Lord Ashcorft has published some new polling of marginal seats, full details here. As with the ComRes marginal poll in the week the seats polled were mostly ultra-marginal seats – in this case, the 12 most marginal Con-Lab seats, the 12 most marginal Lab-Con seats, but whereas the ComRes poll was a single sample representing the most marginal 40, these were 24 individual samples, one from each seat. Ashcroft also added two seats that are less marginal, but thought to be good for UKIP: Thanet South and Great Yarmouth.

The fieldwork for Ashcroft polls was done between the 31st March and the 18th May. During that period the average Labour lead in the national polls was about 3.5 points: that’s the equivalent of a uniform swing of 5.25%. The average swing in the twelve CONSERVATIVE ultra-marginals seats was 5.5%, The average swing in the twelve LABOUR ultra-marginals was 6.5%.

That means that in contrast to the the ComRes poll in the week, the swing from Con-to-Lab in Conservative ultra-marginals is pretty much in line with the national swing, a fraction of a percentage point better for Labour compared to the national figure. In Labour-held ultra-marginals the swing is a little larger, which is what we’d expect to find (parties do a little better in seats they hold due to the incumbency effect of the local MP).

It’s not a very exciting finding – swing in Conservative marginals not vastly different to other seats – but it’s one that gives me some confidence in the poll. The reality is that come general elections marginals as a group are not usually vastly different to other seats. The swing is sometimes a little bigger or smaller, new incumbents normally do a little better, but the contrast isn’t normally vast.

While I excluded them from the sums above (as they were selected because they were unusual, so would have skewed the sample) I should comment on those two extra seats polled – Thanet South and Great Yarmouth. Both were chosen because there was an expectation that UKIP would be doing well, and in both cases it proved to be true – both had them in a very strong third place, with 28% in Great Yarmouth and 27% in Thanet South. Their strongest performance though came in a seat that was part of the normal sample of ultra marginals – Thurrock, where Ashcroft found them at 29% and in second place behind Labour. Thurrock was also one of the seats where UKIP did extremely well in Thursday’s local elections.

UPDATE: Actually I’ve just spotted that the fieldwork in the Tory held seats was done earlier than the fieldwork in the Labour held seats. So comparing the swing in Con-Lab seats to the swing in national polls at the time the polls were done shows no difference at all (both show swing of 5.5%). Comparing the swing in Lab-Con seats to the swing in national polls at the time those polls were done shows Lab doing about 1.5 points better in seats they already hold.


241 Responses to “Ashcroft poll of marginal seats”

1 2 3 5
  1. With regard to your last comment AW on Thurrock’s UKIP gains. I am still a little baffled that UKIP made about 50 gains more than the 100 that they were widely expected to make on a pre-predicted showing in the mid 20 percents. If they have actually slumped by 6 points since last year to only 17% – how come thay have gained 50% more seats under FPTP than was predicted even though they have slumped this much?

    Is it a) They have concentrated their vote much much better than last year?
    or b) The PNS is dodgy with regard to the methodology used for getting the UKIP share right?

    Which brings me back to my question to you of a couple of threads ago. How do the BBC treat UKIP share of the vote when they field only one candidate in a three member ward?

  2. Now we just need the euro results to make the media storm over the last 24 hours look very stupid indeed.

  3. UKIP really need to be ahead, not just strongly placed, in Thanet South (where I am sitting at the moment). Great Yarmouth and Thurrock if they are going to get one or more MPs elected at the General Election. although a Farage effect might push them past the post in Thanet South if he were the candidate

  4. NickP,
    Unless UKIP come first, as most polling predicts?

  5. Unless UKIP come first, as most polling predicts?”

    The media seem to have assumed that result. But I suspect Lab will be top, and if they ain’t it will be fractional.

  6. @TF

    I think that’s a racing certainty.

  7. @NIck P

    On what basis do you believe UKIP will not come top?

  8. NickP

    I agree the media have been over the top about UKIP, as a DM reader, this mornings paper, was really rather crass. If they win the Euro’s it’ll all start again.

    TONY DEAN

    It’d due to London where UKIP are not strong and in many cases did not even field candidates. If you exclude London they probably did as well as last year if not marginally better.

  9. Let us imagine a typical London three member ward

    Lab1) 16%
    Lab2) 15%
    Lab3) 14%
    Con1) 11%
    Con2) 10%
    Con3) 9%
    UKIP1) 8%
    LibD1) 6%
    LibD2) 5%
    LibD3) 4%
    Ind1) 2%

    Should the UKIP share of the total vote be treated as the 8% their one candidate got – or 3 times 8% (24%) as if they had fielded 3 candidates?

