ORB’s weekly poll in the Sunday Telegraph has topline figures of CON 46%(nc), LAB 32%(+1), LDEM 8%(-1), UKIP 7%(-1). The changes since last week are by themselves insignificant, though it’s worth noting that the Labour share of 32% is the highest they’ve managed in any poll so far in the campaign. Precise fieldwork dates are not available yet, but the Telegraph’s write up says it was at least partially before the Labour manifesto leak.

Opinium in the Observer have topline figures of CON 47%(+1), LAB 32%(+2), LDEM 8%(-1), UKIP 5%(-2), GRN 2%(nc). Again, the changes are small, but reflect a narrowing of the lead and the highest Labour score of the campaign so far. The Tory lead is still extremely large, but it appears to be getting a little smaller. Once again, fieldwork for this poll started on Tuesday, so would have been mostly before the Labour manifesto leak. Tabs for that are here.

A third poll from ComRes for the Sunday Mirror and Independent has topline figures of CON 48%(-2), LAB 30%(+5), LDEM 10%(-2), UKIP 5%(-2). The narrowing is much sharper here, but that’s because it’s a different time scale: ComRes’s previous poll was conducted straight after the election was called when most polls were giving the Tories a twenty-plus point lead, so the changes here are echoing the decline from twenty-point leads to leads in the mid-to-high teens that we’ve already seen from other companies. Fieldwork here was Wednesday to Friday.

Overall the pattern seems to be a slight narrowing of the Tory lead, but it’s a case of a truly humongous lead becoming merely a towering one: a lead of fourteen to eighteen points will still deliver a very hefty majority. The election also seems to be becoming more and more of a two horse race. UKIP’s support fell sharply at the start of the campaign and only seems to have gotten worse since then and while many (including me!) expected the Liberal Democrats to increase their support during the campaign, it has yet to happen. If anything, Lib Dem support seems to be being further squeezed.

Still to come tonight we have the YouGov/Sunday Times poll. We’ve also had an ICM poll every weekend of the campaign so far (either for Robert Peston’s show or the Sun on Sunday), but I’ve no idea whether we will have one this week or not. I’m not around tonight, so will update on any other polls that emerge tomorrow morning.

176 Responses to “Saturday night polls from ORB, Opinium and ComRes”

1 2 3 4
  1. These polls suggest that the forecast of the death of two party politics in England was perhaps premature.

  2. First?

    Firstly, polls often narrow in the run up to the general election.

    Secondly, there has been more DIRECT coverage of Corbyn – as opposed to the caricture we see of him in much of the media……so, I suspect that a small number of people are coming round to him.

    Thirdly, Labour’s policy announcements seem to have been met with favourable reception – and the manifesto leak appears to have got them more coverage.

    In the near future, I wonder if the fallout from the NHS Malware will have an effect on the Tory lead.

    In itself, the effect should be minimal, but, given May’s ‘Strong and stable’ mantra (as reported here, pretty much the only slogan from any party that appears to have got through to the public), coupled with today’s stories in the press that there were cuts to spending on IT security, could mean that ‘Strong and stable’ gets a bit of a knock.

  3. If Labour can somehow push up to 35% by election day; they may manage to restrict the Tories to a lead that could be overcome by 2022.

  4. AW: Many thanks for your work, again.
    Someone has tweeted message that at this stage of the GE 2015 GE the Labour Party was on 36%

  5. @CL45

    I don’t think we can compare figures with the same stage of the 2015 GE campaign as all polling companies have significantly changed their methodologies since that time.

  6. I guess part of the excitement among poll watchers is that the post election announcement did seem to lead to an unnatural Con surge, which is now unwinding. Whether this is a ‘real’ narrowing or just a reversion to the true state following some less reliable scores is quite important, as one would mean we can start to talk about Labour’s campaign having gone OK, whereas the other would mean they are flatlining.

    One passing thought about the Conservative campaign. It is so focused on May, that it almost isn’t a Conservative campaign. You have to look very hard on our delivered leaflets to see the word ‘Conservative’, while May and strong and stable are all over it.

    Placing absolutely everything on the leader is a good strategy, so long as the leader performs and nothing untoward happens. May has never run a GE as leader, and in the previous elections she has been an incidental figure.

    While it wasn’t often mentioned, Hilary Clinton had a growing poll lead not up until the email scandal reared it’s head again. Her poll lead had already started to fall a few days earlier, when she appeared to collapse at a public event.

