour as ‘flat cap and whippets’, Conservatives as ‘exclusive toffs’ and the LibDems as ‘well meaning hippies’ are really so prevalent then why such a marked change in outcomes for the LibDems? Why would people be any less disposed to the nice folk who live in the beige house and take in strays than they were five years ago?

    Such soft data is fun, but I sometimes question how informative it really is.

    Focus groups are great ways to tease out responses to individuals, events, policies. proposals, messages in the same way they are used in the commercial sector to develop product, branding, advertising and messaging.

    But when people asked ‘if the each of the party leaders belonged to a pop group which pop group would it be?’ you know your in for more laughs than insights.

  31. “On this basis I would make the Tories slight favourites to win a plurality in vote share but perhaps not by more than half a point. But who knows how the campaign will play out.”

    pretty sensible. but seats-wise I think it’s even stevens…

    http://www.bet2015.com is very good

    Think they currently have lab and con on abt. 275, and it doesn’t fluctuate nearly as much as the polls do…seems sensible to me.

    The ashcroft poll England only shows a con to lab swing in England of 2.75%, loss of about 35 seats on a uns basis…think the swing will be 3% but let’s see.

  32. Smithson’s England Only numbers are based on a single poll (with Tories ahead) which makes them a bit meaningless in my opinion.

  33. “Smithson’s England Only numbers are based on a single poll (with Tories ahead) which makes them a bit meaningless in my opinion.”

    Interesting you say this. I have been tracking england only numbers from various pollsters all year. the average swing is about 4% con to lab….averages are very powerful as they iron out margin of error.

    The swing in England is key, as Pcrawford always says. I don’t see much change there, though the squeeze on ukip should help the tories a bit.

  34. @Omnishambles

    On the 61:39 split, it would appear that the more robust voters are in the North and Scotland, which tends to suggest, does it not, that the Labour and SNP have a more solid hardcore which will be difficult to shift.

  35. James Peel

    The weight of polling evidence suggests a much higher England Conservative swing to Labour than 2.75%.

  36. I was close with my prediction for Ashcroft just one point off, all in all while I agree it’s to early to claim a “debate bounce” for either party it might still be too early to rule one out.

  37. @aberdeenangus

    Yes. Although 2) in this summary http://lordashcroftpolls.com/wp-content/uploads/2015/03/ANP-summary-150330AZ.pdf

    … shows that while Labour does have the most solid support, the numbers for Labour, Tories and UKIP are very similar. There’s not much in it. The Lib Dems, however, are a different kettle of fish.

  38. @J R Tomlin

    I agree it tells us nothing we didn’t know. Presumably a tied poll would mean a swing in E&W of 3.75 to Labour. That would be about 45 seats changing hands I think. But well knew that.

  39. Welcome Oberta. Sorry I missed your post earlier. You’ll find that we all have different political opinions and expertise (in some cases), but we try not to be partisan.

  40. Ashcroft another weekend poll (fieldwork 27 -29) I note.

  41. “The weight of polling evidence suggests a much higher England Conservative swing to Labour than 2.75%.”

    I was referring to the England only figures in the latest ashcroft poll…

    The average con to lab swing in England in polls which have this breakdown is about 4% this year…

    I don’t think it’ll be as high as that in the election but that’s just a hunch it’s not based on data. Even at 3%, this translates to the loss of about 40 seats which is what many believe, not the kellner tory on 295 seat crowd but most other people.

    Interestingly, http://www.bet2015.co.uk which currently has

    Lab 276
    Con 275
    SNP 41
    LD 31

    Hasn’t changed that much in the last few weeks since it was launched….it simply quotes the favourites to win in each constituency.

    As a ball park, this is about right. As i have said on numerous occasions, i think the tories will be a bit below this, say at 270, and labour a bit above at 280, but i’ll be happy to be proved wrong. on a more general note I really can’t see con + ld being greater than snp + labour…

  42. Those Ashcroft focus groups do sound like fun don’t they, though it’s probably unwise to draw much from than the occasional hint or joke. I was however struck by a comment Ashcroft makes himself with regard to the election bumph that has been hitting the doormats when he mentioned “[…] the old campaigning truth that you can’t fatten the pig on market day”.

  43. AberdeenAngus

    There’s no real rule of thumb that answers your question correctly. The variations between pollsters alone can mess things up, even the assumption that the average pollster is right is a big one.

    Ultimately you have to do statistical tests.

    For 1) there is no hint of any reason to take such a test, the polls are level for statistical purposes (people normally append “my party may be a touch ahead” as standard practice)

    For 2) There have been statistical tests done with a curious result. Taking Yougov polls alone there is a significant decline in the UKIP VI. Taking all the polls, there is a decline but it’s so small it might well be due to statistical noise. There is an issue with comparing polls across different pollsters as you need to separate house effects from statistical noise from the underlying trend.

    Even then it’s not definitive. Statisticians like to quote anything with probability < 5% as significant however…

    https://xkcd.com/882/

  44. @Andy Shadrack

    I agree with you. There’s nothing on the face of it that would have the latest YouGov Welsh poll shift Brecon and Radnor to the Conservative side; however that’s how Prof Scully at Cardiff seems to be calling it in his piece for ITV.

    Perhaps as @RogerMexico posits it’s down putting more faith into detailed SVI analysis rather than the contested CVI polling methodology used by Lord Ashcroft.

    I realise that at 27th most vulnerable, this would, on paper leave the LibDems with around 30 seats, but taking Scotland into account, I feel losing B&R would presage a greater collapse in LibDem numbers and, significantly, a larger number of Conservatives in the next HoC.

  45. James Peel

    I think that was the point I was making.

    I find these Ashcroft national polls very hard to take seriously. Last summer, they had the Conservatives on sub-30. Yeah, right.

  46. So, just to be clear…despite the rise of UKIP, and despite the revival of lab, Ashcroft’s poll shows that in England only the Cons have increased their VI from 2010!?

  47. “I find these Ashcroft national polls very hard to take seriously. Last summer, they had the Conservatives on sub-30. Yeah, right.”

    I think they do have strange results. Ashcroft is the only pollster this parliament which had labour and con under 30% in the same poll. That seemed nuts.

    But you can’t disregard them totally…..

    the latest polls suggest a squeeze on the minor parties. I quote http://www.bet2015.co.uk because that shows the quickest snapshot of the individual constituency markets…

    the assumption behind their numbers is about 38 seat gain by labour from con, and an 8 seat gain by labour from ld and 10 gain by tory from ld, and snp 8 gain from ld and 27 gain from labour….

    con 303-38+10= 275
    lab 257+38+8-27= 276
    snp 6+8+27=41
    ld 57-10-8-8 = 31

    That’s the underlying arithmetic behind their numbers…it seems sensible to me, but as i say, i think the labour gain from c might be one or two more, and i think labour will get more than 8 seats from the lib dems. I also somehow don’t think the snp will win 27 off labour, but these numbers aren’t way off to me.

  48. “despite the revival of lab, Ashcroft’s poll shows that in England only the Cons have increased their VI from 2010!?”

    Are you nuts?

    labour got 28% in England in 2010

  49. James Peel

    ???

    Er, that’s what he was getting at.

  50. again as a newcomer to all of this, my reading is that there is little chance of a “right wing” coalition reaching 326, perhaps featuring conservatives, Lib Dems, UKIP and Ulster unionists, unless there is big improvement in the tory polling numbers?

    Alternatively, the “left wing” featuring Labour, SNP, Greens and Respect may just have enough to form a government?

    Sorry if that is an overly simplistic reading

1 4 5 6 7 8 9