Kantar have published a new voting intention poll ahead of the budget, the first I’ve seen from them since the general election. Topline figures are CON 42%, LAB 38%, LDEM 9%, UKIP 5%. Fieldwork was between last Tuesday and this Monday.

This is the first poll to show a Conservative lead since September and the largest Tory lead in any poll since the election. As ever, it’s best to look carefully at any poll that shows an unusual result before getting too excited/dismayed. The reason for the unusual result appears to be methodological, rather than from some sudden Tory recovery, and down to the way Kantar treat turnout. As regular readers will know, many polls came horribly unstuck at the 2017 election because instead of basing turnout on how likely respondents said they were to vote, they predicted respondents likelihood to vote based on factors like their age and class. These methods assumed young people would be much less likely to vote, and produced large Conservative leads that ended up being wrong. Generally speaking, these socio-economic models have been dropped.

At the election Kantar took a sort of halfway position – they based their turnout model on both respondents’ self-assessed likelihood to vote, whether they voted last time and their age, assuming that older people were more likely to vote than younger people. This actually performed far better than most other companies did; Kantar’s final poll showed a five point Conservative lead, compared to the 2.5 they actually got. As such, Kantar appear to have kept using their old turnout model that partly predicts likelihood to vote based on age. The impact of this is clear – before turnout weighting Labour would have had a one point lead, very similar to other companies’ polls. After turnout weighting the Conservatives are four points ahead (the full tabs and methodology details are here).

(Another noticable difference between Kantar’s method and other companies is that they use the leaders’ names in their voting intention question, though given there is not nearly as much of a gap between Theresa May and Jeremy Corbyn’s ratings as there used to be I’m not sure that would still have an impact.)

633 Responses to “Kantar- CON 42, LAB 38, LDEM 9, UKIP 5”

1 2 3 13
  1. Although they say they weight by “region” among other factors, the lack of crossbreaks by region removes a useful check on whether aspects of the poll are broadly in line, or wildly out with other polls.

  2. Good news at last … a poll showing a Tory lead.

    A few more like this and it might tempt the Tories into calling another election :-)

  3. Interesting poll from Kantar. Coming on top of the other two latest polls there will no doubt be some quiet satisfaction for May. The Kantar poll shows very little change from the GE.

    Reading many of the posts here recently I suspect that UKPR is very skewed and does not reflect how the majority of voters feel about UK politics or Brexit. IMO of course.

  4. Cons on a Roll.. :-) :-)

  5. TOH

    I think most of the 52% seem to occupy the BBC Have your say (HYS) comments boards for what I have seen. I would hazard a guess that many of the posters on here come from the 48% – not all of course, as we have to beware of tarring everyone by the same brush.

    I’m not saying the standard of debate on HYS is infantile and bigoted. I would not want to make sweeping generalisations about the standard of education and understanding of the issues amongst 52% of the population but it’s just an impression….

    All IMO of course.

    Still struggling to see any good news or any upsides to Brexit beyond vague platitudes and pie in the sky nonsense.

    Perhaps you could cheer me up TOH?

  6. So Lab lead of 1% in reality with Kantar then!
    The last General Election under estimated Lab polling not Con!

  7. Really interesting insight into how turnout modelling can affect the data.

    Also worth noting that the weighting gives a small but clear undersampling of Remain voters.

    Shall we discuss the Scottish raw subsample with the Tories on 41% of voters expressing a preference and the SNP and Labour tied on 23% each, or shall we let Howard nurse his fantasies for a little longer ;)


    From the previous thread

    3. All countries are independent and organise their economies differently even though we have harmonization of things like consumption tax (sort of) no one says to Hammond don’t build 1 Million houses, no one tells Macron don’t strip labour laws to regulatory minimums All the rules we have are agreed ans some of the silly ones are actually proposed by the UK

  9. @TonyBTG

    Agreed. It is clear that this poll shows that the only rational course to allow the Tories a majority Government is a General Election at the nearest opportunity


  10. Given the state of the Tories, cack-handed policies and Mayhem at the helm this is incredible, even with adjustments. Post election Corbyn bounce appears to be over.

  11. The likelihood weighting turns a 39-38 Labour largest vote share to a 42-38 Tory vote share. I find the weighting issues very interesting

  12. John Curtice on the generational divide. V. interesting.


    tl;dr: long term, the Tories are in MASSIVE trouble (as everyone knows) that may no longer be avoidable.

  13. @Passtherockplease

    Weighting is such an interesting and tricky subject, It’s basically impossible to get right and eminently possible to get wrong.
    I think it’s good that the different companies are trying different approaches.
    That said, though, Kantar were wrong last time.

    know they weren’t as wrong as many competitors, but they were still wrong and I think it is almost certainly wrong to assume that in the current context that the young vote turnout at the General Election hit a high that will not be repeated as Kantar are assuming in not changing a model they know to be wrong.