    What do posters here think is the most accurate method for estimating UKIP support if all voters had had just one vote instead of 3 in a 3 member ward?

  10. Tony – I was a bit dubious about that 23% last year and suspect it was a bit high, which would mean the fall is a little lower. Anyway, apparently in those locations where it it was directly possible for Curtice’s team to compare 2013 to 2014 UKIP had fewer votes this year, so the fall is genuine, even if one is uncertain about the scale.

    The two approaches to dealing with multi-member wards are take the highest of each party, or the average of each party. I’ve no idea which the BBC take, but obviously it doesn’t make so much difference when a party has only one candidate.

  11. raf

    polls were showing ukip support shrinking before the actual ballot.

    We’ll see.

  12. Thanks AW – that’s helpful. Back to the number crunching for me!!!

  13. @ RAF

    Many polls only had UKIP top for the euros based on 10/10 certain to vote; when 6-10 was considered, Labour often came top.

  14. How many of the 26 seats polled by Ashcroft were in London, given that London appears to be trending very differently from the rest of the country?

  15. (As posted on the previous thread)

    On the Lord Ashcroft poll, his choice of Con-Lab marginals is bit disappointing. Of the 26 seats selected, 12 are already Lab held and only 2 of the remaining 14 require a Con to Lab swing of more than 1% for Labour to retake them.

    So it’s not really a poll focused on the key seats which will determine whether or not Labour will get an overall majority, or even become the largest party. With the exception of those two seats (Thanet S and G Yarmouth) it is a a poll of the lowest hanging fruit which Labour needs to pick up, plus the fruit which Labour already had (just) in the bag in 2010. Labour could (and should) win all of these and the remaining seats where it needs a 1% swing in 2015 only to find Cameron and Clegg putting together another coalition with a workable majority.

    The question is: can we extrapolate from a marginals poll mainly of Labour low hanging fruit into the other marginal seats that will really determine the outcome of the GE?

  16. I would think that the fall in the PNS would point to UKIP performing below expectations in the Euros. We’ll at least discover which is the best predictor – the polls or the PNS.

  17. It’s true that media organisations have their own agendas, but let’s not forget that they also have access to privileged sources of information, such as exit polls.

    On the basis of comparisons with 2009, surveys of voting intentions over the last few weeks, and reasonable deductions from material published since Thursday, I expect the UKIP share of the vote in the European elections to be very considerably higher than its share in the local elections. My guess is 32-36%.

  18. @ Phil Haines

    Agree with your post about the selection being a bit disappointing. A nightmare scenario is that the Tories pick up the same number of LD seats that Labour takes from Cons – and we are then back to square one effectively about becoming the largest party in a hung situation!

  19. A fair few CON supporters on Twitter quite downbeat about the Ashcroft poll but it is worth emphasising, as Anthony shows, that it is in line with the national trend.

    A few potential issues here:

    1/ He hasn’t looked at any LD seats. I think we can safely say that a fair few will go LAB but the result could be decided by how many CON can get their hands on. Scottish seats might also be critical, though it is probably pointless polling them until after the referendum.
    2/ The questions don’t talk about local MPs which are always going to be a major factor at constituency level.
    3/ There are lots more seats that could potentially change hands than this sample of ultra-marginals.

    On UKIP it is hard to say too much. They are right up there in a few seats there but who knows which is the best chance? The Farage factor might, I guess, add a few votes if he chooses well (I’m pretty certain he’ll go for Thanet South). I fancy Great Grimsby as UKIP’s best chance actually with Austin Mitchell retiring. However, that poll doesn’t quite support the figures yesterday playing them ahead there.

