To start with, here’s an update of all the pre-election polls (Ipsos MORI, Survation and NCPolitics all published theirs on the morning of election day, after my last post).

Note that ComRes and Hanbury also produced polls during the campaign, but not with fieldwork conducted on or after the final weekend of the campaign. For what it’s worth, they tended to show high Labour support, though we’ll never know what their polls would have shown in the final week.

Needless to say, the pre-election polls varied wildly from one another for all the main parties. Labour had a twelve point spread (13% to 25%), the Conservatives eight points (7% to 15%), the Liberal Democrats (12% to 20%), the Brexit party eleven points (27% to 38%). In the event, the polls that had low Labour scores and high Liberal Democrat scores were closest to reality. Compared to the final results, Ipsos MORI took the laurels, getting close to the correct result for all parties. YouGov were next, getting within a point or two of most parties but overstating the Brextit party. Other companies recorded significant errors, with a few double-digit overstatements of Labour support.

It is difficult to point at a single obvious cause for the wide variation. When there were huge differences between polls at the 2017 election the reasons the were clear: pollsters had adopted demographic turnout models and other post-fieldwork adjustments which backfired and overstated Tory support. There is no such easy explanation for the 2019 polls – pollsters have mainly reversed the missteps of 2017 and, while there are some variations in approaches to turnout, the elaborate turnout models that made such a difference in 2017 have disappeared. Different approaches to turnout perhaps explain differences of a point or two, they don’t explain differences of 10 points. The differences here look as if they are more likely to be down to pollsters’ different approaches to sampling or weighting, and the representativeness of their samples.

From the beginning these European elections were going to be a challenge. They are a low turnout election, when at recent elections polls have struggled to correctly reflect the demographic pattern of turnout. In recent decades most British pollsters have also relied upon past-vote weighting to ensure their polls are politically representative, and this was an election when past vote was a particularly poor predictor of current voting intention.

In terms of what this means for wider polling, errors here don’t necessarily transfer directly across to Westminster polls. The challenges posed by high-turnout elections can be very different to those posed by low-turnout elections and just because some polls overstated Labour support in the European elections does not necessarily mean they are overstating Labour support for general elections. On the other hand, given the recent history of errors, it probably isn’t something we in the polling industry should be complacent about.

15 Responses to “European election polling post-mortem”

  1. Westminster voting intention:

    LDem: 24% (+6)
    Brex: 22% (+4)
    Con: 19% (-5)
    Lab: 19% (-5)
    Grn: 8% (+2)

    via @YouGov
    Chgs. w/ 17 May

  2. I doubt that Electoral Calculus was designed to cope with this kind of VI, but it suggests GB seats as –

    Lab 202 : BxP 141 : LD 119 : Con 110 : SNP 55 : Plaid 4 : Green 1

  3. Oldnat

    Lab 202 : BxP 141 : LD 119 : Con 110 : SNP 55 : Plaid 4 : Green

    That suggests that as you say Jo Swinson would lose her seat, given that there are 56 seats in Scotland and the one non SNP seat is definitely the Orkneys and Shetlands.

  4. Prof Howard

    59 seats in Scotland – presumably 55 SNP 4 LDs.

  5. Oldnat

    Thanks. Then there is a chance Jo Swinson would be one of the 4 Lib Dems to survive.

    Interesting to see Lib Dems on most votes but not most seats.

  6. Interesting that those polls, versus actual results, show that Con and Lab hit their lowest score, LD and Greens their highest and TBP somewhere in the middle.

    If the latest polls are on the same track then the two “major” parties will need to be renamed.

  7. New thread.

  8. I see the LD leadership rules mean they have to be supported by 10% of MPs.

    Given that they have eleven then, so long as they nominate themselves, they’re almost there.

  9. @CIM (FPT)

    “How – practically – do those get solved so that we can have a second referendum, without a preceding GE, in a timescale that the EU would be willing to grant a further extension for. What configuration of referendum question and details cuts through that?”

    We can’t. That’s why I go for revoke and reset. We need to remove the ticking clock, which both prevents patient discussion and also creates a threat weapon used by both sides.

    Instead, we need to make space for the crafting of an acceptable leave option which would then be ratified or rejected in a second vote – which would be on this deal or remain.

    It’s not perfect by any means, but we need to find a way out of the dogs dinner that the Conservatives have managed to cook up.

    It’s also worth pointing out that if we want to leave, we’re going to have to agree the WA. Brexiters need to suck up this simply fact, or prepare to remain. It’s also worth noting that we are all stuck on ‘the Brexit deal’, whereas the conversation we need to have has to start in a very different place if we are going to be able to sort out Brexit.

    We need to work out viable solutions for the relationships within and between the four nations, sort out mechanisms for managing the tensions between the English regions, and deal with an electoral and parliamentary system that is no longer fit for purpose and that actively mitigates against conciliation and finding solutions. This will almost certainly involve some serious implications for Labour and Conservatives as the two main power blocks.

    Once we’ve done that, then we’ll find Brexit can be sorted, hence my plea for a delay – possibly a long one – while we reinvent Britain.

    It’ll never happen.

