So, here goes – the eve of the election means we get the final call polls. We already got Opinium’s final poll yesterday and Ipsos MORI won’t be till tomorrow, but everyone else should be reporting today.

ICM have tended to show the strongest leads for the Conservatives during the campaign – their final poll for the Guardian continues that trend with topline figures of CON 46%(+1), LAB 34%(nc), LDEM 7%(-1), UKIP 5%(nc), a Tory lead of twelve points. Fieldwork was yesterday and today. Note that these are preliminary figures and that ICM are continuing to collect data through the evening, so they will confirm final results later. The tables for the preliminary results are here.

ComRes for the Independent have final figures of CON 44%(-3), LAB 34%(-1), LDEM 9%(+1), UKIP 5%(+1). Fieldwork was between Monday and today. Along with ICM ComRes tend to show the largest leads for the Conservatives, and the ten point lead is actually their lowest of the campaign. Tables are here.

Surveymonkey for the Sun report just a four point lead for the Conservatives: CON 42%(-2), LAB 38%(nc), LDEM 6%(nc), UKIP 4%(nc). Fieldwork was Sunday to Tuesday and changes are from a week ago. Surveymonkey aren’t a BPC member so I don’t have more details, though we should be getting some later. Regular readers will remember that Surveymonkey polled at the last general election and got the Conservative lead right, albeit getting both main parties too low. There are more details of Surveymonkey’s approach here.

Panelbase have final figures of CON 44%(nc), LAB 36%(nc), LDEM 7%(nc), UKIP 5%(nc), GRN 2%(-1). Fieldwork was between Friday and today, and obviously shows no substantial change from their previous poll.

Kantar‘s final poll has topline figures of CON 43%(nc), LAB 38%(+5), LDEM 7%(-4), UKIP 4%(nc). Fieldwork was between Thursday and today and shows a narrowing of the Tory lead to just five points – Kantar have previously tended to show larger leads. Note that there is a very minor methodology change here, Kantar have fixed the share of the 2017 vote coming from 2015 Conservative and Labour voters at 61% – I’m not sure exactly what that means, but it has only a minor effect anyway, increasing the Tory lead by one point. Tables are here.

YouGov‘s final poll for the Times has topline figures of CON 42%, LAB 35%, LDEM 10%, UKIP 5%, GRN 2%. Fieldwork was Monday to today. Minor method change here too – adding candidate names to the voting question, and reallocating don’t knows using past vote (which knocked down Labour support by just over a point). Full details here.

Survation‘s final poll (using their phone methodology, rather than their online one) has topline figures of CON 41%, LAB 40%, LDEM 8%, UKIP 2%, GRN 2% – the one point Tory lead is the closest we’ve seen, though effectively the same as Survation’s last poll. Fieldwork was Monday and Tuesday and tables are here.

BMG, who haven’t polled since back in 2016, have also put out a final poll. Their topline figures are CON 46%, LAB 33%, LDEM 8%, UKIP 5%.

909 Responses to “Final eve-of-election polls”

1 13 14 15 16 17 19

    It’s not just 18-24’s. It’s everyone under 45.

    That is what will make the difference, or not. We’ll have to see later. If turnout in the under-45’s is up a uniform 10%, it’s hung parliament territory.

    Of course it may not happen. The anecdotal evidence is currently promising, but who knows.

  2. Roll a Hard Six –

    Not out yet. Will put up a new post once they’ve arrived and I’ve digested them

  3. I’m sticking with BBC – although not the same without Peter Snow.

  4. Some anecdotal hints of a larger than expected turnout? But this may not to be Lab’s benefit.

    Thinking about the polls before 2015 GE and the exit poll…if I remember correctly it as only the exit poll which indicated a Tory OM. This came as a surprise to everyone including the Tory party leadership.

    The thing which angered me about the 2015 polls – particularly the daily YG polls – was that these made everyone believe that a hung parliament was going to be the outcome. IMO, this perception meant that the parties prepared their manifestos accordingly. In the case of Lab, EM has said recently that this current Lab manifesto is the sort he would have liked to have offered at the 2015 GE.

    Instead, the 2015 Lab manifesto was ‘safe’ and lacking in socialist policies. I attribute this to the deceptive series of YG polls showing Lab were closer to Con than was really the case. But would a more socialist manifesto in the 2015 GE have achieved better Lab polling on the day?

    Another Con govt looks likely. The size of the OM will be important in determinin how TM pursues Brexit and whether she faces down dissenting elements in her party and indeed across the media (ie Sun, and DM). A small OM will mean TM will be weaker than before the GE ws called.

