Realistically there are four main battlefields in the general election. First is that between Conservative and Labour, which is the battle that will really determine how large the Conservative majority is (and how badly Labour are damaged by an election fought when they are at a historic low). Next there are the Lib Dem battles, against Labour and against the Conservatives – polls and by-elections suggest the Lib Dems are staging a recovery, but what is the potential to convert that into seats? Finally there is the position in Scotland, which these days is a wholly distinct battle from the rest of the UK: will the SNP repeat their almost clean sweep of Scottish seats?

Note that EU referendum results weren’t actually counted by constituency – the figures here are all the estimates produced by Chris Hanretty of UEA.

Conservative vs Labour battleground

This is the largest, and the real show in this election. How many seats will the Tories take off Labour, how deep into usual Labour territory will they stretch?

There are forty-two Con-Lab seats that would fall on a uniform swing of five points (including Copeland which the Tories have gained in a by-election already), eighty-two that would fall on a swing of ten points. To explain that to those who I know find talk of swings at elections baffling – a swing of 5 points is the equivalent of one party going down five points and another going up, so if a party has a majority of ten percentage points, their opponent would need a swing of five points to defeat them. This means a five point swing is the equivalent of a seventeen point Tory lead in the GB polls (ten points up from 2015), a ten point swing would be a towering twenty-seven point lead in a GB poll. A victory by twenty-seven points is, of course, a fairly outlandish prospect, but in reality the swing will not be uniform. Labour will hold some seats against the tide, and the Conservatives will take some seats that needed swings beyond the national average.

Looking down the list there are a couple of inner London seats that are heavy with young professionals and voted strongly Remain… but these largely have very small majorities, so might fall to the Tories despite that. There are also those seats that Labour took off the Tories in 2015, largely ripe for being retaken on modest swings – places like Brentford, Chester, Dewsbury, Enfield North, Wirral West and Wolverhampton SW. Birmingham Edgbaston, a perennial Tory target that people assume has remained Labour due to the personal popularity of Gisela Stuart is there (Stuart is not, she is stepping down)

The small number of Labour holdouts in the South outside London are almost all on the list – Hove has a majority of only 2%, though was heavily Remain. Labour’s last seat in the South Hampshire conurbation – Southampton Test – falls on a four and half percent swing, as does Bristol East. Exeter would need a swing of six and half percent to fall. Luton South and Bristol South fall on seven percent swings. That leaves only Slough (a 15% Labour majority and a high BME vote that the Tories would struggle with, though Labour lose any incumbency vote from Fiona Mactaggart’s retirement) and the probably impregnable 22% majority in Luton North.

Most of the list is, however, made up of seats in the suburbs and provincial towns and cities of the Midlands and North. The outskirts of the West Midlands conurbation are well represented, with seats in Dudley, Coventry, Northfield, Walsall and Wolverhampton on the theoretical target list, as is Greater Manchester, with plausible targets in Bury, Bolton and Worsley. There’s another group of marginals in North-East Wales and the Wirral – Wirral South and West, Ellesmere Port, Wrexham, Delyn and Alyn & Deeside. The great Northern cities themselves aren’t there – the Conservatives are not viable in Liverpool, Manchester or Sheffield; Tynemouth is the only seat on Tyneside.

Looking down the list of Labour MPs at risk there are pro and anti-Corbyn MPs in the firing line: John Woodcock who is standing but refusing to endorse Corbyn is in the 8th most vulnerable seat, Corbyn ally Cat Smith is in the 14th. Mary Creagh and Vernon Coaker have only 6% majorities to defend, Lindsay Hoyle – a favourite to be the next Speaker if he survives – has a 9% majority in Chorley.

Conservative vs Lib Dem battleground

The expectation is that the Conservatives will win seats from Labour, but that this will be blunted to some degree by losses to the Liberal Democrats. The Lib Dems have staged a modest recovery in the national polls and a strong recovery in local by-election contests. On a straight national swing we shouldn’t necessarily expect them to win anything from the Tories – the Tory vote has increased as much as the Lib Dem vote. In practice, however, I would expect the Lib Dem recovery to be concentrated in places with a history of recent Lib Dem support and places that voted against Brexit.

That said, the Liberal Democrats will have pull out some impressive swings to get more than a modest number of gains. There are only ten Conservative seats with the Lib Dems in second that have majorities under 10%, another fifteen with majorities under 20%. Few of these seats were strongly pro-Remain: Kingston & Surbiton and Twickenham voted heavily against Brexit and are high on the target list, but the Lib Dem targets in Cornwall and the South West mostly voted to Leave. All of these seats are places that had Lib Dem MPs recently, and I would expect many of those former MPs to seek a rematch – the most prominent, Vince Cable, has already confirmed he is to stand again in Twickenham.