  10. Profhoward,
    I think if the Lib Dems are on 24% in Britain i doubt if Jo Swinson would be in trouble. The Scottish crossbreak in the yougov poll has the SNP up 7% and the Lib Dems up 4% since 2017 which would not overturn a 10% majority. We need a proper Scottish poll though.
    She did not lose by much in the SNP tsunami of 2015.

  11. @BRXT

    The WA has not been scrapped. It is still on the table and the EU say they will not change its basic terms. The question is whether the EU and UK can agree a free trade deal quite quickly, so the Irish Backstop is not an issue. This process of negotiating a trade deal would take time and therefore the WA would not be implemented until there was certainty about when the revised trading relationship would apply from.

    It is not impossible that a new Tory PM secures a free trade deal and the WA agreement is implemented. However. It Is unlikely this will happen before end of October 2019. My guess is that negotiations with the EU could continue into 2020.

    The UK will not leave the EU without a deal. There are too many Tory’s against this and therefore parliament will block by amending legislation.

    If a new Tory PM is unable to agree an EU trade deal that satisfies enough MP’s and they are not happy to implement the WA, then you are looking at General Election first and then a referendum.

  12. Yougov data at

    Although at two days old, it might already be out of date.

    I’d say what I usually say. The libs are leading on the amazing support level of 16% of voters.

    I mean…the most popular political party in the Uk can only engage 16% of registered voters!!!!!

    It is more of a win be default. Dont know continues to be higher, on 18%. I recall commenting how DK on half or 2/3opf whatever the leading party got was enough to reverse the result, and now it is the plain leader. The only popular party in the Uk is the SNP. (apologies if i slight other regional parties for which i dont have the numbers, but it is also true that with overwhelming support in Scotland, SNP could become important in national decisions)

    In the issues section, Brexit tops the poll as we might expect, but it is interesting the environment came in 3rd after health 2nd. The youth/remain vote was the most interested in the environment.

    wrong to leave 46%. Right to leave 41%. Consistent with a larger vote share for remain parties.

  13. On the current numbers both lab and con would be mad to go for an election. Both because many of them would be facing retirement, and because there is absolutely no clear likely winner and therefore policy outcome from a new parliament.

    This constant chasing deadlines is insane, especially when after getting an extension all parliament does is waste the time until lthe next one. if it cannot make a decision, then it has to remove the timebomb and revoke. It is the only sensible thing to do. We know the EUs negotiating position now. If the nation wants to argue and decide what to do, it can do so on its own timetable. Better for parliament to bite that bullet, revoke, and spend the time it has left before the next mandatory election tryng to sort out its position.

    Presumably then we could have a brexit election.

  14. > the Brextit party.

    Oh, aye?

  15. @DANNY
    Lots of talk about removing the ticking time bomb:

    The problem has always been that nearly half the electorate do not see it as a time bomb but the count down to something different. Indeed I suspect at this point they don’t care if it is actually a proverbial grenade superglued to their hand.

    The reason the MP keep p155ing the time away is they do not want to be responsible for the decision. The ideal scenario is that the UK leaves because the EU just kicks us out. It would mean that everyone can point that it is the EU’s fault. It s why I suspect the EU will allow the UK an extension. It loses nothing at this point. How does it help the next PM. it does not?

    This is about shifting the blame. Not about whether we remain or not to the EU. I agree with you in one sense that most of the players for PM understand that they have very little space to have a coherent plan. but the Tories know the name of the game is that they have to show they are going for a no deal

    The House of common arithmetic at the moment says nothing gets through and will continue to do so until something breaks.

    Lastly the UK has a position it is just achievable. This is why no deal is seductive it is the illusion of control, which is after all why people voted to leave in the first place.

    We will have to leave with no deal and the real problem is for the Tories is how they sell it as it was not their fault.


    You raised the correct question yesterday in that why should Labour try and win both Bristol North West and Walsall North. The simple answer is that in the end both want the same thing using the same method Brexit is getting in the way and putting a wedge where there was not one

    Most everyone outside the right and indeed including the right no believes that austerity has gone too far. The point is that the current orthodoxy is really broken and the problem is what replaces it. At the moment Brexit is the simplest answer: It is full of once we leave we can……place your answer here.

    The point is how to move beyond Brexit. As an example do you think that HUGO would want Dominic Raab et al Britain Unchained as a philosophy for post brexit?

    Interestingly one of the many myths about Trump was that he won over the poor I find this astonishing in the face of the facts and polling Clinton won the poor by 11% ($50k or less, she lost the wealthy by 1%) she basically lost the election by a grand total of 200K votes across three states it is turnout of the poor that cost her.
    What was interesting was that when Democrats went on Fox (just think a Gay and a socialist got standing ovations from what is supposedly a trump audience) Clinton was a polarising figure. What I think is interesting about US was that 2 year ago GA FL were seen as not winnable and to get within 15% in TX would have been earth shattering all three thing are happening. Do they win Ol’ Miss? no but I think the world is changing on that front.

    My view is actually another 4 year of Trump seals the deal. Since the GOP will have a base so out there that it makes it difficult to row back

    The UK is in a different state to US in that sense we have to go through the Trumpian phase, My view is that much of the problem in the Tory party hierarchy is that they believe rightly or wrongly that no deal is bad and having to row back is going to be tough once they have left