  14. @Andrew

    I think we’re mired in cultural divide right now. This isn’t about policies at the moment. This is about your tribe.

  15. I can’t believe they’re weighting to the 2015 general election and not the 2017 one? Irrespective of whether the 2015 weighting model gave a good result, turnout patterns in 2015 are now irrelevant – grounding results on turnout patterns in an election long past, and where political engagement has changed massively since, is extremely naive. Speaking as a Statistician, in any other scenario it would look like a researcher was specifically going out of their way to pick a methodology that gave them a result that they liked.

    Their Welsh and Scottish subsamples (both showing Tories well in 1st place) are so bizarre that I’d also question their sampling. Obviously cross breaks have a huge margin of error but when a sample of 172 Scots has Cons on almost twice the SNP score, it doesn’t inspire confidence.

    In fact I’d argue that their (measured) success at GE2017 was coincidental – their methods happened to produce the right result purely by chance – a result of the bizarre contortions which they employed in their methodology. They didn’t reach it in any valid way, like YouGov’s MRP model or Survation did. And yet, even then Kantar estimated the Tory lead to be twice what it was.

    Looking at the crosstabs, it’s still showing mostly the same trends as other polls in terms of churn between parties – small net swing from Con to Lab, LDs and Labour trading votes, more Cons as DKs than 2017 Lab voters, etc.

    Looking at their result before weighting, it’s 39% Lab, 38% Con, 9% LD, 5% UKIP, 3% Green. These are the figures which you want to look at if you want to compare with other polls. Other than that, the actual poll’s result is of little consequence because it uses methods which refer back to a state of politics that just doesn’t exist anymore.

  16. I wonder if Labour’s relatively poor polling performances have anything to do with Corbyn donning the invisibility cloak that he seems to wear between campaigning – whether for party leadership or PM?

    I’m fairly interested in current event/politics and I can’t think of very much that he has said.

    Of course the media could be to blame but, either way, it makes judgements about actual general elections pretty difficult to arrive at. That may change when either one or both of the current main party leaders are no longer in place – which will almost certainly be before the next general election.

  17. he problem for pollsters is that just because turnout by age in 2017 didn’t conform to their 2015-based prediction, doesn’t mean turnout by age in the next election will reflect turnout by age in 2017. It may go back to 2015 levels, or somewhere in between, or somewhere else.

    Or put in another way, changes in voting preferences are relatively easy to pick up, it’s a zero-sum game. Changes in turnout is much more difficult when basically all likely voters are saying 10/10 or 9/10, and you have to guess whether this year’s 9/10 is a stronger indicator than last year’s 9/10…

  18. PAUL

    Yes I agree with that. I don’t think this Kantar poll is particularly credible though when it’s weighting’s totally ignore what happened in June. Labour are well positioned for the next General Election although by no means certain to form the next Govt.

  19. Good evening all from a mild Winchester.

    I don’t believe for a second the Tories are ahead of Labour and worse…I don’t believe for a micro second the Lib/Dems are on 9%.

    It’s one poll and although the Tories might get a mini boost post budget, I rather suspect normal service will resume by December.

    Conclusion. ..Nothing to see in kantar..move on.

  20. ” and does not reflect how the majority of voters feel about UK politics or Brexit. ”

    It’s a blog on polling. That’s a pretty niche interest! So no, it is never going to be representative.

  21. A minor point re this Kantar poll, but naming the leaders only works if you know who they are!

    Kantar still naming Kez as Scottish leader two and a half months after she resigned, shouldn’t give anyone confidence that Kantar even know what year this is!

  22. oldnat

    At the time they did the poll they probably thought Kez was a better option than:

    “one or other of these two blokes you’ve never heard of – once the unions have made their choice.”

  23. Goodish article on Slugger today from ex BBC Brian Walker: Brexit and the border is widening the gap between London and Dublin and depressing further the chances of a return to Stormont

    Three non-contiguous sentences summarise the article pretty well:

    Leo Varadkar is due some sympathy. He has to ride EU and British horses pulling in different directions at once but he can hardly afford to be less zealous on the border issue than Michel Barnier.

    The same article contains a report that HMRC officials appearing before the Commons’ Public Accounts Committee have admitted that they are prevented from planning scenarios for border checks by the absence of ministerial authority.

    Even if the EU is impressed by a nebulous and conditional British offer of more money towards the Brexit divorce bill, the future of the border remains a barrier on progress in the negotiations, as the EU negotiator Michel Barnier again made clear yesterday.

