Forecasting the 2015 UK General Election

CON 252, LAB 231, SNP 52, LD 41, UKIP 38,  DUP 9, GRN 9, PC 8, SF 6, SDLP 2, Independent 1, Speaker 1

Required for majority: 322 (CON 70 short, LAB 91 short)

The upcoming UK General Election has been called “the most unpredictable for almost a century”. There are several reasons for this; the rise of UKIP, the Greens, and the SNP, the collapse of the Liberal Democrats, who nevertheless seem to maintain highly localized popularity in areas where they have sitting MPs, and the fact that Labour and the Conservatives are going in to the election essentially on level pegging.

The focus of the media is on the UK-wide vote shares published by the main opinion pollsters, but as Mike Smithson from politicalbetting.com correctly says, the winner of the election will be settled by 650 first past the post races, and not the winner of the national popular vote. Given the fact that some seats have become three, four or even five way marginal seats, forecasting the winner has become an incredibly difficult task.

The latter-day god of psephology, Nate Silver at fivethirtyeight.com, correctly forecast the outcome of all fifty states in the 2012 United States Presidential election. However, his model was based on aggregating results from state opinion polls. This approach would not work in the UK, as there is not (and could never be) the same level of constituency opinion polls as there are for states in US Presidential elections. An alternative approach is needed.

I have built a forecast model based on the results of council elections across England, Wales, and Northern Ireland since 2011. This has meant combing through thousands of individual council races, and matching the ward level results to constituencies. For Scotland, I have taken Holyrood Election results and matched them to Westminster constituencies in a similar fashion.

I have had to “fill in the blanks” in a few seats, wherever a party did not stand in every ward in a constituency, which was especially prevalent for the Greens and UKIP in 2011 and 2012.

The final part of the model is to mark parties up or down depending on the prevailing national opinion polls. For England, I have taken the simple average of the four most recent UK-wide opinion polls, and the simple average of the four polls after polling day for each election day in 2011, 2013 and 2014, and used these to create a scalar factor to apply to each party in each constituency. For example, Labour’s polling average after the 2014 local elections was 33.8%, whilst at the time of writing it is 32.8%. Therefore all Labour results for 2014 have been multiplied by (32.8 / 33.8 = 0.97), i.e. a three percent penalty to reflect their decline in the polls. I applied similar adjustment to polls in Scotland and Wales, although polls are a lot fewer and far between, so I simply used the single most recent poll and the poll closest to election day to calculate the adjustment scalars. I didn’t apply any adjustment for Northern Ireland at all; polling is unreliable and public opinion tends to be relatively consistent.

Results and Forecast 2

The map above shows the results of the 2010 General Election, and the forecast result for this year’s election using the local election results to forecast Westminster results. A spreadsheet showing the forecast winner, forecast second and third places, and their forecast votes shares can be found here.

There are a few things that are immediately obvious. Firstly, both the forecast totals for UKIP and the Greens are a lot higher than consensus estimates. I think that it is possible that the lack of precedent of the “big two” of Labour and the Conservatives polling such a low combined share (less than 60% in some recent  polls) means that commentators are being (small “c”) conservative in their estimates of how well the Greens and, particular, UKIP may do. I have spot checked some of my forecast results with the constituency level polls published by Lord Ashcroft, and they are in the same ballpark. Or I may be completely wrong. We will see.

So, what would happen if these forecast results became reality? In all likelihood, there would probably be a second election in short order, because it would be very difficult for either Labour to form a coalition, and all but impossible for the Conservatives, even though they would have “won” in terms of seats (but not the popular vote).

Given Sinn Féin’s abstentionist policy, the magic number for a majority in the House of Commons is 322. A Labour-led coalition in such a scenario would require both the Liberal Democrats and the SNP to get 324 seats, giving the coalition a majority of two. A Conservative-led coalition would require all of the Liberal Democrats, UKIP and the DUP to get 340 seats, a majority of 18. However, it is almost impossible to imagine Nigel Dodds, Nigel Farage, and Nick Clegg (or whomever gets the gig of Lib Dem leader) all agreeing to be in a coalition with each other. It would be weird.

Both David Cameron and Ed Miliband will be praying that the minor party surge is only a flash-in-the-pan, and that in the election itself people will return to their “natural” tribe and vote for one of the big two. Otherwise the next parliament could end up being an ungovernable mess. It is very possible to envisage a scenario where Nigel Dodds or (probable SNP Westminster party leader) Alex Salmond wield real power, as their Westminster blocs may be needed to prop up a government. One can imagine that both men are busy drafting their wish lists for what additional resources they would like to see being dispatched to Northern Ireland and Scotland in the event that they find themselves at the centre of national attention in May. They could call it the coalition of the shilling.

