Skip to main content
Coronavirus (COVID-19): Latest updates.


Final Call: the last Welsh Poll of the campaign

7 June 2017

The final Welsh opinion poll of the 2017 general election shows Labour on course to maintain their long-standing dominance of electoral politics in Wales. It also shows the Conservatives on the brink of their highest vote share here for more than a century. Both Plaid Cymru and the Liberal Democrats, however, appear to be facing their worst general election result for many years.

The final Welsh Political Barometer poll of the general election campaign was conducted between Monday and today. Our pollsters, YouGov, have sought to provide our final measure of voting intentions as close as possible to election day itself.

After a campaign that has seen wild fluctuations in the opinion polls, and particularly so here in Wales, this is what our final poll has found in terms of voting intentions for the general election (with changes on the last Barometer poll, conducted just over a week ago, in brackets):

Labour: 46% (no change)
Conservatives: 34% (-1)
Plaid Cymru: 9% (+1)
Liberal Democrats: 5% (no change)
UKIP: 5% (no change)
Others: 1% (+1)

There are no big changes since our previous poll; all movements are well within the ‘margin of error’.

Using the normal method of projecting these polling numbers – computing uniform national swings since the 2015 general election for all forty constituencies in Wales – then our latest poll implies the following overall result (with projected seat changes from the 2015 result shown in brackets):

Labour: 27 seats (+2)
Conservatives: 9 seats (-2)
Plaid Cymru: 3 seats (no change)
Liberal Democrats: 1 seat (no change)

After the first two polls of the campaign had suggested the Conservatives were on course to make substantial seat gains in Wales from Labour, and possibly even come first in both seats and votes, things have certainly changed around dramatically in recent weeks. Labour have fought back strongly. If these figures are born out in the election tomorrow, the Conservatives would gain their largest share of the vote in Wales at a general election since before World War I. Yet they may not gain any seats for this, because Labour – this poll indicates – may themselves be about to secure their highest vote share in Wales since the second Tony Blair landslide victory of 2001. On uniform swings, only two seats are projected to change hands in Wales from 2015: Labour would regain both seats they lost to the Conservatives two years ago, Gower and the Vale of Clwyd.

As has been the case in all previous polls during this campaign, uniform swing suggests that Plaid Cymru and the Liberal Democrats remain on course to hold the few seats they currently have, but will not make any gains. But both parties may also be facing their vote share decline from that in 2015: in Plaid’s case to the lowest level seen since the general election of 1992, and for the Lib-Dems the worst general election they or their predecessor parties have ever experienced in Wales.

We must always show some caution about both the polling figures and about how these figures are translated into seats. The general election just two years ago demonstrated that the polls are not flawless; this one will be the first major test of some of the changes that the various polling companies introduced in the wake of the 2015 election. Recent weeks have seen some of the major polling companies produce quite different findings in Britain-wide polls: all of them have shown the Conservative lead to have fallen substantially, but they disagree about the extent of that fall and the size of the continuing Tory lead.

We must also remember that uniform national swing provides simply a baseline method for projecting polling results onto the electoral map. Even if this final Welsh poll turns out to be perfect in terms of vote shares (and I suspect everyone, including YouGov, would be pleasantly surprised if the poll truly is perfect), then the results in terms of seats may well not follow those indicated by uniform swing. Particularly effective candidates and campaigns, or indeed strikingly ineffective ones, can make a significant difference. In 2015, a uniform projection of the actual swings that occurred across Wales would not have indicated that the Conservatives would gain either Gower or the Vale of Clwyd from Labour: the Tories needed above average swings to capture both seats. The one thing we can say for certain about changes in Wales’ forty constituencies since the 2015 election is that they will not be absolutely uniform for all parties in all places.

A final note of necessary caution concerns from where the main parties are getting their support. There has long been a pattern of Labour tending to do relatively better among younger voters and the Conservatives gathering more support from older ones. But the extent of that disparity in recent polls, include this final Welsh poll, is truly extraordinary. The sub-sample of voters aged 18-24 in our new poll shows fully 73 percent intending to vote Labour, and only 14 percent supporting the Conservatives. Among voters aged 65 and above, some 56 percent intend to vote Conservative and only 27 percent Labour. Older voters have long also tended to be more reliable in actually coming out to vote on the day. Having enthused many younger voters, the key task for Labour must now be to ensure that they actually turn out; if they do not, Labour could still face serious difficulties in this election.