    If everything is pinned on May, and particularly on May as ‘strong and stable’, any hint that she isn’t physically strong is going to raise questions.

    Exactly the same applies to Corbyn, as a 71 year old (I think) but for Labour, they have a raft of policies, no one seriously thinks Corby will be PM, and he isn’t making grand claims about himself.

  7. An English team has just won the European Cup!

    Yes, Saracens beat Clermont Auvergne 28-17. A great game with some thrilling tries but really brutal in the contact areas. Billy Vunipola man of the match.

  8. @CL1945 – as @RAF implies – the really fascinating point would be to see what Labour would be on now under the 2015 polling methods.

  9. Something to watch in the coming weeks: if Amber Rudd gets the blame for outdated NHS IT systems she could be vulnerable in her marginal seat of Hastings and Rye. This is especially so as (i) the Greens have stood down and (ii) as the most prominent Remainer in the Government she is unlikely to receive any Ukip bonus votes.

  10. JC is 67.

  11. A couple of intriguing details from the Comres poll. A majority of respondents do not believe May when she says she wants to help those who are just about managing and a majority expect to pay more tax under a Tory government than under a Labour.government.

  12. @Alec

    Clearly the Tories did a lot of internal polling and have seen that leading as the Theresa May Party is the way to go! That bus is pretty funny. You can just see the word Conservatives in small type on the entrance door…if you look carefully.

    @S Thomas

    Seeing Clegg being dispatched by the returning officer is a fantasy that I have not yet indulged in. But looking at the 2015 result he is clearly vulnerable.

  13. Alec

    That’s fair comment on the campaign so far and I agree the Tory effort has all been about May. This probably explains her 2% rise in net approval in the Opinium poll.

    May 47/32 rating +15
    Corbyn 23/51 rating -18
    Farron 16/37 rating -21
    Nuttall 9/47 –rating -38

    Yes interesting findings, but you missed these:-

    Best to represent Britain on the world stage
    May 49% Corbyn17%
    Most likely to deliver improvements to the NHS
    May 28% Corbyn38%
    Best to lead Britain’s negotiations over Brexit
    May 50% Corbyn 15%
    The one I’d most want to be stranded on a desert island with
    May 26% Corbyn 19%
    Most likely to keep Britain safe from terrorism May
    47% Corbyn
    Best to look after the interests of hard working families
    May 32% Corbyn
    Most likely to raise school standards
    May 35% Corbyn 28%
    Most likely to reduce net migration to the UK
    May 39% Corbyn 9%

    So Corbyn has leads on NHS (Labour almost always do) and has a small lead on the hard working families question.

    And I think this is a real danger for Labour:-
    “A Conservative victory on June 8th is widely anticipated, with 86% of British adults saying so when forced to choose between that and a Labour victory – this figure includes 71% of Labour voters.”

    If that is mirrored in he other polls going I will be looking to increase the size of the Tory majority in my forecast.

  14. Hireton

    Sorry I seemed to lose some numbers when i posted last. They are:-

    Most likely to keep Britain safe from terrorism May
    47% Corbyn 14

    Best to look after the interests of hard working families
    May 32% Corbyn

  15. @Andrew Myers
    ‘Can anyone honestly see Corbyn polling better (32%) than Miliband (30%) or Brown (29%) on the day?’

    These polls are GB based – not UK. Under Miliband Labour polled 31.2% and under Brown 29.7% in 2010. No great surprise that Labour is up a bit compared with 2010 because the LibDem vote is so much weaker. There is very little change from 2015 , but Labour is likely to be picking up some support from UKIP and the Greens as well as from non-voters. This appears to be offsetting direct switches from Labour to the Tories.

    Were 2015 methodologies still being used , tonight’s polls would be showing Labour in the range of 32% to 34%.
    The polls are now clearly suggesting that in England & Wales Labour is polling higher than in 1983 – 1987 – and 2010. They may also be marginally higher than 2015 according to ORB and Opinium.The problem for Labour is that Tory support is so much higher – based on UKIP’s collapse. It is far from clear,however, that the residual UKIP support will favour the Tories -ie the pro-UKIP Tories may already have switched.

  16. Best to look after the interests of hard working families
    May 32% Corbyn 35%

    Got it at last!

  17. TOH
    ‘If that is mirrored in he other polls going I will be looking to increase the size of the Tory majority in my forecast.’

