Last night’s YouGov poll for the Sun has topline figures of CON 32%, LAB 44%, LDEM 10%, UKIP 8%.

There is also a new TNS BMRB poll out with topline voting intention figures of CON 31%(-1), LAB 40%(-2), LD 11%(+1), Others 18%(+2). Changes are from their last poll a week ago. Up to now TNS have been conducting voting intention polls every month or so, but I understand this is the start of a shift to regular weekly polls.

193 Responses to “New YouGov and TNS BMRB polls”

1 2 3 4
  1. 1st hah!

  2. LAB +12 in YouGov… consistency of lead over the past few months is good sign for Labour…

  3. LAB have more support in Rest Of South (34%) then CON polling figures (32%),,, solid stuff for Labour.

  4. What the betting that the media will report the headline:

    TNS BMRB Media Alert: Labour lead shrinks to 9 points over the bank holiday weekend[1]

    rather than the article’s opening words:

    There was minimal movement in TNS BMRB’s final voting intentions poll of August 2012, which was conducted over the bank holiday weekend. Labour continues to hold a commanding lead of nine points over the Conservatives whilst the Liberal Democrats showed little movement and the smaller parties managed to pick up a couple of percentage points between them.

    All of which goes to illustrate the point of Anthony’s brilliant and no doubt universally ignored article on the previous thread, which even someone as picky as me was unable to find a word wrong with (and the perfect graphic as always).[2]

    [1] Scroll to the bottom of the page to click through for the tables. Yay I’ve finally found where TNS-BMRB hide their data. That means they must be about to move it again.

    [2] You will notice that as usual I have managed to end up commenting on a piece on the thread after, or indeed before, it.

  5. It will be good to see regular weekly polls from another pollster.

    Who is paying for them?

  6. I wonder if this idea will get very far?

    Might the powers come to appreciate Scotland, or might the Scottish electorate love the chance to see (egg) them? ;)

  7. Most of the pollsters who conduct polls on a monthly basis, or even more infrequently, have suggested no great movement between their intermittent surveys (TNS/ICM/Opinium) whereas YouGov, unless I’m seeing things that aren’t there, are hinting at an ever so slowly widening lead over the last two weeks. Anthony’s UKPR polling average hasn’t been updated for two weeks, but the last five YouGov polls that feature in his table, and which were conducted between August 8th to 16th, show a run of Labour leads of 8,8,10,9 and 9; clearly in the 9-10% average lead territory that Anthony referred to as recently as three days ago.

    However, let’s look at what’s been going on since the 16th August. The run of Labour leads in the seven polls conducted since then has been 11,10,10,9,12, 9 and 12. It seems like we’re inching slowly but surely into 10-11% average lead territory now, doesn’t it? What might be worth keeping an eye on is the occurrence of two 12% Labour leads in the last three polls. May well be something and nothing and we’ve just had two top end MOE scores almost in succession, but the last two weeks have seen, predominantly, double digit leads where, before, they were much more intermittent.

    No great surge, I grant you, but there may be something going on and I shall be very interested in the two remaining YouGov polls this week.

    One last thought. Looking at the YouGov trend, and the recent ComRes/TNS/Opinium and IPSOS/MORI polls that are more or less in line, these ICM polls are standing out like absolute sore thumbs now, aren’t they, certainly in terms of the gap between the two parties?

  8. @CrossBat,

    Re: true, but certainly not anything significant. An average 9/10 lead has turned into a 10% one – or an average of 10.1% to be more exact – by Billy’s own calculations. I agree that there has probably been a minute increase in Yougov leads but there has also been a slight decrease in Labour leads with most other pollsters.

    Very little change really.

  9. @Q/Ali

    Not seen you post on here before. Welcome aboard!

  10. @AmbivalentSupporter

    Thanks for the name check, but my posts are impressionistic at best. I’ve been averaging the headline Labour lead, so my figures probably don’t stand up to a deeper statistical analysis.

