Polls tonight

There is an interesting response from Peter Kellner here about something that looks set to be in the Telegraph tomorrow. I’ll have more to say about it later.

Tonight I’m expecting at least two new polls.

UPDATE: Almost as I hit submit, I found yet another poll – there are also new figures from TNS BMRB. Topline figures with changes from their previous poll are CON 38%(-1), LAB 33%(+2), LDEM 19%(nc). I think the fieldwork was between the 24th and 30th, so all post-budget. The other polls tonight are ICM and YouGov.

UPDATE 2: Corrected TNS BMRB’s dates to match those they quote on their website.


124 Responses to “Polls tonight”

1 2 3
  1. Secretely – and I am being honest here – I have been predicting a small Tory majority for quite a while now come May 6th (though I have kept this to myself). I certainly haven’t seen anything to change my mind.

    I think law and order, immigration and taxation, in particular, will all come back to haunt Labour in the GE campaign.

    I think Labour would do extremely well to get a hung parliament.

  2. Robin – that’s only true up to a point – it’s a well-known truism. Opinion polls scientifically carried out are the best way of predicting election results that we have. Obviously in the end, however, it would be silly if pollsters were to get annoyed that the electorate spoke in a different way from the polls!

  3. Seriously the thing is to convaluted for words. The Tory Party still use and love YG. The Daily Telegraph, and I have been a reader for 40 years, have delighted in dissing the Tory election campaign and revelled in poor poll results for the party. In blunt terms if YG come in tonight with a 3% lead the Telegraph can and will say, “there you are gentle reader, Cameron and Osborne are w….r’s. Its a very funny do.

  4. Peter Kellner has boosted my confidence in him. His explanations are music to my ears.

    It does seem that some unsavoury people are out to discredit him, but it won’t work -so roll on tonight’s YG

    ;-)

  5. I see that AR’s past vote weighting is now Lab 39%, Con 33%, LD 22%. Previously, it was Lab 36%, Con 33%, LD 23%, so perhaps they are taking account of false recall, now.

  6. Looks like Mr Kellner has his facts wrong, which isn’t helpful, especially after the damage done to YouGov’s reputation by this affair. YouGov’s weightings have certainly come in for a fair bit of criticism of late.

  7. @EOIN CLARKE
    I am delighted for once to totally agree with you Eoin, Kellners a big boy with a big job, he is not there to muck about trying to make a dying duck Labour government look good.

  8. David in France, some of that may be true, but you seem to be painting with an awfully large brush and trying to smear anyone who has ever questioned YouGov weightings. Who you were pointing at in particular I don’t know, but I don’t much like your implications.

    I have questioned YouGov weightings. I still do. They be right, but frankly, I’m still a long way from comfortable with one of their admitted practices and wonder if there are others they aren’t admitting.

    That doesn’t mean I have a political axe to grind. I AM an American, but god knows have NEVER been a Republican. Wash out your mouth, my son.

    Nor do I care much whether Labour or the Tories win (although I dislike GB because Iraq and Afghanistan but don’t see that Cameron would have done differently). I’m merely an interested observer. Personal opinion if I were laying a bet (well, I have) I’d bet on a hung parliament. *shrug*

    But I just think you need to cut back on smearing people who happen to have the nerve to question YouGov. They’re not god. A long way from it.

  9. Is it really a surprise that Tory supporting businessmen write a
    letter in support of Tory the proposal on NI?
    No doubt they all also support a cut in the Inheritance Tax proposal, wonder why?
    But wait —what about the essential need to reduce UK’s deficit?

  10. @Gooey Blob

    Peter has not got his facts wrong. Why do strange named people appear from nowhere to ridicule Peter?

    YouGove methods have a proven record. Can you substantiate your claim. If you can’t , would you leave. tata

  11. Not surprised at the business leaders supporting the NI cut. Any business employing staff is going to be against an increase onn taxation/jobs. I don’t think this is party political at all. I think if the Cons had just come out and suggested increasing it, there would be letters appearing in The Guardian saying it was a bad idea.
    For me, Labour have just got slightly wrong footed on this.
    Whether its affordable is obviously another point for debate.