    Alternatively, the fact that no one expects JC to become PM could neutralise some of those poor leadership scores. Some who don’t want Mrs May to have a free hand on domestic policy may vote Labour even though they don’t think much of Corbyn.
    Time will tell.

  18. Quick note on %s. UKIP not standing in many seats will mean their outcome on the day is about 2/3 what they are polling (ie something near 5% will mean closer to 3%).
    If you consider LDEM are way off the 20% area they used to achieve and sum of NATS/others is roughly 9% then you could get

    CON 50%
    LAB 30%
    LDEM 8%
    UKIP 3%
    others 9%

    UKIP and LDEM slipping and LAB standing still in actual votes results in a higher LAB %. Others is fairly fixed so if you make a quick estimate that CON+LAB =80% then LAB ending up 50%.

    PS Thanks for the early poll info!

  19. error in above (seems the less than and greater than symbols messed up). Last bit should read LAB ending up below 30% would imply CON above 50%

  20. Cloudspotter

    “Some who don’t want Mrs May to have a free hand on domestic policy may vote Labour even though they don’t think much of Corbyn.”

    I am sure that will be true of some but overall those who are ambivalent about voting will probably be much even less likely to vote.

    As you say we will see on the 8th June.

    The one I’d most want to be stranded on a desert island with
    May 26% Corbyn 19%

    That’s the only surprise of the list. Sharing a desert island with
    “Foxy” May the huntress would seem a much more challenging situation than with “pacifist” Corbyn.

    I’d take my chances with the latter!

  22. RAF and ALEC.
    Yes, thanks for the opinion poll methodology change clarification.

    I remember in the 60’s and 70’s that the ‘Sarries’ were not a top side, while London Welsh were; back in the day.

    On Corbyn: Caitlin Moran nails it, IMO

  23. Hmm. Anecdote alert.

    Canvassing today from 2 households, Corbyn as PM no thanks but Tories going to win anyway so will vote for current Labour MP so majority not too big.

    Might help in a few seats, possibly he says clutching at a rather thin and worn straw!!

  24. BZ

    We can’t even agree on that one!


  25. There seems to be a trend in recent opinion polls internationally to understate the support for the centre right. See the state elections in Germany and also the success of Macron in France, where the polls underestimated Macron’s support by about 6%. It will be interesting to see if this trend is repeated in our General Election


    Whilst I could not possibly vote for Corbyn in the GE, if I had to choose him or May to be stuck on a desert island with, he’d win. His skills as an allotmenteer will come in handy growing food and maybe I’d even get a ready supply of jam.

  27. CHRISLANE1945

    Your quite correct but Sarries are a great side now. Successfully defending a European Cup really takes some doing. I support Harlequins but I have to “take my hat off to Sarries” today.

  28. @Alec It is very rare for me to dare question you on a matter of fact (or judgement for that matter). However, I was in Canada in the run up to the Clinton- Trump contest and obsessed with the polls which to my relief were showing a widening gap in favour of Clinton. Immediately after the Comey announcement this trend was reversed. (http://election.princeton.edu/2016/12/10/the-comey-effect/). The effect of the ‘stroke’ incident was as far as I remember also real but some time earlier. I have no idea whether she would have won without this announcement (she won the national vote anyway) but I am certain that it had an effect and possibly a decisive one,

  29. The polls suggest that Tory support is pretty solid. Labour seem to be up slightly and LD down by the same amount. UKIP voters are either staying loyal ( assuming they have a candidate to vote for on Election Day) or jumping to the Tories. The latest polls do not suggest any huge movement other than its between 2 main parties.

    In the end it will come down to who we want to lead the country as PM.

  30. What strikes me about these latest polls is how very close they are to each other. Are they all wrong, or are they all more accurately reflecting the true position, in which case by definition they should converge? Only time will tell.

    However I don’t read anything significant in the polls tightening. Its exactly to be expected, its what normally happens, and besides the Tories have, no doubt deliberately, let Labour make the running in the first half of the campaign. If you’re ‘strong and stable’ you can, to some extent, rise above the hustle and bustle on the ground. Also, the Tories now know whats in Labours manifesto before disclosing theirs, so have the option of adjusting their offering if they deem necessary.

