Lord Ashcroft has published a new batch of constituency polling. I hesitate to call it marginals polling, since we’ve moving up into some less marginal territory with today’s polls. Ashcroft has polled four different groups of seats in this set (all the tabs are here.)

The first is the next cohort of Lib Dem -v- Conservative marginals, this group are those seats with a Lib Dem majority of between 9% and 15% over the Conservatives, so we are no longer looking at ultra-marginals. The average swing from the Liberal Democrats to Conservatives in these seats is 2%, nowhere near enough to win seats like these. However, as we’ve seen in previous Lord Ashcroft polls of Lib Dem marginals there is an awful lot of variation between individual constituencies – some seats (Carshalton & Wallington and Thornbury & Yate) are actually showing swings from Con to LD. At the other end of the scale two seats are showing large enough swings for the Conservatives to win the seat (North Devon and Portsmouth South, which has a chunky 9 point swing from LD to Con, presumably at least partially connected to the scandal around Mike Hancock).

The second group of seats consists of two more Lib Dem seats with Labour in second place. Lord Ashcroft’s previous polling in LD v Lab seats essentially showed a complete Lib Dem collapse, raising the possibility of an almost complete wipeout for Liberal Democrat MPs where Labour was the main opponent. One of the seats here – Burnley – follows that pattern, with a ten point swing from LD to Lab. The other, Birmingham Yardley, represented by John Hemming, bucks the trend. There is still a 2.5% swing from LD to Lab, but it is smaller than we’ve seen in other LD -v- Lab seats and would be small enough for Hemming to hold on.

The third group of seats is two unusual seats – the close three-way marginal of Watford, and Wyre Forest, an Independent seat between 2001 and 2010. Neither of these really fit into any broader category, but looking at them as individual seats Watford shows little relative movement for the three main parties – all are down a little, UKIP are up a lot but still in fourth place, meaning the Conservatives retain a narrow lead. Wyre Forest was held by Dr Richard Taylor between 2001 and 2010. He’ll be standing again come the next general election for the National Health Action party, but I think under the same Kidderminster Health Concern label that he won on in 2001 and 2005. Ashcroft’s poll currently has the Conservatives holding the seat on 32% with UKIP in second on 27%, Labour 16%, Lib Dem 7%, Green 5%, Other 13%. The others aren’t identified in the poll, but is presumably largely Dr Taylor’s supporters.

Finally Ashcroft polled three of the four seats that will be contested by the main party leaders come the election – Sheffield Hallam, Doncaster North and Thanet South (presumably he didn’t do Witney because he thought it would be too boring… it would seem there comes a point when even Lord Ashcroft saves his money!). Party leaders normally do pretty well in their own seats. It is extremely rare for them to lose their own constituency and they very often outperform their party nationally. Such is the collapse of Liberal Democrat support however people have seriously raised the possiblity of Clegg losing his own seat – Ashcroft’s poll has it very close. Clegg is on 31%, Labour on 28%, just three behind (and this is on the question prompting people to think about their own constituency, the standard voting intention question had Labour a point ahead). Ed Miliband’s Doncaster North seat is traditionally a very safe Labour seat that should pose no concerns for him, but there was some speculation about how well UKIP might do. The BNP have held their deposit there at the last two elections and their was some significant support for the English Democrats too, with the far-right parties now collapsing and UKIP hoovering up that right-wing protest vote it looked as if there could be some potential. In fact Ashcroft’s poll did find UKIP in second place in Doncaster North, but 12 points behind Ed Miliband. Finally Thanet South, the seat where Nigel Farage plans to stand at the general election. Current figures there are CON 34%, LAB 26%, LDEM 7%, UKIP 29% – so UKIP in a strong second place, but not currently quite enough to send Farage to Westminster.

529 Responses to “Ashcroft polls of the Lib Dem battleground and leaders’ seats”

1 7 8 9 10 11
  1. @OldNat

    All the best with to forthcoming hospital sojourn. I hope all goes well. We don’t always agree about much, and I’ve pulled your leg recently about the frequency of your posting, but you’re the sort of knowledgeable and witty contributor that makes this site worth visiting. Accordingly, get well soon.