  20. *sighs*
    I’m just going to repost my commentary from the other thread here since my timing was off…though this gives me additional material, added at the end:

    To swing back to UKIP, one worthwhile exercise for right now is to take the combined votes of the BNP, UKIP, and English Democrats from 2010. One thing I am finding on looking back is that in some places where UKIP did well this time, the combined vote of the three is >15% (and it is often >10%). For the moment, I will take a few examples:
    Rotherham: BNP: 10.4%, UKIP: 5.9% Total: 16.3%
    Doncaster North: BNP 6.8%, ED 5.2%, UKIP 4.3% Total: 16.3%
    Thurrock: BNP: 7.9%, UKIP 7.4% Total: 15.3%
    Don Valley: BNP 4.9%, UKIP 4.4%, ED 4.0% Total: 13.3%
    Doncaster Central: ED: 4.4%, BNP 4.2%, UKIP: 3.4% Total: 12.0%
    Sheffield Brightside: BNP: 7.8%, UKIP: 4.1% Total: 11.9%
    Sheffield South East: BNP: 5.7%, UKIP: 4.6% Total: 10.3%
    Sheffield Heeley: BNP: 5.5%, UKIP 3.7% Total: 9.2%
    Great Yarmouth: UKIP: 4.8%, BNP: 3.3% Total: 9.1%
    Basildon and Billericay: BNP: 4.6%, UKIP: 3.8% Total: 8.4%

    Doncaster is interesting because it was the EDs piling up the votes there this time. Worth considering is that if UKIP doesn’t stand everywhere, the EDs might end up doing well in those areas.

    (I exclude Castle Point because Bob Spink was an incumbent MP and nominally not UKIP’s candidate, but adding his campaign to the BNP vote gets nearly 32% there! This was also a strong seat for UKIP in 2005.)

    This isn’t a hard-and-fast rule, obviously, but it looks like where those three did well in 2010 combined, the support has mostly consolidated to UKIP. This does broadly speak to the narrative of UKIP largely consolidating a scattered vote on the right that had been going multiple ways. It also points to a good pile of votes that were “lost” years ago, and which have likely been augmented by a couple of factors centering on UKIP’s rise.

    This actually leads to a worthwhile question: Why didn’t UKIP do better in places like Barking and Dagenham (where the BNP pulled 12 seats in 2006, and where in Barking in 2010 UKIP+BNP was 17.7%)? My best answer there is that UKIP just didn’t bother to stand enough candidates (for example, they only stood one candidate each in Alibon and Goresbrook, two of BNP’s wards that elected two councillors). This has been said elsewhere, but that actually might hint that UKIP’s “London Problem” was in no small part an organization issue, not solely a lack of support.

    Added footnote: Ashcroft’s poll gives the above some more support. Were he doing the poll today, I would be extremely interested in seeing what would come out of those Northern areas UKIP did decently in. One thing of interest is Walsall North (UKIP on 32%). It isn’t one I had looked at on my own, but here’s what I have:
    Walsall North: BNP: 8.1%, UKIP 4.8% Total: 12.9%
    UKIP picked up three seats here…so it is an accidentally good pick for looking at UKIP’s support in those seats (and it incidentally gives the highest VI for UKIP anywhere!).

  21. RogerH – at the risk of extrapolating it a bit, I think it suggests quite a close finish.

    We don’t have local election polling, but if we assume a relationship between Euro vote and local vote that’s similar to the relationship between Euro vote and Westminster vote then we’d expect about a quarter of those Tory voters to back UKIP at the Euros (so CON about 22, UKIP an extra 7) and about one in ten Labour local voters to back UKIP (so Lab 27 or so, UKIP and extra 3 or so). That would give us something in the region of Con 22, Lab and UKIP 27. But then theres the Lib Dems and the effect of the Greens and all those minor parties…to many assumptions piled upon assumptions to really go further…

  22. Re the choice of seats made by Ashcroft I guess he probably went for these because he did it for the Con Home ‘Securing a Majority’ conference. So the focus was on how CON can win, not how LAB could win. Obviously an equivalent Labour event would prob look more to the second tier of marginals – the likes of the Enfield seats and Loughborough.

  23. “A nightmare scenario is that the Tories pick up the same number of LD seats”

    Is that likely, though? Why would the Tories perform better than the LibDems in a LibDem seat?

  24. @AW: “RogerH – at the risk of extrapolating it a bit, I think it suggests quite a close finish.”

    Thanks. I have £6 on Labour coming top!

  25. I’m still sticking with a Labour majority in May 2015. Their great strength is that their electoral base is wider than any of the other parties: first-time voters and ethnic minorities are more likely to vote Labour and Labour are the largest party in Scotland (in terms of Westminster seats).

  26. @AW
    “……too many assumptions piled upon assumptions to really go further…”

    I did spot something that would add further to the mix when the Euro papers were sorted yesterday. I’ll try and comment on Sunday once the polls are closed, before the results are announced.

  27. @ RogerH

    I think it is perfectly possible for the Tories to gain LibDem seats but lose ground to Labour at the same time unless the LibDems recover a bit!