  24. I have no idea if the Kantar poll is accurate or not it doesn’t seem to be saying anything out of the norm anyway given the last 4 polls are 2 Lab 1 tie and 1 Con.
    What I do find amusing is the number of those on the left lining up to denigrate a poll that shows Lab behind , you rarely see those on the right attack polls that show a Lab lead. Maybe the cult of Corbyn is on the wain or maybe despite all the wishful thinking by some, both political parties are pretty even and the public are content to wait till after brexit to show any real interest in who governs.

  25. Paul Croft

    They could reasonably have said “Vacant”. That’s a pretty good description of the post, whether there’s a branch manager in place or not.

  26. As for polls in the HoC ce soir it appears the tricky CON one has been fudged and the other ones have seen CON wins, lowest margin being 10votes.

    I’m keeping track of the rebels on both sides as+when they update but since Grieve pulled his amendment we haven’t had a clear sign yet of how many CON MPs beyond Clarke and Soubs are prepared to break whip over Brexit. If/when it comes to a confidence vote that is a whole different ball game but one step at a time!

    if you missed the link to where to find the names of the rebels then repost here:

  27. @ COLIN (from last post) – people want a number. I don’t know how they get to 38-40bn. The 20bn was clearly 2 more years of full payment with a bit of rounding. The 38bn sounds like someone has given it a little thought. 40bn a rounded 38bn?

    @ PRTP – ??

    OK, I can’t resist – LDEM resurgence? Remainers flocking back to the only party that actually wants to Remain in the EU?

  28. TURK

    I rubbished the poll but I’m assuming you’re not accusing me of being on the left. Just because people disagree with this poll and moreover, bash the Tories, doesn’t mean they are on the left.

    Now moving on…What about those on the right who are backslapping each other on a poll showing a small Tory lead!!

  29. So. After voting that animals aren’t sentient, and so have no rights, MPs have voted to make sure that humans don’t have fundamental rights in the UK either.


  30. AC

    “I rubbished the poll but I’m assuming you’re not accusing me of being on the left. Just because people disagree with this poll and moreover, bash the Tories, doesn’t mean they are on the left.”

    I get the feeling that TURK feels everyone is a rabid left winger, even TOH.


    “I get the feeling that TURK feels everyone is a rabid left winger, even TOH”

    I’m not surprised. Turk is so far to the right he makes Trump seem like Chairman Mao.

  32. Norbold
    Turk still hasn’t told us how Mrs Turk got on dealing with the cottonmouth in the hen house!

  33. I think you are all being very unfair to Turk.

    His pastoral images are quite relaxing. When he mentioned “Corbyn on the wain”, I immediately thought of this


  34. @ Paul Croft

    ‘I’m fairly interested in current event/politics and I can’t think of very much that he has said.’

    I suspect that you do not look at Jeremy Corbyn’s twitter feed or facebook page because in the last 24h, it seems that:

    Corbyn has requested Philip Hammond to earmark 1bn for sprinklers in tower blocks

    Demanded that the gov’t acts to end the suffering of the Yemeni people

    Commemorated Trans people who have been killed by hatred and called for ending transphobic violence and supporting equal rights and justice for trans people.

    Speech calling on Tories to use Wednesday’s budget to end cuts to the education system…. and spelling out Labour’s plans.

    LP broadcast which included Britain’s productivity crisis, Brexit and neglect of Britain’s manufacturing base…. very much aimed at challenging Hammond over tomorrow’s budget.

    OK I’m bored of checking now but I’m sure that I could find more online …. but as you suggest for the lack of voice, ‘the media could be to blame’ …..

    And I very much agree that the absence of media coverage ‘makes judgements about actual general elections pretty difficult to arrive at.’ So much for the media’s participation in the democratic process.

  35. @ToH You say that a lot of people are very happy we are leaving. This is clearly true. For good measure there does not seem to me much evidence that anyone is changing their mind on whether Brexit is a good or bad thing. That said there are some groups such as sheep farmers for whom Brexit does threaten serious economic loss. Chris Riley appears to have particular experience of their problems and so his views on this should be welcome. To your credit you have always said that Brexit would bring short-term economic difficulty but in your view this should be short-term and was a price well worth paying. I would have thought that you would accept that these short-term problems would affect some groups more than others and that it was incumbent on people holding your views to recognise these difficulties and see what could be done about them,

  36. I think the polls are derivatives of quite pessimistic outlooks (half of the respondents think that that the worse will come in the near future), the lack of trust in government and parliament (dropping by 7% in a year) and the despair over public services (something like 70%).

    These all pull to different directions (or just let the reins loose).

  37. SYZYGY

    As you rightly acknowledge I did add the caveat about the dreaded media.

    However, as it is obvious that the majority of the population wait for news to come to them, rather than seek it out, it does seem to me that there must be something that the Labour leadership can do to at least try and prompt coverage.

    I AM interested in such things but, despite using the internet a lot, I don’t follow anyone’s twitter feeds – I have enough with Trump’s being publicised each day – and I don’t expect many of the older people we need to attract do so either.