    Welcome back. You’ve been missed. Has Arsenal’s win on Wednesday night brought you back? :-)


    Your post was a scholarly piece of work, if I may say so

  2. @ Unicorn

    Great post. That was the question I was asking the other day about how well your model worked and clearly it has worked well!! I think there is still a bit question mark hanging over what happens when the LD’s start to squeeze with two horse races etc but certainly on the info you have available you seem to have done a good job.

  3. @Unicorn

    Very interesting analysis.

    When the see the formula for UNS vs your regression equation, it’s very easy to see why the regression works better.

    The UNS formula takes no account of concept like minimum level of support. If you take the 2010 levevl of support and knock off the average fall, seats with a higher 2010 LD support may not fall by enough, and low 2010 votes will end at zero or less (clearly nonsense).

    I have looked at the LD VI since 2010, and it has been very stable (stable bad). Therefore a good regression fit look quite possible. The challenge for regression is how it handles the Conservative vote, where definite peaks and troughs have occurred. Based on You Gov data in 2014 the R Sq for each party is as follows:

    Lab 0.57
    LD – 0.46
    UKIP – 0.36
    Con – 0.04

    Straight line regression with an R Sq of 0.04 ought to be dreadful.

    I think your analysis shows that the UNS model looks hopeless, given the different factors in place this Parliament.

    Very well done.

  4. Of course, the other issue with regression with as a whole is the the fact is built on lots of hard to measure ‘soft’ variables.

    Should, for example, one party announce a left field policy that radically moves VI, the assumptions that support the regression model have fundamentally changed. Therefore, an analysis mixing based on data before and after this change will be measuring apples and pears.

    I use regression (I’m a Quality Engineer in plastics), and it works because the key processes work in a known way. Applying x pressure for y time with cooling of z degrees is based on firmly established chemical and physical models.

    I’m not too sure the soft variables behind VI are reliable and repeatable, as they based on human perceptions and various rational and irrational judgements that can change.

    Still, good work anyway :-)

  5. Thanks all (esp for the smartphone advice)

    I’m now getting that musical ad. Seems someone is wasting their cash, since while “You can get it all here” they don’t say where “here” is.

    Of course, it could just be Anthony promoting this site – in which case we already knew that you can get everything here. :-)

  6. @The Other Howard

    I’ve seen much to disagree with here sometimes, but I draw the line at this.

    Jeremy Brett’s portrayal of Holmes, as good as it was, is not an accurate portrayal of the original Holmes. Who was described in the text as a bohemian figure more akin to Shelly and Wilde. He wore shirts *without collars*!

  7. Apparently we are going to be getting a lot of new Polish British citizens before May 2015 as an immediate outcome of yesterday’s speech. Not sure who they can vote for though…presumably not UKIP or Conservative.

    MsTPole @MsTPole
    Everyone at Polish school this morning is talking about getting a British passport.Id be worried, Cameron – they’ll be allowed to vote then.

    The Lib Dems and Greens should be organizing a citizenship/ voting registration drive!

  8. Amid the continuing confusing message from all these polls, I thought I’d look at whether a trend could be detected pre and post Rochester. To simplify matters, I’ve only looked at the average VIs for Labour and Tory and the average Labour lead. It makes quite interesting reading.

    I’ve taken as a start point the first poll that showed a Tory lead on November 9th. After that, and more or less right up to the Rochester by-election, there were a total of 17 polls in which the Tories led in 6 (!), and Labour 11. The average of all those polls more or less indicated a dead heat with the Tories on 32.1% and Labour on 33, giving Labour an average lead of only 0.9%. As I said, more or less a statistical dead heat.

    Since Rochester, we’ve had 9 polls, two of them ties, one with a Tory lead and 6 showing Labour ahead. The Tories have slipped to an average VI of 31.3% and Labour have increased to 33.5%, stretching their average lead to 2.29%.

    Nothing dramatic, but a definite trend developing post Rochester.

    The weekend polls will be fascinating.

    Off to watch some footy.

  9. Has anyone here/elsewhere tried using logistic curves instead of a linear model of how vote numbers go up and down within a constituency? These vary between 0 and 1, with the middle being pretty much the same as a linear model.

    With linear UNS it’s possible to get locally to get negative voting predictions, with a logistic curve that problem would disappear.

    Most large parties (large enough to bother modelling) seem to have a maximum and minimum level of support nationally which are not 100% or 0%, and it seems reasonable that this is true at a constituency level too.