    As regards Lab…in the event of a Tory OM I suspect we will see another left of centre party arise. It may mean the fracturing of Lab.

    The LDs…their coalition venture with the Tories has destroyed them. But the remnants may well be minded to join a new party.


    SNP? No idea.

  5. My guess is a Conservative majority of 40-60 seats but with an electorate as volatile as it is nothing would surprise me including a hung parliament or a majority of over 90 .

  6. Tories have fallen in this poll and in fact most polls any drop in labour hasn’t gone to the Tories so we have to be careful in using ‘swingback’

  7. Turncoat

    That’s the opposite of the point I was making. I too anticipate that some people who would otherwise vote Labour just won’t be able to bring themselves to vote Corbyn in the booth.

    But Ipsos’s evidence shows just the opposite. Support for Jeremy Corbyn, to be PM, is at it’s highest level ever, and May at it’s lowest. Which suggests this kind of thought process would be less likely to happen.

    But as I said I still predict a 10 point tory win, around 370 seats.


    Young people tend to work or attend university. If there is a real surge in youth turnout, it’ll happen after 6pm.

  9. edge of seat

    “However you want to approach forming a government, the point of a democracy is surely to reflect the opinions of the voters?”

    Well, clearly not if what you are happy with is perpetual Tory government………………….

    ps I like the idea mentioned above about individual parties that offer “pre-formed” coalitions. You then know what you are voting for, in a broad brush sense, but can influence the flavour to your own taste.

    Unless an alternative coalition gets most support which would also be fine by me.

  10. Having bet the SNP to get 46 or less, and the LDs to get 10 or less, I find very widespread support amongst UKPR posters.

    Are you guys herding?


    they will be in the pub by 6 or the students union. I love the straw clutching but give it up now .. its gone or it was never there. Corbyns strategy of beating Millibands vote share I think will work so he will stay where he is .. biut he will lose by 80 I reckon. Extra votes in safe labour areas wont mean more seats.

  12. @Magosh

    Matt Singh had three different models to explain VI in his epic post before the 2015 election.

    One of them was the “Best PM” ratings.

    But one of them was using real votes (as in the local elections in the year before the general election).

    I think he used the assumption that voters like to give the govt a kicking in local elections, and how much of a kicking predicted the general election in the following year.

    In 2014 Lab got 31% of the projected vote, with Con on 29%, which he pointed out was the best a governing party had done since 1982 (the ’82 results predicting the 1983 GE win).

    In 2016, Lab got 31% of projected vote, with Con on 30%.

    That points to an improvement in the Conservative General Election result.

    (And of course just a month ago Mrs M got 38% to Lab’s 27%)

    For those who want to re-read his post:

    This time round, he seems to think the lead is about the mid teens:

  13. I wish they had an election night channel that just dealt with the stats.

    What would be so difficult about having a network of gossip from reliable tellers and party agents “and now we’re going over to Stamford Bridge for the latest score”.

  14. Polling Station update…SE Cornwall.

    5 voters ahead of me and two following me in at 9.30 in small village of 3000. I am 52 and was about 40 years younger than the average….?….No sign of the bloc youth vote….Sheryl Murray …strongly pro Brexit MP will walk it I suspect…..

  15. My prediction after the final poll of campaign by ipsos mori

    43% 340

    37% 232

    9% 9

    4% 0

    2% 1

    4% 49

    0.5% 1

    ?NI 18 seats

    Conservative majority of 30

  16. Just back from my 2nd visit to my local polling station, turned away on the 1st occasion due to the fact that I wasn’t allowed to take my, ( new ) bike, in. Having had 3 good bikes stolen, I now affect the appropriate paranoia when parking, and here, ( Southwark ) I’m particularly vigilant, the only young people I encountered were the usual suspects on the lookout for bikes and phones to snatch, not a bad strategy when one thinks about it, soft targets concentrating on democratic issues, easy meat.
    Anyway, good luck everyone, thanks AW for your forebearance at this testing time. ;-)

  17. Amidst all the polling excitement, yesterday we broke a record!!

    From the Beeb…

    “Renewable sources of energy have generated more electricity than coal and gas in the UK for the first time.

    National Grid reported that, on Wednesday lunchtime, power from wind, solar, hydro and wood pellet burning supplied 50.7% of UK energy.

    Add in nuclear. and by 2pm low carbon sources were producing 72.1% of electricity in the UK.

    On Tuesday, a tenth of the UK’s power was coming from offshore wind farms – a newcomer on the energy scene whose costs have plummeted far faster than expected.

    So much power was being generated by wind turbines, in fact, that prices fell to a tenth of their normal level.”