Labour vs Lib Dem battleground

Given the catastrophic performance of the Lib Dems in Labour areas in 2015 there are actually very few Lab-Lib Dem marginals on paper. In many seats that we are used to thinking of as Labour held Lib Dem targets (Sheffield Central, say) the Lib Dems collapsed to such an extent they are no longer in second place. There are plausible places where they could come from third place to win if they do particularly well, such as Bristol West and Norwich South. While there are Leave seats on this list, the most plausible Lib Dem pick ups are the mix of inner-city and university seats that voted overwhelmingly for Remain – the sort of young, well-educated areas where the Lib Dems have traditionally excelled like Cambridge, Bermondsey and Manchester. Again, these are all seats that the Lib Dems previously held and in some cases former Lib Dem MPs will be standing again – Julian Huppert and Simon Hughes are confirmed as the candidates in their former seats.

The SNP Defence

In 2015 Scotland was a crushing victory for the SNP, sweeping almost all before them. It was an effective lesson in what happens under the First Past the Post system when a new political cleavage becomes dominant, people on one side have a clear main party to vote for and the other side is split between three different parties: the side with a united vote utterly smashes the other side.

In terms of support the SNP are in the same sort of dominant position they were in 2015 and I think we can be relatively safe in predicting another easy SNP victory. The interesting thing in terms of seats will be the behaviour of the other parties. The defence list of SNP seats is below – but don’t just look at the majorities, look at the shares too. For example, in both Dumfries & Galloway and Paisley & Renfrewshire North the SNP has a majority of 12%. However, in Paisley that represented an overall majority of the vote, in Dumfries the SNP vote was ten points lower, but the Unionist vote was split.

The SNP will almost certainly win the vast majority of seats, but whether they manage another almost clean sweep depends on if the Unionist vote remains split, or whether Unionist voters vote tactically for the party best placed to beat the SNP. Since 2015 the Scottish Parliament elections and the subsequent polls suggest that the Scottish Tories have sneaked past Labour to become the most popular Unionist party in Scotland… whether Scottish Labour voters are willing to vote tactically for a Tory is an interesting question.

Other interesting seats

Besides those main battlegrounds there are, as always, various other seats that are interesting in their own unique ways and worth keeping an eye on:

  • Thurrock was an extremely tight three way marginal between UKIP, Conservative and Labour in 2015. If the UKIP vote collapses towards the Tories it should be safe for them.
  • Richmond Park was a Conservative seat in 2015 but has already been won by the Lib Dems on a huge swing after Zac Goldsmith’s resignation. Can they retain it without the focus of a by-election and a proper Conservative candidate against them?
  • Brighton Pavilion is currently the Green party’s sole seat – with Labour in retreat they should hold it.
  • Bristol West and Sheffield Central both have the Green party in second place. In Bristol the Greens are very clearly the main challenger, 5000 ahead of the Lib Dems in third place. Sheffield Central meanwhile is a very pro-Remain university seat where the Lib Dems have traditionally done very well.
  • Manchester Gorton was due to have had a by-election, cancelled because it was overtaken by the general election. The Lib Dems were reporting a strong performance there and they will, of course, already have done a lot of leafletting and campaigning there.
  • Ynys Mon and Edinburgh South are the two Labour seats at most risk to Nationalist candidates – Plaid in Ynys Mon and the SNP in Edinburgh South, their sole remaining Scottish seat
  • In Southport the incumbent Liberal Democrat MP John Pugh is standing down. Given their reliance upon the personal vote of their Members of Parliament the Lib Dems have sometimes struggled to pass on seats when an MP retires (though they have done so in Southport before, retaining the seat when Ronnie Fearn stepped down in 2001)
  • Orkney and Shetland is the Lib Dems sole seat in Scotland, a Liberal seat since 1950. Alistair Carmichael survived a failed legal challenge to his election in 2015, concluding that he had not committed any illegal practice, but that he had lied. Whether that saga has any impact on Carmichael’s support remains to be seen.
  • Finally there is Clacton, UKIP’s sole constituency at the 2015 general election. It was held by Douglas Carswell at a by-election when he defected from the Conservatives and again at the general election. Carswell himself has since left UKIP and endorsed the Conservatives at the next general election, though will not stand himself. That may mean it is an easy Conservative gain, though Arron Banks is still to confirm whether or not he will go through with his intention to stand there now Carswell has stepped down.