, , ,

  • Ernekid

    These estimated figures seem totally wrong there’s no way UKIP will win 38 seats.

    There’s only one poll that matters and that’s on election day. Trying to project polling on a national level is meaningless

  • LucidTalk

    This article is hardly worth commenting on. It’s total rubbish. You can’t take council elections and project them into a national UK election.

  • Muiris

    ‘It’s just a bit of speculation, with a touch of science thrown in, like wondering which of two raindrops will reach the end of the windowpane first (or in this case, the order in which 6-7 raindrops will do so). Of no importance, good for a few ‘I told you so’s, if near the bull’s eye on election night, and likely no less accurate than loads of other punditry out there.

  • Catcher in the Rye

    38 UKIP seats is many multiples of what the serious commentators are suggesting. I stopped reading after that.

  • Zeno

    We’ll find that out in a few months time when the results come in.

  • tmitch57

    Salmon, good luck with your model. I believe that trying to predict how many seats UKIP will get is a bit like trying to predict how the Know Nothings would do in the 1854 midterms in the United States: they had no record to go on or project off of and most of the Know Nothings were actually running as Whigs or Democrats. UKIP may prove to be every bit as ephemeral and temporary as the Know Nothings were. The thing that UKIP and the Know Nothings share in common is that both exploited/will exploit a great popular dissatisfaction with the existing parties and politics in general. And UKIP could also prove to be of great historical importance by allowing one of the two main parties to exploit the results just as the Know Nothings ultimately facilitated the rise of the Republican Party during the 1850s and led to the Civil War.

  • Dan

    Let’s hope it’s multiples of 38 on the day.

  • Robin Keogh

    I feel bad that u have spent hours and hours designing this 🙁

  • Nicholas Whyte

    Salmon,

    Thank for this refreshng perspective. I think you probably do flatter both UKIP and the Lib Dems, but at least you have provided detailed data to back it up. Did you consider also factoring in the European election results? They are available (with a little digging) at local council level in England, Scotland and Wales.

  • salmonofdata

    Thanks. I fully expected to get a kicking for predicting such a high UKIP total, but I think that the fact that they are unpopular by many is leading to widespread confirmation bias; people don’t want it to be true that UKIP could pick up many seats, so they don’t believe it to be true. UKIP have polled 20% in recent days, with Labour and the Tories polling less than 30%. It seems highly unlikely that UKIP wouldn’t start to pick up seats if these vote shares were replicated in the election, especially considering that their vote is concentrated in clusters such as Kent and Essex where they already have sitting MPs.

    I didn’t include the European results on the grounds that it might have flattered UKIP by including the votes of those who only vote UKIP in European elections. Although, if people think that UKIP can’t win first past the post contests in nationwide elections, they might want to look at this map, showing which party came first in each English council area in the 2014 European election.

  • Paddy Reilly

    Who is purple and who is blue?

  • salmonofdata

    Blue = Conservative, Red = Labour, Purple = UKIP, Orange = Lib Dems

  • mjh

    Salmon Of Data

    Firstly respect for the enormous amount of work that this represents. It certainly highlights the complexity and diversity of the political landscape.

    I can just see those “We are winning here” leaflets that will be produced by UKIP and the Greens in the constituencies where your figures put them ahead.

    Others have pointed to the issue of using Council and Holyrood elections as the base for Westminster predictions – there is plenty of evidence that a significant number of people vote differently in different in different types of elections (even when they are held on the same day) and there is also the potential that the low levels of turnout at local elections affect the results. The consistently better performance of the Lib Dems at local level suggests that this could also apply to the other smaller parties.

    Would it be possible to look at the level by which these parties total local results have historically outperformed their total Westminster results and apply a factor to your constituency predictions to account for this?

  • Nicholas Whyte

    Looks like Tim Farron is safe then, if nobody else…

  • James Martin

    As much as I enjoyed looking through this, I am afraid to say that I suspect it is pretty far off what the reality will be. UKIP will be lucky to win five seats let alone 38 I suspect, as their support will be too evenly spread (like the Lib-SDP Alliance in 1983). That said, I found your analysis of NI very interesting. Do you really think that SF will take North Belfast? And that the DUP will take South Belfast? I find both of those results unlikely due to tactical voting in those areas. I would like to hear whether you have factored in any (a) incumbency bonus and (b) the idea of tactical voting.