  12. @RICHARD O
    Well more than slightly wrong footed Richard, the BBC news has just given Mandy a real good bum smacking. “A major political error”, Oh dear.

  13. The intervention by business leaders will backfire.

    It’s fast becoming;

    Labour V Big Business Bullies (some Con donors)

    The public don’t have any sympathy for high flying CEO’s any longer, much the same as they’ve lost faith in Bankers.

  14. Barnaby Marder

    “I am sure that any undue weighting would be against the BPC’s rules.”

    Which explains why YouGov ADMITTED it was not correcting its incorrect weighting for Scotland in UK wide polls. Yes, that produces a only a small error–but an error none the less that has to be added in when considering the regular margin of error.

    And they only admitted that after substantial pressure from people including some who are regulars on this forum and these had been weighting errors that went on for years. Now I don’t know that they have an axe to grind. I frankly doubt it. That does NOT mean that they are right or are willing to correct anomalous weightings for whatever reason (and companies CAN be strange beasts with complex internal workings).

    I simply trust other pollster such as ICM more. While there is some convergence–I just have a hard time with ANY polling company that says they know part of their weightings are wrong but can’t be bothered.

  15. Eoin – TNS use face-to-face interviewing, which almost unavoidably takes a long time (interviewers need to get to all those houses and sit there for half an hour).

    Sean – I thought Angus Reid were now members, but they aren’t listed on the BPC website. My guess is that is an error.

    Barnaby – the BPC’s main rule is that methodology must be transparent and tables released promptly. They don’t really pass judgement on methodology beyond being satisfied that something is likely to produce represenative data.

  16. Kathryn
    I see TNS BMRB are quoting a smaller poll lead for the Conservatives than Yougov. Presumably we can look forward to the same people rubbishing their polls too. Presumably they are also the same people that imaging AR have been got at as they have so dramatically cut the Tory lead too. Go on believing it will be a landslide if youmust, its unevidenced and it won’t happen, but you knock yourself out. Yawn.
    —————————–

    You will find those [ particularly the hysterical ones at PB ] will keep quiet regarding AR’s revision. Basically, it means that AR was wrong ! Their recent conversion to geographical weighting effectively reduced the in-built Tory bias. This was primarily lowering Labour vote. I am not saying that AR was doing it deliberately. But they are new here and soon realised that adjustments had to be made. So if you have to see the early AR polls in that light. There never were such commanding Tory leads !!

  17. JR Tomlin – no YouGov didn’t. Firstly, I think you are probably referring to something I said, and as it says on the sidebar of this site “The views and opinions expressed herein are solely those of the author and in no way reflect the views of YouGov plc”

    Secondly, you presumably missed my response to oldnat when he took the same meaning as you did from my original comment. YouGov’s GB weighting targets were already updated in 2009, so they were not out of date in the same way as the old Scottish weighting targets were.

  18. New graph on Guardian web-site suggests Labour below 30% in latest ICM poll

  19. @gooey Blob – “especially after the damage done to YouGov’s reputation by this affair. ”
    Not sure why you think YG’s reputation has been damaged? They spotted the shrinking lead first, and as no two pollsters agree precisely I’m not quite sure what you mean? All reasonable minded poll watchers accept polls, whether they like what they say or not.

  20. I think Peter Kneller is perfectly correct to distinquish between the group of established pollsters and the new comers. Harris is an established company but has been out of political polling for sometime and, I would speculate, revising its methodology as it goes on.

    While all the pollsters have different methodologies, AR and Optimum are new and dont have a track record. Personally I put aside whether they happen to produce results that I like or don’t. I have to vieew them with greater scepticism because I don’t have anything to judge them by.

    As Peter Kneller says, You Govs recent record has been impressive.

  21. Sean Fear
    I see that AR’s past vote weighting is now Lab 39%, Con 33%, LD 22%. Previously, it was Lab 36%, Con 33%, LD 23%, so perhaps they are taking account of false recall, now.
    ———————————-

    No, I think they are adjusting for the “disloyal” Labour voter of 2005 exactly like YouGov, for which they were heavily criticised. Of course, now that AR have done so , this will not be mentioned.