32 Responses to “The election battlegrounds”

  1. Thanks Anthony. A useful rundown (with so many variables in place, forecasting is a mug’s game).

    Things may become clearer after a few week’s campaigning – and once local election results are known.

  2. I suspect the Scottish local elections are going to tell us less about Unionist voters willingness to vote tactically, night be less than one might assume.

    1. EU citizens and 16-17 year olds can vote in May but not in June and, while there is some evidence that more EU citizens will vote this time round, it’ll be long after the UK GE before the analysis has been done.

    2. Turnout will (presumably) be considerably lower in May than June. and many of those who will vote only in the UK GE are less interested in politics, and less likely to be affected by ideas like tactical voting.

    However, in most wards, there will be only 1 SCon, 1 SLab and (perhaps) 1 Lib-Dem. Theoretically, that allows us to look at the 2nd preferences of Unionists and perhaps indicating a willingness to vote tactically.

    But there are lots of Independent candidates, and there is a strong possibility that Unionist party supporters will prefer to give their 2nd pref to an Independent.

    Full Scottish polls (especially those with geographic crossbreaks – for all their deficiencies) might tell us more about the likely outcome in June.

  3. Anthony.
    Thank you for the best complete and definitive seat list

  4. Anthony,
    In Norwich South The LibDems fell to fourth place in 2015 – not third! In reality the Tories are the only serious challenger to Clive Lewis , and they last won the seat by 1700 in the 1983 landslide.

  5. @Anthony

    Do you think you could have a chat with Lord Ashcroft. He’s just posted this:

    “Opinium National Poll in Scotland SNP 43% CON 32% LAB 14% LDEM 8%”.

    National Poll? It’s a subsample.

  6. @Graham

    If Clive Lewis were to lose it would be a devastating blow to Labour going forward. Lewis is a future leader.

  7. I’ve said it elsewhere but I think a lot of the lib dem losses in 2015 were put down down to tactical switches to the Conservatives to fend off the spectre of a Labour/SNP coalition. Now that spectre has faded and many of the libs will have been remainers I can forsee big swings back to them. Presumably this is what Crosby’s alleged polls were seeing.

  8. RAF

    Number Cruncher has been making the same point on Twitter.

    He has also heard rumours that a proper Scottish poll is due out tomorrow. If not, I don’t imagine it’ll be too long before one appears.

  9. @ RAF

    Just out of interest I ran the numbers on the subsample and it would give the Tories a theoretical gain of 11 seats and (hallelujah) a LD gain of 2 seats.

  10. Ynys Mon is one to watch. On paper it should fall from Labour to Plaid Cymru but it voted Leave

    GE 2015
    Lab 10871
    PC 10642
    Con 7393
    UKIP 5121
    LD 751

    This seat, given it,s a Leave constituency and given the size of the Con/UKIP vote, could be a shock result and go blue.

  11. Britain Elects? @britainelects

    Westminster voting intention:

    CON: 50% (+4)
    LAB: 25% (-)
    LDEM: 11% (-)
    UKIP: 7% (-2)

    (via ComRes)

  12. @Raf
    Norwich South is number 69 on the Tory v Labour battleground. I know of Green voters who will be tactically supporting Lewis to fend off the Tories – they have no expectation now of taking the seat themselves.
    As for his leadership prospects, I suspect his chances would have been better without a snap election and a leadership election taking place next year.Many are likely to have backed him to avoid a poor result in 2020. Now that we have a very different scenario, I suspect that if Corbyn leads Labour to a very heavy 1983 or 1997 style defeat other options are likely to be appealing. In those circumstances the membership could probably be persuaded to go for a much clearer break with Corbyn – Keir Starmer and Yvete Cooper come to mind.Effectively May could well have ended his leadership hopes.

  13. via Britain Elects

    Upcoming polls (tonight and tomorrow):

    Panelbase – Scotland
    YouGov – GB-wide
    Survation – GB-wide
    Survation – Scotland

    Plus, of course, the Welsh poll on Monday.

  14. Looking at the Comres tables I think Labour voters just staying home could be a big problem for them.

    Unweighted by Turnout

    Con 41%
    Lab 24%

    Certain to Vote

    81% of Cons
    72% of Lab
    67% of Lib
    66% of UKIP

  15. Sheffield Central and Manchester Gorton have a great deal in common and those things make them rather less safe for Labour than it seems from the simple size of majority last time out.