  • LucidTalk

    Farage said on Sunday Politics today that they were aiming for 3-5 seats – it’s hard to get from that up to 38. + Greens will hold Brighton Pavilion & that’ll be it ie 1 seat for them.

  • salmonofdata

    Cheers. I’m aware of the limitations of using local elections as a means of predicting general elections, but due to the paucity and expense of polling individual constituencies I don’t think there is much else. Although I believe there is a strong link between the two types of elections; for example, it is striking how well the Lib Dems are doing in town halls where they have a sitting MP. It is through the graft of building up a local support base that the Lib Dems were able to establish a foothold in Westminster in the first place. UKIP have managed to build up bastions of council election support even since the last general election.

    I thought about trying to weight for parties outperforming at local level as opposed to the general, but there is a lack of historic data to extrapolate from for UKIP, so I didn’t. One of the interesting things I thought was an Ashcroft poll that showed UKIP well in the lead in Plymouth Moor View for preferred party at Westminster, but they lost support and got pipped by the Conservatives because a number of UKIP supporters didn’t think that the UKIP candidate could win. I think that concerted campaiging of the “winning here” sort could close that gap. That goes for the Greens as well.

  • LucidTalk

    When did Ukip poll 20% ‘in recent days’, as you say above?

  • salmonofdata

    Opinium/Observer General Election Poll, 16/01/2015;
    CON 28% LAB 33% LD 7% UKIP 20% GRN 6%

  • Guest

    If you’re interested in this subject then why not one along

  • LucidTalk

    Thanks, I missed that one! However the Ukip average is around 15-17% in recent polls, and under the FPTP system that means they could end up with 3-5 seats on that score & the LibDems on 7% could end up with 30 seats.

  • LucidTalk

    If you’re interested in this subject then why not come along this Thursday evening (29th January – 6.15pm) to the Holiday Inn, Ormeau Avenue, Belfast, and hear the views and predictions of Peter Kellner (YouGov) one of the UK’s top polling experts. For more details, and how to register: http://www.lucidtalk.co.uk/latest-news/latest-commentary-from-peter-kellner-yougov-president/

  • Ian James Parsley

    Unfortunately I have to go with LucidTalk here – albeit for a different reason.

    You simply can’t project overall poll results (be they ballots actually cast at recent elections or opinion surveys) on to the 2015 General Election.

    For example, Ashcroft’s polling shows that in 13 key LibDem seats (11 where an incumbent MP is defending at two where a well-known local LibDem is challenging), the LibDem party share is about 20% but the number stating an intention to vote LibDem taking into account the candidate rises to 35% (note well Nicholas Whyte!) – national polls take no account of that kind of thing, yet it is increasingly relevant.

    In the absence of serious well-known local candidates, I would suggest UKIP will win 3 and the Greens 1. They will be no more relevant to Government Formation than our own local parties in such a scenario (i.e. not at all, frankly).

  • Zeno

    I think it’s a little bit early to make solid predictions. I imagine you agree with that. Though you do seem pretty certain about the Greens. I’ve no idea how many seats UKIP will get, and haven’t even looked at it, but I’d say while the Greens would be largely predictible UKIP are not at the minute.

  • mjh

    I take your point about the difficulty of adjusting your predictions to take account of the difference between local council voting patterns and Westminster, particularly in the case of UKIP. It means that you would have to make some assumptions – but that is no problem provided that the assumptions are clearly stated, their impact is made clear, and a margin for error is indicated.

    For example the elections in England in May 2009 for County Councils (representing more rural and small town areas) gave:
    Con 38%, Lab 23%, Lib Dem 28%, Others 11%

    Those for Metropolitan and Unitary authorities (the urban areas) in May 2010 gave:
    Con 35%, Lab 27%, Lib Dem 26%, Others 12%

    While the Westminster election on the same day gave:
    Con 40%, Lab 28%, Lib Dem 24%, Other 8%

    Crudely in the Westminster election the, the Con vote was about 3% points higher than would have been expected from the council results; Lab was also about 3% points higher; while Lib Dems and Others were about 3% points lower.

    Very, very crudely this would suggest that the vote totals indicted by the council elections in England should be adjusted upwards by about 10% for the Conservatives and Labour and down about 10% for the Lib Dems. For UKIP and the Greens it should be down somewhere between the 10% of the Lib Dems and the 30% of the Others.