    In a way I’m keeping track of the day to day diet of “Labour lead strong… Labour lead on 10” etc type of commentary, and how that in itself feeds into the political climate (as opposed to public opinion per se).

  11. Mike N

    It will be good to see regular weekly polls from another pollster.

    Who is paying for them?

    A good question. The latest one is of VI only, so I would assume it was something tagged onto a weekly omnibus poll. In that case you don’t need a client and, if you’re asking about social issues, VI and/or 2010 vote (which they also ask) it could be useful to control your sample or to provide further analysis of it. I think Opinium’s figures come from this sort of exercise, and of course the pollster gets some free publicity from it.

    That said I can’t see anything on the TNS-BMRB website about a regular omnibus (some pollsters promote them to encourage smaller clients to ask a few questions).

  12. @Crossbat11
    “these ICM polls are standing out like absolute sore thumbs now, aren’t they, certainly in terms of the gap between the two parties?”


    On the previous thread, AW reminded us that it’s only after the 2010 GE that ICM have applied a policy of halving responses from those who didn’t vote in 2010, a policy which is so far unique to ICM. Typically that as yet untested approach reduces the Lab lead over Con by about 2%.

    I’m wondering why ICM saw the need to change their approach. The uniformity of the adjustment is what bothers me, being applied regardless of whether a party has recovered or lost support since 2010.

  13. @Roger Mexico

    “All of which goes to illustrate the point of Anthony’s brilliant and no doubt universally ignored article on the previous thread…..”

    Indeed, I wait with baited breath for the “great unwashed” amongst journalists to actually start to pay attention to the kind of advice that Anthony (amongst others) offers. I am not optimistic. The degree of downright false reporting of polls is fairly sickening. It is not simply that they do not understand, they *wilfully* (IMHO) refuse to understand – it would restrict the degree of elbow room they desire when it comes to that “eye-catching” headline.

  14. And as for ICM’s assumptions regarding the voting “dont knows”, in their latest poll they allocate no less than 37% of them to the LDs (of the three main parties) even though only 12% of those polled actually responded as being willing to vote LD.

  15. @Frederickl

    “it would restrict the degree of elbow room they desire when it comes to that “eye-catching” headline.”

    Agreed. Sadly, the ‘news’ doesn’t represent the dissemination of the information that the public should be aware of; rather the useless tittle tattle that sells newspapers and increases hit count.

  16. The polling results of Yougov and ICM are more alike than some think.

    ICM have a slightly dubious way of reallocating ‘don’t knows’ which benefits the Libs at the expense of Labour. This may have been justifiable before the last GE, but as the Lib vote has collapsed ICM’s current methodology probably is slightly more hard to justify IMO. On the other hand, Yougov don’t currently do any take into account someone’s likelihood to vote at present – something that boosts the Labour vote until they change their methodology once a GE/local election campaign is under way.

    So as people are more likely to know who they are going to vote as an election nears, the proportion of people who are in the ‘don’t know’ category tends to fall and so so far fewer are reallocated on the basis of previous voting assumptions (as per ICM). At the same time, as election campaigns start, Yougov start taking into account someone’s ‘likelihood to vote’. Thus, Labour’s vote (according to Yougov polls) tends to fall slightly.

    If you take the above facts into account, it’s only logical that as an election draws very close the results of ICM and Yougov polls does tend to converge. The number of don’t knows falls (and hence aren’t reallocated using ICM methdology) and Yougov takes into account a voter’s likelihood to vote.

    As I said earlier, if you take into account the different methodologies and how they are being applied at present, they are both amazingly similar in what they are showing. Both are showing an underlying Labour lead of around 7% to 8% (once everything is taken into account).

  17. @Phil

    I’m wondering if Peter Kellner’s article couldn’t be interpreted as a criticism of ICM’s method:

    “The Lib Dem core vote has always been tiny. They add to their support at general elections and, even more spectacularly, by-elections by attracting the tactical votes of people who identify with other parties, and a large slice of the people who don’t identify with any party.”