    The key factor in both is the large proportion of the electorate who are students. Their voting behaviour is far more volatile than other groups. Paul Blomfield’s 17000+ majority in 2015 was almost entirely the result of the collapse in the student vote for LibDems (in 2010 he got elected by 500). He is at risk from three factors: general disenchantment with Corbyn’s leadership (even though he’s part of the ‘soft’ anti-Corbyn group); a LibDem campaign that concentrates on Remain grievances and (rather more simply) a potential collapse in student participation because of the timing of the election, which will occur after many students have left Sheffield and will be voting in their parental home constituencies instead. If I was the Labour agent in Sheff Cen I’d be putting a lot of work right now in getting postal votes for students. At least he’s been spared the boundary changes which would have made things even more difficult.

  16. PS I’m completely baffled as to why people keep talking about Clive Lewis as a potential leader of the Labour Party. For God’s sake why? He’s like Gertrude Stein’s description of Oakland: when you get there there’s no there there. Why should you elect someone who got lost coming back from Glastonbury?

  17. Britain Elects?

    Westminster voting intention:

    CON: 48% (-)
    LAB: 25% (+1)
    LDEM: 12% (-)
    UKIP: 5% (-2)

    (via YouGov)

  18. AW,

    Thanks for this.

    However, shouldn’t Batley and Spen be based on the 2015 Jo Cox figures? (That’s a 6% swing needed to take it).

    Tracy Brabin’s majority is artificially high as basically most non-Labour voters didn’t even vote last October.

  19. Canmanjeff – that is the majority from 2015 (the 12% there is the majority, it would need a 6% swing to overturn it)

  20. AW,

    Sorry, too tired to read correctly.

    Thank you.

  21. The analysis I did gave Conservatives a 4% lead (about a 4,000 majority) in Batley and Spen

  22. Thanks for such a helpful post Anthony. Are you able to make the spreadsheet used to compile these tables publicly available?

  23. @Graham – “In Norwich South The LibDems fell to fourth place in 2015 – not third! In reality the Tories are the only serious challenger to Clive Lewis, and they last won the seat by 1700 in the 1983 landslide.”

    They should distribute bar charts!

  24. @AW – Excellent post – thanks

  25. I’m surprised more wasn’t made of Gordon in Scotland.

    It’s a prize the Lib Dems would love to win. Alex Salmond is the MP and in swing terms alone it’s not necessarily out of range for them.

    ‘Nationalist’ Support is 48%. That however doesn’t necessarily mean that all the obvious Unionist Vote has to come together to defeat him.

    It’s quite possible that a number of people who voted SNP there in 2015 are not pro Independence.

    In fact only 39% voted YES in 2014 and even that figure included 16-18 year olds. In 2015 the voters there might have only voted for him tactically to defeat the Lib Dems.

    Salmond however is a major Bete Noire for Unionists and if they are going come together anywhere, and vote tactically to defeat the SNP it will be here.

    And it’s a lot easier for Tory and Labour voters to stomach voting tactically for the the Lib Dems than it is to vote for each other.

    There’s also the advantage for Unionists, that Salmond is a real living proxy for voting against Sturgeon.

    On balance, I don’t expect him to lose, but he could nevertheless, be in for a shock. If the Lib Dems can defeat him with quiet Tory, Labour, and the few people who voted UKIP in 2015’s acquiescence, it will be major blow to Sturgeon.

  26. @Andy; I’m not sure Ynys Mon voting Leave is a big factor (it was practically 50:50) but more a question of where the UKIP and Labour votes go as they fall (and how far Labour fall); I certainly agree the Tories coming through the middle is possible but much will depend on local candidates as it always does on the island – and Plaid candidates can vary from centre right to far left which means they appeal to different groups

    Full Welsh poll out tomorrow night; from Roger Scully’s hints expect it to look awful for Labour

  27. Further points about Sheffield Central: a correction to my earlier note, Blomfield’s majority in 2010 was 165. The three main candidates are already known: Paul Blomfield (Lab), Natalie Bennett (Green) and Shaffaq Mohammed. Conservatives are an endangered species in Sheffield Central but could benefit from the collapse of the UKIP vote and there will be the usual fringe candidates (probably including the appalling Maxine Bowler, notorious from the ‘Comrade Delta’ SWP business). Blomfield is a popular and, crucially, respected MP locally. It was his Private Members’ Bill that led the way for the controls over the payday lenders. He is assiduous in his constituency duties, takes great care to keep in touch with local issues and he has successfully distanced himself from the Corbynistas (he was for example one of the signatories of the recent anti-Livingstone letter). He was one of the main people in the Labour Remain campaign. He is especially popular with students, which is useful as there are more of them in his constituency than in any other. He has a general reputation as a principled thoughtful politician. One of the good guys. The Sheffield Central Labour Party (one of the largest constituency parties in the country) has been largely immune from the activities of Momentum and had none of the problems that have afflicted places like Leeds or Bristol. Bennett has a high recognition factor (and zero charisma) and has made little impact in Sheffield since becoming candidate a couple of years ago, but assiduous work by the party locally and the student vote has given them four council seats, three of which are in Central. Mohammed is the leader of the LibDems on the council and is an inveterate carpet-bagger. He took three goes to get back on the council and had to be found a safe seat. He stood in the Brightside by-election last year and got 6%, no improvement on the GE figure, he might benefit from Remain sentiment but he will have to fight the Greens for that.