    In the end I still think we will see May get an enormous majority, at least 150 possibly even 200. It seems to be a ‘perfect paradise’ (as opposed to a perfect storm) for her. Corbyn is deeply unpopular with all but the left of the Labour party. Labour has a manifesto which promises a huge increase in spending and many people are at best doubtful that it will be properly costed. Planning to dramatically increase business taxation at the same time as businesses are having to cope with whatever Brexit throws at them is perhaps unwise. Labour are perceived as weak on defence and security. No matter what the official party policy on the nuclear deterrent is, it is common knowledge that Corbyn is a lifelong unilateralist. Most of his MPs voted no confidence in him less than a year ago. Many of his top team are not thought of highly by the majority of voters e.g. Abbott, McDonnell. And we’re about to elect a government that will be responsible for carrying out the most important negotiation this country has done in, probably, a lifetime. I doubt many would consider Corbyn to be a skilled negotiator, including many Labour supporters. And nobody really knows where Labour stands on Brexit. All in all a disastrous offering for Labour with the voters.

    We have never voted for strongly left wing Labour governments except when the country is already in a real crisis. 1945 and 1974 are the obvious examples. We’re not presently in such a ‘crisis’ no matter what a few on the left might think.


    Nail on the head.


    I think that’s a fair summary of the situation as of now. I certainly think May should get at least a 100 seat majority but I’m a little sceptical of as high as 150, although you could turn out to be correct.

    His skills as an allotmenteer will come in handy growing food and maybe I’d even get a ready supply of jam

    I confess I hadn’t thought of that but you’re right. I was more concerned that with his “oppo” it would be survival of the fittest.

  34. Yep , I’m agreeing with Baldbloke too.

    Why do the polls tighten at this point in a campaign?

  35. Bald bloke

    Having read that I find it very difficult to believe that Labour will actually poll 30% + at this election. Will the Corbyn effect lead to a significant number of Labour voters staying at home.

  36. @Alistair Long,
    “There seems to be a trend in recent opinion polls internationally to understate the support for the centre right. See the state elections in Germany and also the success of Macron in France”

    That strikes me as a controversial categorisation of Macron. It may be correct; I know some people who think so. But it is not clear cut.

    (Macron is an ex-Socialist and a former member of a Socialist government, albeit a very rightwing – in comparative terms – member of that government. And a large proportion of République En Marche parliamentary candidates are former Socialist members. We can say relatively objectively that Les Républicains or the CDU or the Tories are centre-right parties, because there’s a lot of history and there are institutional reasons including their relationship as more or less sister-parties – notwithstanding the Tory departure from the EPP. To count Macron as centre-right, by contrast, can only be based on an analysis of his policies, which is bound to be subjective.)

    Anyway, the fact that opinion polls are not allowed during the final day or two in France, coupled with Le Pen’s poor performance in the final debate, and the fact that in the second round he had no candidates to the left of him are all relevant factors.

  37. One factor that I had always wondered about in online polls is the probability of a user pressing the wrong button in answering a question. This article on Nate Silver’s web site shows that such mistakes can cause people analysing data to draw erronious conclusions.


    No-one would have noticed the original paper if it had not been for the Donald trumpeting it in his election campaign.

    There is a good deal of work in, for example, the chemical industry on human errors of this kind:


    Kletz reports that he found that he pressed the wrong button on a beverage vending machine about one time in 50 (p25).

    So I have a question for AW. Do you know what the human error in answering your online questionnaires is? Presumably it is smaller than your statistical errors, but is it a little smaller or much smaller?

  38. TOS

    I think that, amongst the undecided and those toying with possibly voting Labour, in the end most of them will vote Tory as it will be seen as the lesser risk choice and maintains the status quo. I just cannot believe that Corbyn’s Labour will get close to the voter support that either Brown or Miliband got.

    I also believe that when faced with the ballot paper, almost everyone who voted UKIP previously because of their anti-EU stance will find themselves voting Tory this time in order to secure Brexit. And that is most UKIPers. Hardly any former UKIP voters will return to Labour because most of those who went to UKIP in the first place did so because of their views on immigration, and on that Labour still has very little to say. Corbyns previous statements on immigration are well known and not exactly attractive to anyone who has voted UKIP in the past. So I think the UKIP collapse will be even greater than currently the polls are showing – and it will virtually all go to May.

    Which all points to an enormous Tory majority.