    I don’t like this “disloyal” description. They are probably the most loyal to Labour’s aims. They could not stomach their party being led by a warmonger.

  22. Just read the Kellner/Winnett exchange, very interesting

    I speak as I find having followed these polls for many many many
    years!

    Both ICM who have been operating a long time and You Gov also
    for several GE’s are to my mind the polls to watch and believe
    whatever they pronounce, whether it suits me or not.

  23. ICM – Con 38 (-2) / Lab 29 (-2) / LibDem 24 (+4)

  24. 38-29-23 in the Guardian

  25. This also means that “Others” have come down to 10% – which I think is close to normal. SNP must have gone down.

  26. Labour seem to be slipping slightly at the moment and the LDs seem to be the beneficiaries. I think Vince Cable’s appearance in the debate probably has a lot to do with that.

  27. 38-29-23 in the Guardian/ICM. This according to UKPR gives Tories 20 seats short. However, given the marginals story, they could be doing slightly better. The Con/Lib marginals might as well be a No/No for the Tories.

  28. JR Tomlin

    There are some posters on this site who discuss issue, including questions on weighting, sensibly. Then they are those who not offensive but silly (see my previous thread posting on some vodoo maths by one poster).

    However what we also have are people who are completely partisan, offensive and aggressive. They caome from both ends of the spectrum but are quite happy to attack a polling company because it doesnt agree with the straw poll they have done in the ‘Dog and Duck’ or somewhere. I believe it was that sort of person that David in France was getting at.

  29. Another 38 for Con adds weight to their mini-revival. 19 is a good number for the Lib Dems too but for me the 33 for Lab is their best result in a fair few polls – not that a single result means much out of context. My unweighted rolling average of the last 10 polls shows a 7 point lead. The opinion polls are resolute for a hung parliament.

    There’s an interesting story on the BBC website about the possibility of a Lib Dem outright win. It makes a change from the stories on here of outright Con or Lab wins.

  30. Some of the bickering comments are petty. The graphs on this site give a clear picture of underlying trends, which have far more meaning than day to day ripples. Some die-hards won’t like the long-term narrowing much. The GE is not decided by a long way.

    Re those people who now seek to rubbish YouGov. Were these same people rubbishing YouGov when they showed a large Tory lead? Smacks of ‘shooting the messenger’ when you don’t like the news.

  31. Colin

    I played around with the seat predictor and it is interesting how FPTP works for the Lib Dems. because of their even spread as opposed to the concentration of votes of Labour and Conservatives, there vote can go up a long way before they see real benefit. Then, all of a sudden, they get 400 seats.

    And before anyone says things like that happen, remember the Canadian Tories.

  32. How big would the switch from CON to LD be if folks thought LD could win & ‘keep Labour out’? – That seems to be a large component of the CON vote.

    The unremitting government bashing may lose CONs this election in an unexpected way :-)

  33. There are some partisan observers from both sides that is true enough. What I find gets a little boring is when Valerie and David In F come on here and bleat about it consistanly when we all know it’s happening and we trust Anthony to moderate.

    I don’t like either side coming on and being partisan but I wouldn’t kepp making the point of coming on here just to bitch about it……..I find that as irratating as the biased partisan comments…just my opinion. Sorry..

  34. Colin Green
    There’s an interesting story on the BBC website about the possibility of a Lib Dem outright win. It makes a change from the stories on here of outright Con or Lab wins.

    It wasn’t put on there before 1200 today was it Colin?

  35. Im not bleating, just explaining.

    It is the people who produce ridiculous pseudo-maths that get my goat.

  36. Al J @ 6-21….

    I think you are mistaken in that notion, do you have any evidence to back that theory up..?

    What do we make of the ICM poll by the way..? doesn’t it just show that the polls are STILL very varied …?

  37. Wasn’t aimed at you Pompey..sorry.

    I find you post some good stuff… :-)

  38. @ MITZ

    Woah!!! Only 5 points now??!! LOL!! This is DEFINITELY SIGNIFICANT – Labour are gonna win a MASSIVE MAJORITY – believe it!!! :-) !!!!!
    ————————————————
    ROLFLOL

    You’re sending me up, aren’t you? :-)

  39. Perhaps we could get totally bloody ridiculos and see how Labour votes flooding to UKIP will get the Greens in power.