    The critical issue, however, may be something that is local but has nothing at all to do with Central or Paul Blomfield: trees. The local council has made a complete pig’s ear of a tree management plan in the suburbs with people getting arrested at 7 am for blocking Amey tree-fellers. This has added to a general impression held of them as incompetent and arbitrary (library closures, demolition of the Don Stadium, a botched town centre revival plan, proposals to sell off the iconic and much loved Central Library building to a Chinese hotel group, you name it they’ve buggered it up). They’ve managed to alienate substantial numbers of both key labour voting groups in the city: traditional working class in the North and East and progressive liberal middle class in the South and West. The Greens and the LibDems have made much of the trees debacle and it’s their local leaders who are standing against Blomfield. There are no general council elections in Sheffield although there is a by-election (outside Central) on May 5th in one of the wards that make up Sheffield Brightside. The results of that will be looked at with considerable interest locally.

  28. The Lib Dem target list looks a bit light

    Richmond Park was only 32 on the most leave list per Chris Hanretty

    https://docs.google.com/spreadsheets/d/1b71SDKPFbk-ktmUTXmDpUP5PT299qq24orEA0_TOpmw/edit#gid=579044181

    One only has to look at a twitter search of Hoey Lib Dem to see the wonderful photo of Kate Hoey and Nigel Farage sailing the Thames, and read the corresponding outrage to see that Vauxhall (number 10 on the list) is going Lib Dem and that doesn’t even feature?

  29. Summed up by Stephen Tall

    http://stephentall.org/2017/04/22/election-notebook-3-lib-dem-targets-tory-landslide-upside-coalition-of-chaos/

    “The question is: does Brexit upset that rule-of-thumb? Should the party be looking less at how we did in 2015 and more at which seats have the highest Remain vote-shares instead? In which case, suburban seats like St Albans (Tory majority 15,316) or urban seats like Vauxhall (Labour majority 12,708; 22,466 over the Lib Dems) come into genuine contention.
    After all, in 2005 the Lib Dems scored some spectacular swings against Labour in seats where the Iraq war was especially unpopular (such as in my own then home of Oxford East) while our so-called decapitation strategy against top Tories with slim majorities — including one Theresa May in Maidenhead — proved an almost complete failure. (The sole exception was Tim Farron in Westmoreland.)”

    So we should be looking at any seat with a large remain vote as a potential Lib Dem target, especially if the MP is opposed to Brexit.

    And probably also doing the reverse to add to Tory targets.

  30. A lot of useful work done there. Thank you.

  31. Re Scotland it’s important to look at the 2016 Holyrood results. The constituency boundaries are different to the Westminster seats, but you can still see trends. For example the Gordon situation: the equivalent seat was indeed lost by the SNP, but it went to the Tories, not the LibDems. And the Tories gained quite a few seats, mainly in the south of Scotland. The LibDems meanwhile gained North East Fife and Edinburgh West, so I reckon they should repeat both in June. They also held Orkney and Shetland easily (67% in both seats), despite intensive SNP targeting. I think the general view is that the Nats’ attempt to undermine Carmichael over ‘leak-gate’ has backfired in O&S because they made it too vitriolic and personal. People there don’t like that kind of intensive political attack, and they generally regard Alistair as a decent guy and a good MP who made a mistake, for which he apologised. They’re also not particularly keen on independence. So I think the LDs will be fine there. They should also gain EdWest, where they have selected an excellent candidate in Christine Jardine, and possibly East Dunbartonshire as well now that Jo Swinson is re-standing.

  32. I agree with Richard that the Lib Dem target list could do with being a bit more sophisticated. Places where Brexit is popular are likely to be a wash but places with a high remain vote have much more potential.

    I was surprised to not see St Albans on the list and had to look up the result, which reminded me how convincing the Tory victory was.

    It will take a big push though as the Labour and Green vote will need a massive squeeze. I’m not sure the Lib Dems will have the time resources or leadership to make it happen, could be close though.