    This ties in with Mark Pack’s analysis that LD VI falls away steeply between GEs… it could be that they are dependent more on attracting a large new cohort each time, rather than being able to entice back those who lent their vote last time…. hence the necessity of presenting themselves at each GE as “the new politics” in contrast to “the two tired old parties”.

  18. @Billy Bob,

    “I’m wondering if Peter Kellner’s article couldn’t be interpreted as a criticism of ICM’s method:”

    I think it probably is. I don’t think he necessarily meant it as a direct criticism, though.

    I must admit I always find Peter Kellner’s – rather like AW’s – analyses fascinating. He may be a staunch Labourite but he always seems to analyse in a totally unbiased way and currently sees the likely outcome in 2015 as in the balance and very open. I must say it is so refreshing to read the opinions of someone who is able to put his own partisan views aside and write a meaningful and insightful anaysis of the current state of affairs.

  19. Just a quick question: What impact does it have if a political party has troops on the ground, knocking on doors and getting the vote out? Significant or marginal? Has nay research been done on this?

  20. Kevin

    Only thing that comes to mind is Justin Fisher’s work (he contacts the agents for all candidates after the election with a questionnaire to get data on how much campaigning work was done in the constituency, and uses that to look for effects).

    There’s a paper on it from him on the 2010 election here. Conservative campaigning strength didn’t seem to make much difference, Labour a small impact, Lib Dems more so –

  21. Kevin and AW

    Thanks for the link Anthony. I will study it later certainly. Kevin did not draw a distinction between GE, local, etc.

    I can report that in a previously sometimes walkover local Con seat, I campaigned with a few helpers (but very intensively) and raised the turnout to 67% instead of the usual average 40% over the surrounding seats and it went to me by a smidgeon.

    I retired and the seat (actually with boundary changes but this is within a politically homogeneous area) has reverted to Con with, guess what, a 40%, turnout. Using a LD paper candidate, the Con gets 67% of that turnout and the paper candidate one third. So it is now a Con cert once more.

    I conclude that if you have total targeting, you can pull off a win anywhere where the population is reasonably comfortably off.

    Surely in a GE, the national publicity has to take over, without voters being highly conscious of what is a marginal and what is not? Otherwise why get 755 turnouts in Con certainty seats? I suppose people just feel they have to ensure victory just in case their colleagues were all on holiday?

    Clegg mania (which I dispute more than a bit as a label) proves the point – it’s just awareness in a GE surely?

  22. It is no doubt true that by dint of effective organisation Labour did manage to snatch defeat from the jaws of annihilation in 2010.

    My own thoughts fwiw:

    Local campaigns bought into media narratives about the Labour leader being a drag on party fortunes. They concentrated on the virtues of their constituency candidate and touched on national issues, but studiously avoided any mention or pictures of Gordon Brown, and as a result they created the impression that there was a vacancy for the post of PM.

    Conservative campaign leaflets, in contrast, took the form of a job application, accompanied by pictures of a demure looking David Cameron, head bowed at his desk.

    As with attacks on Gordon Brown from the Sun which backfired, Labour could have been bolder: “He may be a one-eyed marker pen wielding obsessive, but he is a better bet when it comes to the economy”

  23. Billy Bob


  24. Billy Bob:

    I just dunno why they didn’t go with that line……………

  25. @Billy Bob

    If memory serves me correctly, the economy wasn’t one of the big Labour sell stories in 2010.

  26. It is being tweeted that Andrew Strauss will stand in Corby as the Tory candidate.

  27. LizH – no, it’s a rumour that Guido has plucked out of the air. It appears to be entirely unsubstantiated.

  28. @statgeek

    In ten years Brown only ever scored two negative ratings as chancellor (-3 and -13). Darling was at -15 in 2010.

    Osborne has accelerated to -30 in short order:


  29. Anthony

    Is YG now releasing polling at 7am each day, rather than at 10pm ?

    Andrew Strauss as far as I know, has no interest in politics. Think Guido has bowled a googlie.

  30. AW

    The #Strauss4Corby seems to be a popular topic at the moment.