    As the level of support gets near the limits, fewer people will change their minds: either they are diehards for the particular party in question, or else diehards for someone else and won’t shift. A logistic curve model would seem to follow this pattern much better than UNS (which purports to predict the votes of people who don’t exist).

  10. @Unicorn Congratulations as everyone else has said. After you have done so much work it is hard to ask for a bit more, But is it possible to assess the effect of whether the opposition is Conservative or Labour? Other things being equal, one might expect left leaning Lib Dems to be less likely to defect when the alternative is a Conservative opposition than when it is Labour. What happens if you put a dummy variable representing this into your regression?

  11. @ OldNat

    Best of luck for next week!

    If my experience of hospitals is anything to go by, you’ll be far too busy people-watching to worry about wifi access.

    Looking forward to seeing back here again soon.

  12. couper2802

    You must be very young as Neil Kinnock’s treatment was far worse than anything thrown at EM & he didn’t have the benefit of social media #ImBackingEd campaigns

    Alas I’m not that young and can remember the nasty campaigns against Kinnock. But Kinnock actually got a lot of support at the start, it only got really vicious in the last few years as he became more likely to get elected. Whereas Miliband was under attack from day 1, possibly for the same reason. And the attack on Miliband has been from all sides, including a lot of sniping those media that would normally be favourable or neutral (Guardian, BBC and so on).

    Whether it is as effective is another matter. The Press is less powerful than it was and the very fact of its being so relentless means that there is little left in the locker to throw (unless you’re impressed by the thought of Pressman’s secret celebs). And as you point out social media can act as a counterweight – though the failure of the Yes campaign, despite its dominance in that area, should remind us that social media don’t reach everyone.

    Whether it been dealt with well by the Labour Party is another matter. They seem to be relying on a “With one bound he was free” strategy, when the voters will suddenly realise Ed is a superhero after all during the election campaign when he will get a fair hearing. Certainly there have been some episodes where he has been effective and that has shown in short-term improved ratings. But shifting public opinion is slow and heavy work and you need to be relentless to counter the relentless attacks.

    Labour have also missed a trick in allowing their underdog position to be hijacked by UKIP. After all they are the ones who the real establishment elite are attacking and most of the stresses that people are complaining about are caused by the policy of the current government (admittedly often only intensifying those of their Labour predecessor). And the ‘big issue’ of immigration is one driven by the demands and failures of employers – Labour’s traditional antagonists.

  13. seeing you back, of course.

  14. Jayblanc,

    Who cares? Sherlock Holmes is just the poor man’s Sexton Blake.

  15. Actually, now that I think about it, a lot of modern adaptations of Sherlock Holmes basically ARE Sexton Blake!

  16. Roger Mexico
    I am exceedingly angry with David Cameron and have lost a proportion of my desire to protect his reputation. He has backed himself and the party, right up Excrement Avenue regarding immigration. The over confident positioning in 2010, has led to a huge failure, therefore all future promises on immigration, are believed by Tories alone, and not all of them. If Dave thinks that one kipper will be turned by his excellent speech the other day, he is mistaken.
    He has promised to much to often and fallen flat on his face. We Tories might very well give thanks for Miliband, but Farage gives thanks for Cameron.

  17. Sorry for the amteur/newbie question but what nights do the regular trackers usually come out on? Are we due a YouGov tonight?

  18. amateur even.

    Additionally, how often does AW update the poll average on the right?

  19. @Funty

    Mon-Thu nights from Tue to Fri morning polls (4 in total). The Sunday poll comes out on Sunday, funnily enough,making five for the week.

    I think AW updates it periodically (i.e. when it suits him).

  20. There is left wing social media now which is a negative for [the Conservatives] but then they will do a job on UKIP during the campaign while plenty of yoof commentators will bang the Green drum.

    The campaign will make 92 look like kindergarten and unlike in 92, is being coordinated between News UK and two of the other Tory backing papers. Not surprisingly [the Sun’s] forte will be the celebs ! Though [they] will also have a key New Labour architect involved.

    [Can you please stop posting like a NewsUK spokesman, and please stop revelling it in so much – it just provokes other people to make partisan and sarcastic replies and is generally non-conducive to non-partisan discussion – AW]

  21. Cheers Statgeek

  22. The irony of Poles getting passports is that they tend to be conservative. My evidence of this is their party system (centre-right vs right wing). So that would be another set of conservative immigrant votes that the Tories have thrown away.