  40. ICM poll

    Very interesting. They normally do put the Lib Dems higher than YoGov and equally the timing seems right to have picked up on any ”Vince bounce”. They also tend to put Labour a little lower but You Gov and ICM are within the same MoE. However, they does seem to be a trend of Cons static, Lab down, Lib Dem up or steady.

    The others dropping is interesting. I felt that was likely as soon as you got the debates and election coverage started.

  41. ICM is just always a little more of a gap between Con and Lab generally than e.g. YouGov. This cannot be due to anything other than a minor difference in method, namely telephoning and LTV weighting.

  42. @Al J,

    In short, I agree.

    mandelson has little to fear from attacking big business.

    Labour is not looking to win 42% of the vote as they did in 1997, frankly it does not need to…

    Following the multi-trillion crash caused by the likes of these companies pumping their profits into hedge funds, I do not think the 30% currently supporting Labour will lost any sleep over Labour bashing them…

    Tactically, there is only one member of New Labour who could havemade that attack without it being translated as class warfare and that is peter mandleson.

    Voters seem to be prepared to listen to two separate messages comming out of both parties…. they do not demand responsibility witnessed in the days of Lord Carrington.

    Thus, Cameron’s undermining of Osbone today or MAndleson’s undermining of Darling does not seem to cause the same backlash among voters it once did….

  43. @POLLY TICKS
    Forgive Al he is a very nice person, but his reason goes when Labour are threatened. Me, I am just a hunk, who roles with the punches. (I am frightened of spiders).

  44. The YouGov party identifiers in Scotland were clearly wrong and I still think that the recent correction is inadequate.

    However their willingness to change on the evidence suggests that the general allegations of bias are wide of the mark.

  45. Polls actually seem to be steadying a little around 6-8 points…..

    The best comments I’ve heard on the Big Business vs labour and NI was on R5 from Sir Digby Jones…

    he said, Tax profits by all means, tax income by all means, but don’t tax on employing people, because its irrespective of profit, so a smalll business struggleing pays the same per person as a large profitable business.

    Fair point I thought, he wasn’t saying don’t raise tax, just don’t do it this way…..

    Isn’t the biggest payer of NI the NHS ?

    I am just starting to wish it was election night to get it over with….

    Oh, I just did that vote thing online, 53% Tory, 49% Labour, 43% Lib Dem……

    made me laugh

  46. Peter Mandleson..is he still Trade and Industry ….? Or is he just the Deputy Pm…….Serious question..

    What is his title as Google gives me conflicting answers….? Anyone..?

  47. Matt – “I think law and order, immigration and taxation, in particular, will all come back to haunt Labour in the GE campaign.”

    Why on earth would law and order bite Labour on the bum??????
    Please, give me a reason why on earth a 40% reduction in overall crime could be anything but a positive for Labour?
    Remember these figures were confirmed by almost every statistical association when that dreadful oik tried to imply crime had gone up under Labour. I can point you to a fantastic article in the Economist if you like?

  48. Dont take the bate Sue.

    some people cant get beyond “I feel strongly that this is true, so everybody else must”.

  49. Strictly speaking Harris are not an “established company”. Harris Research, for whom I used to work years ago, were taken over (at least in the UK – I don’t know about worldwide) by WPP, who also own TNS-BMRB. Harris Interactive has only the most tenuous connection with the former Harris Research and it is not unreasonable for Peter Kellner to describe them as an inexperienced company in opinion polling.

  50. The Lab vs Big Business is the main reason I am not voting labour at the next election…..

    I have always believed that we need succesfull big business in the UK, succesfull big businesses equals bigger taxes equals more money for everyone. that includes through taxation….

    But why the hatred towards others based on their success or their background or even there lack of success ?

    This was only brought home to me by local labour canvesers who made some assumptions about me because of my north east accent…

    I think this coming election the results will be all over the places, LAB holds where you won’t expect them, CON gains where you won’t expect them and the other way round. UNS may mean nothing.

1 2 3