  38. laszlo

    “Despair” is the appropriate word I think. I really believe the system is close to broken.

  39. @ Paul Croft

    I’m in no way criticising twitter avoiders….. and I’m sure that the Corbyn team try hard to prompt and would welcome more mainstream coverage.

    Nevertheless, my experience from twitter is that he says a great deal because I see, read or watch Jeremy Corbyn utterances most days. And I would guess that many of the younger generation do too (as well as oldies like me).

    As you suggested, the absence of mainstream media coverage ‘makes judgements about actual general elections pretty difficult to arrive at’ . Large cohorts of the electorate are receiving very different streams of information …. and I think that the 2017GE demonstrated the impact of that. The Labour Summer documentary was notable for me, in that the four MPs featured seemed oblivious of the organising and activism that was happening online. (Actually, they didn’t seem to know what was happening in their own constituencies either …… ).

  40. For all the opposition hyperbole about the withdrawal bill the government remains undefeated and i have a sneaking suspicion that Jezza is riding shotgun on the brexit stagecoach . Where is all the fluff about a hard time in the Commons now? the new pathetic yelp from those who oppose the democratically expressed view of the people is proudly to say that the undemocratic house of Lords will destroy it.The last bastion of the “great and the good “which in any decent country ought to have been swept away 100years ago is recruited against the will of the people.
    Well i strongly suspect that they will huff and puff and open themselves up to ridicule by the public before having their meaningless little rebellions squashed once more. Having said that one good may come of it- their demise.

  41. Charles

    I too have noticed that Chris Riley has an affinity with sheep.

  42. Catching up –

    Chris Riley (8:04pm)
    Re your link – one reason that younger voters are more in favour of immigration than older ones is that more younger voters ARE immigrants. It’s a bit like those surveys that show that areas with high immigrant populations are more in favour of immigration. This is not a sign that people are more tolerant when exposed to more immigrants, but that more people in those areas ARE immigrants themselves.

    Syzygy (10:50pm)
    “I suspect that you do not look at Jeremy Corbyn’s twitter feed or facebook page because in the last 24h, it seems that:”

    I know these things are very trendy, but do we know how many people actually get their information from there compared to (say) newspapers (paper and online), TV news, radio and so on?

    “So much for the media’s participation in the democratic process.”

    Does the phrase ‘the media’ not include Twitter and Facebook?

    G’night all.

  43. PeteB
    The broadcast media (tv, radio ) and print media ( newspapers, magazines, advertising posters) are by and large controlled as to their output by very few people, usually very rich individuals or state apparatchiks.

    What is different about social media is that pretty much anybody can send out information. Some of this is so called ‘narrow casting’, but most is effectively broadcasting via shares, re tweets, re posts etc. It feels different to conventional media, and is,in that the old ‘gatekeepers’ are not there. No editors, reporters, proprietors etc.

  44. RJW

    I think Mr B knows that – it just didn’t suit his purpose to acknowledge the fact.

  45. RJW

    I think Mr B knows that – it just didn’t suit his purpose to acknowledge the fact.

  46. Given that weighting by place of birth is unusual in polling (except sometimes for London and YG Full Scottish polls) they are normally blind to that aspect of the demographic.

    I understand, from what Anthony has said previously on the issue, is that the numbers born “elsewhere” are too small to make a difference.

    However, it would be interesting to know if there is even the remotest possibility that Pete B’s claim as to the age distribution could have an effect.

    While my instinctive reaction is that probably has little basis other than the fantasies of those who want to limit/stop immigration to England, it would be useful to have some evidence, one way or another.

  47. I am fully aware of the difference between social media and more traditional news outlets, whether those are controlled by proprietors or BBC apparatchiks.

    My point however was that despite declining consumption of traditional media, I suspect that more people read the Daily Mail or watch the BBC news than follow Jeremy Corbyn’s Twitter feed. I am of course open to correction.

    This is really g’night all, as I have better things to do at this time.

  48. Pete B

    Actually, let’s usee the DM as an example. According to http://www.newsworks.org.uk/Daily-Mail its daily readership is 3.05m. Over two-thirds of that readership is 55 or older.

    According to twitter JC has 1.6m followers. I don’t have demographics for that group but I am will to bet that a good majority are under 55.

  49. An interesting radio programme from last night, titled ‘Weakened Prime Ministers’, comapring today to Lord Rosebery in 1894:


  50. ROI
    If the FT is right then the ROI us going to receive a lesson in realpolitik.The recent more aggressive stance by them was no more than the desperate shouting of a drowning man.
    i very much fear that the ROI is going to find itself in the EU like a schoolboy whose elder brother who protected him has now left the school.And we all know how that ends.

1 2 3 13