    You are hardly comparing like with like.

    Brown was CoE (despite being CoS :-) ) during the bubble.

  32. LizH – all sorts of things are popular on twitter that are complete and total nonsense. I like to think of it as the font of all bullshit.

    RHuckle – for the time being, yes.

  33. @Billy Bob,

    “In ten years Brown only ever scored two negative ratings as chancellor (-3 and -13). Darling was at -15 in 2010.

    Osborne has accelerated to -30 in short order:”

    Taking money from people and making sweeping cuts is never going to be popular – especially if many people see them as unfair and/or not necessary.

  34. Ukip has a candidate for the Corby byelection:

    Margot Parker was previously a Libertas candidate in the 2009 EU elections, her father moved to the Corby area from Scotland.

  35. Billy Bob

    Where did Andy Sawford’s dad come from?

  36. @Old Nat

    I don’t have any information on Andy Sawyer’s dad.

    Rigel ? Orionis?


    Thought not.

    It’s understandable that, in a by election, candidates try to stress some local connection with the voters, and push that in their propaganda. All candidates did that heavily in the Inverclyde.

    I just wondered why you chose to include that particular bit of puff. Excessive copy and paste?

  38. @Old Nat

    I included it because Parker lays stress on the fact on her website. Also other posters have asked about a supposed ‘Scottish factor’ in Corby.

  39. I agree with most of the comments above. The recent polls are good news for Labour.

    This policy of keeping Ed Miliband under wraps is really paying dividends. Especially since, and I have to admit it, the Tories are making a lot of mistakes at the moment.


    Of course she does!

    Same way that the LD candidate stresses that she “has lived in Northamptonshire for more than 35 years”

    This is normal by-election puff and doesn’t need to be copied elsewhere.

    From the lack of such local connections being pushed by Labour, I presume Sawford has little – not that I think it will make any difference to his winning the seat.

  41. @AW

    Shame, hope it’s only temporary.

    @Howard, AW, Kevin

    From direct experience in the 1980s and early 1990s, local campaigning has the considerable benefit in a multi-party FPTP system of signalling to potential supporters that a vote for your party is worthwhile and not better applied tactically elsewhere.

    In the local context here, it was about keeping Labour credible as the main challenger in a Conservative parliamentary seat. First priority posters in windows/on stakes, second priority volume of leaflets. What was written on the leaflets mattered much less. Labour posters in strong Conservative areas were particularly important to avoid a tactical squeeze, and some canvassing was done purely with the aim of identifying supporters willing to display them, in areas which would never be “knocked up” on polling day. It eventually worked in a seat which could easliy have become a Con-LD contest but for such groundwork.

  42. Billy Bob:

    I’m shocked to see you have been repeating “puff” – especially about Bonnie Scotland.


    Shouldn’t “puff” be more accurately allocated to Wales? Their dragon is magic!

  44. ON:

    No idea what it even means.

    More interested in Theo W staying with the lads.


    Alas, you may to be too young to remember Peter, Paul and Mary’s recording of “Puff, the Magic Dragon” in the 1960s.

  46. @Old Nat

    I’m sure i’ve heard that… And I can assure you I was not around in the 1960s

  47. RAF

    Though, to be fair, neither of is a classical guitarist. Paul Croft probably has higher musical taste than us. :-)

  48. @AW

    “all sorts of things are popular on twitter that are complete and total nonsense. I like to think of it as the font of all bullshit.”

    Great idea. I’ll try to use that description with some of the Twit-erers I know. :)

    @Billy Bob

    Yes, GO has lower ratings. He’s the guy that got the job of sorting the problems of the last lot. Anyway, I don’t want to get partisan. Suffice it to say, during Election 2010, Labour were not talking much about how the economy got to be in the state it was.

  49. Will US Republicans see the encouragement of people to take a bite of the apple at the Paralympics ceremony as a deliberate attack on US fundamental Christianity (for want of a better term)?

  50. Arrgh! Anthony; my quoting your post has me in the moderation queue. Sigh.

1 2 3 4