  39. I posted this on the previous thread, but it seems more appropriate here.
    All the’ most recent polls are showing UKIP being squeezed further. This tranche of voters seem to be splitting more evenly between Lab and Con. It will be interesting to see the tables.
    There has been much speculation here about the destination of UKIP voters who discover they can’t vote for their preferred party. Are these latest polls an indication that they shouldn’t automatically be added to the Con column?
    It is possible that Con leaning UKippers could have left long ago, perhaps the remaining ones are not keen on Con.’

    Comparing the tables of this week’s and last week’s Opinium polls, there isn’t much evidence of GE15 UKIP moving to Labour, but rather GE15 Labour voters who had gone to UKIP a week ago returning.
    Labour’s slight improvement is due to retention increasing and some switchers from Lib Dem.
    However, Labour can not win capturing without former Conservative or UKIP voters and there is no sign of that.

  40. “in the end most of them will vote Tory as it will be seen as the lesser risk choice and maintains the status quo.”

    That *might* be true but the only thing is, I remember believing the same thing with respect to the Brexit referendum and it wasn’t the case there.

  41. Baldbloke – I can’t disagree with much that you say, including the desert island bit! I can’t see Corbyn polling as much as Brown did even with the LD collapse. Maybe 27% on the day but possibly less? Tories should pick up at least 8% from UKIP to put them on 46% minimum. Gut feel says a 19-20 point lead on the day still seems possible.

  42. It’s odd. I don’t see myself as particularly left-wing, more so I know than many on her, but still not one who would relish the sight of capitalists swinging from lamposts. And I didn’t vote for Corbyn and thought and think that the Labour party was foolish to make him keader. But for all that I have liked the way he has conducted himself so far in this campaign, I like the policies, although I think there are too many of them, and I gather that individually they ae popular with people as well. I don’t think that from his point of view things are hopeless and I am interested in what he might do to turn them round.

    Personally I think he needs four things: a crisper, more concrete way of putting across his policies, a more definite stance on Brexit, a ckear and convincing account of how he is going to pay for all this, and a counter-narrative that while Mrs May may be a nice, if rather wooden, Vicar’s daughter, she is also, as Mrs Merkel implies, a seriously deluded one, and as such a danger to herself and others,

    If I am not the only person on this site who believes these prequisites can be met I am probably close to being the only one. So as everyone is always wisely saying ‘time will tell;.

  43. Well if it came down to being stuck on a desert island I’d go for May over Corbyn.

    There doesn’t seem to be much meat on Corbyn and I’d feel less guilty eating a Tory!


  44. LeftieLiberal
    “One factor that I had always wondered about in online polls is the probability of a user pressing the wrong button in answering a question.”

    Wouldn’t errors tend to cancel each other out? i.e. for every ‘Yes’ that was meant to be a ‘No’, there would be a good chance of an opposite error by someone else?

  45. https://twitter.com/britainelects/status/863492244300550145

    Westminster voting intention: CON: 49% (+3) LAB: 31% (+1) LDEM: 9% (-2) UKIP: 3% (-2) (via @YouGov)

  46. I’m pretty sure that’s the lowest we’ve ever seen UKIP (in the YouGov)

  47. Candy
    That backs up the polarisation that AW was talking about. Assuming LibDem and UKIP can’t go much lower, Labour would have to convert 9% to achieve equality in %ge. This is nearly 20% of Tory voters! A bit unlikely, though they would probably win even if they achieved a few % less than the Tories.

  48. P.S. Cartoon from 1872 (I think from Punch), showing Disraeli being questioned by Lord Abercorn about Conservative policies:


    Mrs May is channelling Dizzy.

  49. Peter Cairns

    “I’d feel less guilty eating a Tory!”

    Especially after the tweets by that new Stirling Tory councillor!


  50. Opinium have analysed their ottish crossbreaks (they obviously don’t follow the UKPR mantra!).


    “To model a situation that our polling suggests is possible, how well would the Tories do if we took half of the vote of the other unionist parties at the last election and added it to the Conservative total? What we found is that there could be 5 Tory gains, one of them being the last Labour seat in Scotland.

    This shows how the simple swing information is not enough to adequately tell the story in Scotland. Berwickshire is a very thin marginal anyway, but here and in harder to reach East Renfrewshire there are big unionist votes to squeeze for the Tories – In short, they are seats where the SNP won almost purely because of a divided unionist vote.

    However, in seats like Perth and Moray the SNP previously had half the total votes cast meaning the Tories can go on squeezing what few Labour and Lib Dem voters there are and never catch up with the SNP.”

1 2 3 4