  23. @Lurker

    Not necessarily — some of them may decide the Tories are the lesser evil compared with Labour, and the Tories may find it worth their while to moderate their stance.

    I think that most of the Tory posturing isn’t worthwhile for them, given that a lot of the switch to UKIP is people who will say that they are concerned with immigration, but don’t necessarily base that on a lot of information about the parties’ respective platforms. If this is the case (and the Tories realise it) they may be able to row back a little and placate this set of voters.

  24. Lurker

    So Irish folk living in UK vote according to the Party choices available in the Republic, and those from Jersey search for non-party affiliated candidates?


  25. @CMJ

    Thanks for your encouraging and helpful comments.

    I fully agree that there are risks in using regression in this way. I used a sample of summer polls to establish the parameters of my LD VI-shifting regression model and the underlying assumption is that all data points are drawn from the same overall population. If a significant event occurred half way – as you imagined – then the analysis would try to impose a single pattern on the two disparate subsets. So you’re quite right: it is subject to all the GIGO (garbage-in-garbage-out) problems that you have to handle in exercises of this kind.

    In this sense it is no better or worse than UNS. UNS tried to work out current VI by using a single parameter: absolute change since the election. The linear regression model uses two parameters – some vertical change (intercept) plus a slope parameter (to capture the change linked to the initial VI). Multiple regression could take this further independent variable (e.g., incumbent = 1; non-incumbent = 0). The regression equations are just using extra parameters to predict future VI. (Modellers endlessly discuss the downsides of adding extra parameters).

    I should also point out that using the regression approach could well return with a verdict that uniform swing is a good model of VI change. If the best fit has a slope of zero, then this shows that VI change is completely unaffected by starting value, and here the two models become indistinguishable.

    In your later comments about low Conservative R-squared values I suspect you are refering to a completely different exercise: one in which the independent variable is time/date and not the initial VI we are using here. However you general point is well taken. In order to be helpful the regression equation has to account for a good proportion of the variation in the data. In relation to this, I should repeat that the figure *was* quite high: r-square = 0.88 and so I think there are good grounds for using the equation here.

  26. “Apparently we are going to be getting a lot of new Polish British citizens before May 2015 as an immediate outcome of yesterday’s speech. Not sure who they can vote for though…presumably not UKIP or Conservative.”

    Or Labour, who have announced similar policies to the Tories, so just leaves the LDs.

  27. Hoof hearted

    “so just leaves the LDs”

    Or the Greens. I note that the E&W Greens membership now exceeds 27,000.

    Other growing parties also exist in some locations. :-)

  28. @Unicorn

    I think you have judged the risks wisely. It is important to understand what your model doesn’t say as much as what it does say. In my professional capacity, if I over stated what the data is saying, I would quickly get shot down in flames. Therefore, I do apologise if I come across as picky, it’s just apply my caution with data.

    I do commend your application, as sifting the data as you are can result in many dead ends. With a bit of luck you will find a golden ticket!

  29. @Unicorn

    As you have the dataset from Lord Ashcroft’s polls, what do your results for the other parties look like?

  30. HH,

    The crucial difference between the Conservative and Labour positions is that Labour are not talking about deporting non-working EU citizens. That is what is driving the new rush for UK passport applications.

  31. Roland is correct, ukip voters won’t be swayed by anything DC has to say, all it will do is justify their views and keep immigration and the EU in the headlines, and this can only strengthen ukip, especially as Labour are also following the same retheroric. As a result both Lab and Cons, could lose voters( to not voting, greens or LD)

    DC is leading us to the EU exit, he has already failed because he can’t get a cap through, and his reforms will at best only be accepted by the EU, again another failure. He is alianating the UK from his European partners again.
    We also see European(German) politicians on the news saying the UK is less important than than France and not indispensable! Exactly why so many want out of the EU.

  32. @Hal

    I think the proposal to make it harder for spouses to accompany partners will also be driving that. Currently you have to earn over a certain amount to bring your spouse to live with you here. Polish families don’t want to see their families torn apart so will be forced to apply for citizenship. Unintended consequences and all that…

  33. @Hoof Hearted

    I suspect that former Con and Lab voters who share UKIP’s view on Europe and Immigration have left already, so in that sense can do little extra damage to their VI.