This has been an extraordinary campaign, for an unanticipated election. Not least of the extraordinary features has been the substantial turbulence in the opinion polls. Our final Welsh poll, however, suggests that we may be on course for ‘business as usual’, at least in terms of who wins the seats. Labour have come first in both votes and seats at every general election in Wales from 1922 onwards. The last person to defeat Labour in a general election here was Lloyd George – and he had just won a world war! If our final poll of the campaign is broadly correct, tomorrow the Welsh Labour party will score its twenty-sixth general election victory in a row. That may not be enough to put Jeremy Corbyn into 10 Downing Street, but after a campaign that started with Labour looking under greater electoral pressure than for a century in Wales, it would still be a remarkable achievement.

The poll, for ITV-Cymru Wales and Cardiff University’s Wales Governance Centre, had a sample of 1074 Welsh adults and was carried out by YouGov from 5-7 June 2017.


  1. Richard

    Some pollsters will have egg on their faces Friday I guess ?

  2. K. Vivian

    For nearly a hundred years the Welsh have consistently been conditioned to vote Labour. For the same period of time Wales has been one of the most deprived regions in western Europe, regarded as being sinisterly foreign even at Westminster (“f….ing Welsh,” according to Tony Blair) yet it will be similarly acquiescent tomorrow, You can’t teach a daft dog new tricks.

  3. GDR

    I think tomorrow the result will be a lot closer but probably a narrow lead for Welsh Labour.

  4. Rosemary

    ‘Conditioned to vote Labour’? That always make me laugh. Trouble with Plaid Cymru is that they always talk Wales down and never actually feel that voters have minds of their own and can make a reasoned choice. Wales and it’s people, culture and heritage should be celebrated. Labour has been strong for 100 years? Obviously there is no appetite for the alernative so some may want to go back to the drawing board and start actually listening to the electorate rather than thinking they know what is best for Wales and sticking to their rigid ideaology come what may.
    I do believe Labour will do well here. I do not believe Labour will do so well in England and Scotland outside of the major metropolitan areas and will lose seats in the heavily pro brexit areas of the North and Midlands. Unfortunately it looks like we will again have a Tory government so the other parties must steer away from tribal politics and work together for the sake of the UK. Turbulent times, indeed.

  5. JohnT

    Well the scientific predictions are done and dusted so bearing in mind some comments above perhaps I may be allowed a final stab in the dark from a gut reaction point of view. I feel that Labour will regain one of the “surprise” seats lost to the Conservatives last time but probably not both as voting intention focuses down to those very few seats that could possibly change hands. For the same reason i.e. a concentration of electoral strength in key areas I still expect Plaid to gain one seat and for much the same reason for the Lib Dems to keep what they have. In short raw data does not account for locale, for candidate and campaign quality locally nor to reaction against assumptions by a small but possibly conclusive percentage of electors in tight seats. I think Rosemary is being a little partisan in her comments regarding Plaid because in truth I’ve only ever heard their people talk Wales up….not down.

  6. Phil Mcg

    Rosemary – ask any labour politician a question on independence and see what talking Wales down sounds like.

  7. Richard

    Well as I said – it’s seats that count ! Egg on a few faces and three party politics in Wales now – no longer 5

    Cyd weithio ymlaen ! I hope so

  8. J.Jones

    Good poll. Actually as close to perfect as you can get although obviously individual seats moved differently as they always will.

  9. J.Jones

    Survation seems to have it spot on for the UK. Everyone now remarking that the young vote, probably galvanised by the EU referendum loss when they didn’t vote, have decided to take a grip on their future.

    A good day for democracy although probably the LibDem loss in Ceredigion is down to Plaid Aber students.

  10. K. Vivian

    Much to expectation and delight of the over fed elite of Westminster the turkeys in Wales continue to vote for Xmas

Leave a reply

Your email address will not be published. Required fields are marked *