    Scotland could be adjusted on the basis of the difference in voting intentions for Westminster and Holyrood over a given period. For example the latest poll which asked the same people for their preferences for both was Survation this month. It also showed that the Conservatives and Labour had an advantage for Westminster. While 12.2% of the sample would vote Con in the First Past the Post Constituency vote for Holyrood 14.2% would do so for Westminster. So the Cons had a Westminster advantage of 2% percentage points, Lab of 0.5%, while the SNP dropped 4.2% points.

  • Julia

    You cannot take those type of voting patterns and project it onto a GE!
    E.g. The LibDems target seats built on strong activist zones built up over many years. The extrapolation of the council data leads to a figure of three times their likely GE performance.

    It’s not worth any further comment!

  • John Earl

    Then why comment?

  • LucidTalk

    Zeno – I agree, the situation is very fluid & it’s risky for anyone to make predictions. However the first-past-the-post system is very difficult for Ukip & the Greens – not so much for the SNP, DUP etc. as their vote is concentrated in a specific region. Frankly at this point, there is only 1 certain seat that Ukip will win and that’s Clacton – Yes Farage is favourite for Thanet South, but even he is not 100% acc. to the polls and the bookies.
    John Earl: I said it was ‘hardly worth commenting on’ not that it was ‘totally worth not commenting on’!

  • salmonofdata

    Cheers for the tip about the Peter Kellner talk, that should be pretty interesting.

    Clacton, as you say, is pretty much nailed on for UKIP. They are strong favourites in Thanet South and Great Yarmouth as well, and have a very good chance of holding on in Rochester. The UKIP vote is actually quite clustered regionally, they have pockets of support in the Southeast, the South, the East Coast and the West Midlands. They are virtually friendless in London and the North West. Add in the fact that it is quite possible that a rise in support for the Greens could split the left vote, and it is quite possible to start winning seats in a first past the post system with quite low shares (i.e. in the thirties), if there are five parties competing in the double figures nationally.

    The Ashcroft constituency poll at http://lordashcroftpolls.com/wp-content/uploads/2014/12/LORD-ASHCROFT-POLLS-Marginal-seats-poll-report-December-20141.pdf is interesting reading. This poll put UKIP ahead or marginally behind in Dudley North, Great Grimsby, Plymouth Moor View and Rother Valley. If UKIP can win these seats, there is nothing to stop them winning dozens more if their national share starts to climb higher than the ~17% it is now.

  • Callum Traynor

    no hes not aiming for 3 to 5 hes never said 3 to 5 hes never gave a number

  • Callum Traynor

    just shows what a joke our system is

  • Interesting to see this forecast! Will definitely be a close call with the results and to see if we would have a 2nd election too.
    For those interested in a short summary of the paries in this 2015 election, then check this out: https://hmsies.wordpress.com/2015/03/03/uk-elections-2015-making-your-mind-up/

  • John Lennox

    38 seats for UKIP? You are utterly deluded. Try 4 or 5.

  • Just found this …
    Fascinating prediction, well done for the work that must have gone into it.

    With the advantage of two months of extra information, I’d suggest:

    1. As others have suggested, UKIP support is not even, and where there is support, it seems to grow from nowhere, overnight – I live near Thanet and Dover (not to mention Rochester!). So your prediction is probably on the nail.
    2. On the other hand, Green support seems far more evenly spread, making significant gains unlikely, whatever their vote total (My local kent greens are putting much more effort into planning to shore up Brighton, rather than the futile local campaign).
    3. As always, I reckon predictions of LibDem collapse are premature; indeed, people may come to recognise their role in moderation, despite slurs from both Tory and Labour.
    4. SNP I reckon you’re dead on: regardless of the referendum, Scots can’t trust any party except the SNP to protect their interests in Westminster – in future years, we may see the SNP beaten at home, but still all-powerful at Westminster, provided they play their cards right. And Salmond has proven time and again that the more LABCON attack him, the stronger he is back home (they never learn!).

    But of course, as you admit, it’s all conjecture; it’ll be fun to compare your predcition with the real thing!

  • Koopalover9000

    I think your overestimating. UKIP are just another Tory party And they will get 5 seats at the most. The greens will get maybe 2 if another labour area turns green. I think we will have a labour minority government.

  • Koopalover9000

    I agree