  28. @ Des Brown

    “I’m still sticking with a Labour majority in May 2015.”

    I wish I could share your optimism! After Thursday and recent polls I feel it is slipping away……

  29. Some further thoughts on the Ashcroft poll in particular:
    (1) UKIP generally does better in the Labour seats than in the Tory ones not picked as UKIP targets.
    (2) In Dudley North, Great Grimsby, and Walsall North UKIP beats the Tories. They are hot on the Tories’ heels in Southampton Itchen and Telford. The same applies to Thurrock.
    (3) As Anthony said, there are a lot of factors that are leaning towards UKIP in Thurrock. To his factors, I would add that only 52% of Tory voters ruled out voting UKIP, and the strongly-correlated performance of the BNP+UKIP in 2010. This raises the specter of tactical voting, especially with a credible poll like this out there. The other seats I listed have similar phenomena:
    -The Tories are blown into third
    -The Tories are also not competitive with Labour (this may well foil UKIP in Great Yarmouth, which shows as a three-way marginal)
    -Strong local results for UKIP where applicable (not all seats had underlying local results this time)
    -UKIP+BNP+ED is pretty high:
    –10.8% in Great Grimsby
    –13.4% in Dudley North (+0.4% for a National Front candidate)
    –12.9% in Walsall North (as noted above)

    If I was a UKIP backer, I’d be working hard to arrange polls of similar seats, and then I would make no end of hay about “[Tories/Labour] can’t win here, but we can stop [the other party]” come election time.

    Final thought for right now: Ashcrofts poll doesn’t surprise me at all. I’ve been saying for quite some time that I expected that UKIP would find competitive areas based on the regional breakouts in polls.

  30. TOH/NFR

    Surprised it wasn’t easy to understand.

    Suppose someone GAVE you a million quid, with the provision being you had to bet on largest party in 2015 GE and then give the money to your favourite, very very needy charity.

    You HAVE to bet. Which party you prefer is irrelevant.

    What would your head tell you to do? What do you think most people who’ve thought about it seriously would do?

    I wouldn’t even think about it – even if it was the Ole-Gyp-Of-The-Knee-Foundation.

  31. (When my last post gets moderated through, I would love to know what tripped moderation of it…)

  32. @ Phil

    I did spot something that would add further to the mix when the Euro papers were sorted yesterday. I’ll try and comment on Sunday once the polls are closed, before the results are announced.
    ———-
    Now we’ll be in suspense x 2 awaiting the actual result and whatever it is that you’ve spotted!

  33. That’s great Anthony.
    Very helpful-thank you.

  34. As each of the Ashcroft marginals, has an equal sample size, it should be possible to work out the average swing using a simple average. The sample splits as follows, using data from Ashcroft’s summary table:

    Lab held seats (12): Con-Lab swing 6.5%
    Con held 1% or less marginals (12) Con-Lab swing 5.5%
    Gt Yarmouth/Thanet S (2) Con-Lab swing 7.0%

    Make of that what you will.

    On that basis, the whole sample gives a swing of 6.1%, not the 6.5% quoted by Ashcroft, which is puzzling.

  35. @TD: “I think it is perfectly possible for the Tories to gain LibDem seats but lose ground to Labour at the same time unless the LibDems recover a bit!”

    Why, though? Are LibDems going to switch to Tory? I just don’t see it happening in any significant numbers (if at all).

  36. @roger

    we’re already seeing it in places like kingston and portsmouth

    Tories gaining from Libs as some of their supporters ditch and go to Labour splitting their vote more.

  37. MiM

    NickP has certainly been consistently bullish for his parties prospects .

    I suppose there is something admirable in that sort of partisanity -and absolutely vital in the street campaigning one imagines.

    I heard GO at the ConHome Conference this morning , telling his troops that hard work on the streets amongst the voters is the order of the day.

    Never having been in that role or ever wanting to be , I can only admire the commitment in those who are.

    I think there is some sort of distinction to be drawn between partisanity at the coal face-and partisanity behind a key-board.

  38. @ Gray

    “One thing of interest is Walsall North (UKIP on 32%).”

    That was the first question (standard VI question). For the second question, respondents were prompted with four of the parties likely to contest Walsall North (Con/Lab/LD/UKIP) with the fifth option, “another party”. If “another party” was the reply, they were prompted with Green/BNP/”some other party”. UKIP’s support slipped to 30% on this “local” or “constituency” VI question.