  18. Constant stream of reports coming in now of decent young people turnout. If continues to hold, polls are going to be wrong.

  19. Shevii

    A kind of political Sky Soccer Saturday for geeks…would pay good money for that !!!

  20. I’m hearing from my local polling station that it’s like an X Factor queue outside the door.


    You seem pretty ignorant of young people if I may say so. In fact the under-25’s are now the least likely demographic to drink alcohol.

  22. Ok, I’m going to post something here and I know you’re all going to shoot me down in flames but i can’t help it.

    All these differing polls and all this talk of methodologies/weightings leads me to conclude that for clarity, what would be useful is if polling companies publish two very clear sets of results

    1. The basic OPINION POLL. The percentage breaks that people tell them for each party. no messing about with likelihood to vote and other weightings. Just the raw preference data. At the vey most, they could weight the numbers according to the population levels for the age of each person contacted.

    2. An addiotional set of numbers, the ELECTION FORECAST. This will be as the name suggests, that polling companies estimate of the VI if there was a GE on the day the poll was taken.

    We could then have clarity. The first, the raw poll will be a useful guide to public opinion, especially as the campaign progresses. The true preferences of the public will be clear. People who want to weight raw polls in a rolling average in proportion to their sample sizes and date of sampling can do so.

    YES, I can see that I will get responses that this data is there if you look at the data tables for each poll – but the average Joe is not going to go trawling through data!

    It’s just that I think the two seperate approaches would give a clearer view during the campaign and would also enable people like us to see a clear distinction between a measure of peoples opinions and a measure that has been processed according to a polling companies assumptions about demographic preferences/turnouts etc.

  23. Thanks Anthony.

  24. @Ken

    You should be driving your hydrogen-powered car to the polling station!!

    Even if they nick it, they’ll find it a bit confusing when they go to fill it up at a regular station…

  25. @drmibbles,

    keep going, this is manful stuff. lol.

    It’s definitely the Knight from monty python Holy Grail analogy.

  26. TonyBTG
    I agree that publishing the raw data would be useful but it needs to include don’t know’s and will not vote numbers so we can see the trends in those as well


    “There are no American Tanks in Baghdad”…(Copyright Comical Ali)

  28. @ AW

    Thank you for your explanation above. In fact, I thought that I had observed methodology changing in the penultimate polls in previous elections and that it was a ‘standard sort’ of practice… but I didn’t understand why. Now I do and it is delightful to be given an insight into the considerations which underpin your polling decisions.

  29. @Philotes

    You mean all the vainest and most self-regarding people are turning out to vote and everyone sensible is avoiding it like the plague?

    Could have a significant effect on the polls….

  30. @CARFREW
    I’ll be driving my electric car to my local polling station. Even if they steal it they wont get far!

  31. @Duncan

    Yes absolutely agree. That way we can see how the don’t knows and “will not vote” numbers vary.

    This would give the media something else to talk about too. Imagine polticians being grilled during the campaign as to why the “will not vote” segement of the latest polls is growing.

    Why is it that this campgian is engaging less and less people?

  32. Very quiet in the gambling pits. Some volume going through in the bigger markets (most seats, majority, etc.) but not much change.

    Those putting money down aggregating out to (with odds converted to %s):

    Most seats CON 95% (LAB 5%)

    CON Maj 85%
    Hung P’mnt 12%
    (within that lots of different outcomes CON, LAB minority and LAB+SNP coalition the three most likely at about 3% chance each, the balance of 3% split between the very fragile set-ups)
    LAB Maj 3%

    For any late betters, selection of the “better ways” to express a view (IMHO):

    LAB fans. LAB minority at 32-1 would seem a more likely outcome (given the seats changes required and Corbyn’s likelihood to try and go it alone if he can) than LAB majority at nearer 40-1 (both with best odds on betfair) but it’s your money!

    Close callers: either hung parliament at equiv of 8-1 or CON 300-349 seats at equiv of 4-1

    CON fans: not a lot looks good odds TBH. At CON 365, LAB 200 seats “mid-ish” your not getting great value on any “slightly bigger majority” kind of bets so given higher than normal chance of a shock you might want to consider the long-shots. CON over 399.5 at 6-1 or for a smaller bet and much better odds try your luck picking one of the small majority size buckets (e.g. 150-175 at 25-1)

    All above odds from betfair who will almost always have the best odds.

    Individual seats markets are totally dead. Sign-up for 3x10quid free bets on Paddy Power if you haven’t already (and if you have just find more friends and/or bank cards to get more “opening promo” free bets!).
    Apart from a very few number of seats (eg Twickenham!) very little happening on betfair individual seats.