    Interestingly, in 21 out of 26 seats the UKIP vote was lower on the second question than the first, and only in two (Thanet South and Great Grimsy) was it higher on the second question.

  39. Also, in Walsall North, Conservative support fell -1 point between questions 1 and 2, UKIP (as mentioned) were down -2, LD up +3, and Others up +1.

    I should mention however that 30% was still UKIP’s highest score of all the constituencies on the second question.

  40. @ Roger & Mitm

    Not only as Mitm says to Labour, but I’ve noticed that where LibDems did well in Camden and Islington in 2010 the Greens are significantly up too in 2014!

    So if ex-LibDem Lefties like me go Lab or Green, the Cons could pick up some LibDem seats quite easily, especially where the pre-2010 LibDem support was largely Left-wing based.

  41. I don’t suppose there’s a table anywhere telling us who parties lost/gained council seats to/from? Those net changes don’t begin to tell the whole story.

  42. Yes Robin – I would be fascinated to know how many LibDem council seats were lost to the Conservatives!

  43. Very surprised Lutfur Rahman has been re-elected

    I really thought (and hoped) Labour were going to win the mayoralty there after it was revealed what he was like, guess targeting certain demographics really can work.

  44. I was at the count for the council elections for one of the Labour held marginals which Ashcroft polled. A third of the councillors were up for election. Except in one ward where two were. A very high proportion of voters had not cast their votes twice for the same party. UKIP had only fielded one candidate. (A cock up by them. They had not realised that there were two vacancies until the day nominations were due in.)

    Even the postal votes, where one might have thought that voters had time for considered reflection of how to vote, threw up a lot of odd results. Eg multiple examples of voting for: UKIP and one of the Lib Dems; one Conservative and one Labour; UKIP and a Labour candidate with a clearly Asian name who had campaigned as “a proud British Sikh”. It was a Lib Dem hold with Labour a close second, which was what was expected. Both Lib Dem and both Labour candidates lived in the ward and one of the Lib Dems was their candidate for the constituency in 2010 and their PPC for 2015, so there was a lot of scope for ‘personality voting’; sure enough, a high number of ballots with one Labour and one Lib Dem vote.

    The point of all of the above is that IMO it seems to offer a concrete indication that a surprisingly high proportion of real voters in a real ballot are torn between what we might consider non-intuitive options. It will be interesting to see how they decide next year when they only get one vote. (In each of the General and local elections.) One might assume that it will average out, and it probably will, but the insight into the mixed ‘loyalties’ was fascinating, educational and a little bizarre.

    PS. Some of you may well have worked out where I am writing about.

  45. “we’re already seeing it in places like kingston and portsmouth”

    Lots of odd things happen in local elections. I still don’t agree that the original suggestion – “A nightmare scenario is that the Tories pick up the same number of LD seats that Labour takes from Cons” – is remotely likely in a GE.

  46. “I think there is some sort of distinction to be drawn between partisanity at the coal face-and partisanity behind a key-board.”

    Not sure meself as there is no such word – not even in forrun.

  47. From Ashcroft polling:

    Which parties have you ruled out voting for

    LDs 56%
    UKIP 52%
    Cons 49%
    Lab 38%

    (I assume this is answered by DKs)

  48. @AW
    “UPDATE: Actually I’ve just spotted that the fieldwork in the Tory held seats was done earlier than the fieldwork in the Labour held seats. So comparing the swing in Con-Lab seats to the swing in national polls at the time the polls were done shows no difference at all (both show swing of 5.5%).”

    Nonetheless, if the Conservatives are banking on newly elected incumbents in these ultra-marginal Con seats building up a personal vote of several % to help counteract a national swing to Labour, the evidence from Lord Ashcroft is that it isn’t yet happening.

  49. Of course, if one is ridiculously partisan there is the word

    “Partisinsanity”

    Like some people are about good old Arsenal for example.

    “THAT WASN’T A FREE KICK !!!!!!!!!!!!!!!!!!!!!!!”

    “THAT WAS A CORNER !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!”

    “THE FIRST THREE GOALS WERE LUCKY !!!!!!!!!”

    etc etc

  50. @PHIL HAINES

    I did spot something that would add further to the mix when the Euro papers were sorted yesterday. I’ll try and comment on Sunday once the polls are closed, before the results are announced.

    I think I have a good idea of what you spotted but I’ll be sure to tune in.

1 2 3 5