  33. @ Bardini

    Yup. Big Lab/SNP turnout. Coalition of c*cks.

  34. @TonyBTG

    “All these differing polls and all this talk of methodologies/weightings leads me to conclude that for clarity, what would be useful is if polling companies publish two very clear sets of results”


    Well even if it aids clarity, it might not aid their influence. The press often commission polls, and they are not noted for a desire to avoid influence. If you want polls to influence, you would do them the way they are done. Including publishing them immediately. (Unless you don’t like the result in which case you hold them back…)

  35. Followed the discussion on here with interest for 2 elections now. First post an anecdote and a prediction. Here in Wansbeck, I was the “youth” vote at 0930 this morning and I’m 53!!
    Prediction from me is a comfortable but not huge Con majority of around 50-60. Its so un-trendy to be Con at the moment that the “shy” effect will be larger than before


    Give it a rest. It’s pathetic for an older person to take delight in cheap shots against the younger generation. They are our country’s future, you should have some pride in them and try to see the best in them.

    Plus as Dr M. said, kids these days don’t drink that much. Perhaps you should get out more and talk to some of them.

  37. It’s all good and well saying “loads of the youth vote here” but that’s all a bit meaningless unless we have last year’s stats to compare it with..

    Great, there might be 10,000 people in an X-Factor style queue… but if last year there were 20,000 in that queue, then no.. that’s not the youth turnout coming in force..

    Then of course, there’s the element of who was in the seat before.. if it was Labour previously and you’re assuming these hoards of youngsters are also voting Labour then great, the majority in that seat just rose..

    So, take eye witness reports with a pinch of salt

  38. DrMibbles – Care to share where these “anecdotes” are coming from? Also which seats are they in?

  39. It sounds like the youth turn out is strong but patchy. Some areas swamped others nothing. Will be interesting to see how that effects the result

  40. MIKE88

    I’ve seen reports on twitter from Bedford, Ipswich and Birmingham all saying the same thing – higher than normal turnout of young voters for this time of day

    It may not hold. It could be nothing. But all the data said it was going to happen, and I will be quite surprised if it doesn’t.

  41. “Plus as Dr M. said, kids these days don’t drink that much. Perhaps you should get out more and talk to some of them.”


    Yep, more and more becoming teetotal according to surveys. It’s the drunk, rude boomers barging into me at gigs. The youngsters were much nicer…

  42. @sceptic,

    I suspect you are right. Just a gut feel, but expect massive turnout in inner cities, but same old everywhere else.

  43. Ipsos Mori Poll:

    CON: 44% (-1)
    LAB: 36% (-4)
    LDEM: 7% (-)
    UKIP: 4% (+2)
    GRN: 2% (-)

    (via @IpsosMORI) Chgs. with 01 Jun


    Trouble is all those were probably voting tory.

  45. I’ve received loads of facebook updates today imploring us all to vote Labour but not a single Tory? hmmmmm…. As there does not appear to be any Tories left does this mean that there will be a Labour landslide?

  46. Is it still possible to bet on a hung parliament? I can’t seem to find it anywhere. It’s been removed from OddsChecker.


    Bedford and Ipswich are both ‘leaning Labour’ on the yougov modelling


    My idea is designed to specifically remove the opportunity for polls to influence.

    A POLL should be a measure – a snapshot – of public opinion taken at a specific time.

    Becuase polling companies produce polls for customers, who can be sure that their decisions on methodology are not skewed by the political bias of the customer.

    EG. a poll for a right wing paper, might want to show a poll with a narrower Con lead to make sure that con suipports are not complacent. Or, if preferred, show a larger Con lead, to reassure Con supporters that their party is on the right track.

    I’m not saying this happens. All I’m saying is that if we allow polling companies to only publish the processed data, how can the public be sure that they are trustworthy.

    Better to be up front about it it.

    First Poll : This is what people’s current preferences are.

    Second Poll: This is what our chosen methodoly is predicting.

  49. RE: Young voters and turn out.
    There are two competing narratives. That young people are coming out in droves. And that they’re not (or I’d they are, it must all be in safe Labour seats).
    These are exactly the same narratives as the *predictions* of turn out.
    Until tonight nobody really knows. Anything else can be prefaced with ” I hope that…”

  50. Tonybtg – I think every company does (or at least, every company publishes tables were you can see what the voting intention is with only demographic weights, before any assumptions are made about turnout, reallocations of don’t knows and so on.

    People don’t pay much attention to it unless they are hardcore geeks (and they are quite correct not to, pollsters make these adjustments for a reason and people will end up getting mislead by looking at figures that don’t have reasonable adjustments made), but the data is there.

1 13 14 15 16 17 19