    I don’t expect UKIP to score much above current levels – in fact come the GE I expect a few points being knocked off.

  34. @Unicorn

    ” Multiple regression could take this further independent variable (e.g., incumbent = 1; non-incumbent = 0).!

    So what about taking it a bit further and adding a further variable 1 main opponent conservative and 0 main opponent Labour? Might help you to explain difference between general voting intention and specific intention in that constituency.

  35. Average Labour leads this month:

    All polls: 1.33%
    YouGov: 0.75%
    Non YouGov: 1.95%

    The non YouGov figure is boosted by Populous, without which it would be 1.45%. Even so, It used to be that YouGov showed some of the larger leads. What’s changed to now have them showing much lower leads than the average?

  36. ” i suspect that former Con and Lab voters who share UKIP’s view on Europe and Immigration have left already, so in that sense can do little extra damage to their VI.”
    I disagree, because millions of immigrants must be turned off by Labour and Conservatives’ rhetoric, either not votng , or voting LD/ green.

  37. Good news for UKIP. An endorsement by a former party leader: http://www.buzzfeed.com/jimwaterson/nick-griffin-is-now-voting-for-ukip

  38. @ Thiswillhavetodo,

    Populus are (and have been since they began their online panel) the least responsive pollster to movement from the big parties to the small parties, because of the way they do their weighings. They failed to pick up the rise in Ukip in 2013, which made them favourable to the Tories for most of that year, and they’re currently failing to pick up the rise in the Greens and the SNP, which is making them favourable to Labour.

    Apart from YouGov the other pollsters don’t poll frequently enough for us to distinguish trends from noise at these time scales. Over the course of a month, some of them would only have one or two polls. In addition, some of them like Ashcroft have small sample sizes that make them particularly volatile. A few random good polls for Labour can easily inflate the average lead for the whole group.

    Without looking at the averages over a longer time scale, I’m not confident that YouGov are actually showing lower leads than the rest of the pack once you eliminate Populus.

  39. @ Thiswillhavetodo,

    Taking a quick look at the figures for the month, it seems Yougov have higher Green figures than the other polls, so I would guess the difference lies there, as higher Green is probably related to lower Labour.

    And the reason for that I would guess lies with past vote weighting of unknowns/ turnout weighting. You only have to look at the last Comres to see how that changed a Green score from 6 to 3. So it probably has something to do with the rising Greens this month.

    You can see the Greens rose over the month with Yougov on Statgeek’s chart here


  40. @ Unicorn,

    Impressive work! Now we have an extra reason to look forward to Ashcroft’s next round of marginal polling- I’m eager to see how your model performs in the Con/Lab seats.

    @ Old Nat,

    Good luck with the operation, and may your recovery be as swift as the rise in SNP membership. ;)

    The trouble is, Brussels is its own worst enemy. It behaves like the Mafia and its maladministration is so shocking, that it is beyond parody. So, however good you, (or I ) think the idea is, it has become a symbol of graft, fiddling, greed and nest feathering to the British worker. Cameron’s desire to change huge chunks of it, is perfectly valid, however, nobody believes he can, and the kipping community don’t believe he means it anyway.

  42. Roly

    Good to see that you are maintaining your reputation for never overstating things. :-)

  43. OLD NAT
    I considered stating what I really thought.

    You just take care.

  44. Nick Griffen tweets that he will vote for Ukip.

    This should help the Tories win back a few deserters to Ukip. No-one wants to be seen to agree with Griffen.

  45. RAF

    ” No-one wants to be seen to agree with Griffen.”

    Would that that were true. However, every No voter in the Scottish referendum did precisely that – and every poster on this board who opposed a Yes vote was a Griffen ally.

    (Not that I’m suggesting any of those folk were racists – just pointing out the fallacy in your argument).

  46. Tonight’s YouGov
    CON 32%, LAB 34%, LD 7%, UKIP 15%, GRN 6%

  47. Roland

    I am exceedingly angry with David Cameron and have lost a proportion of my desire to protect his reputation. He has backed himself and the party, right up Excrement Avenue regarding immigration. The over confident positioning in 2010, has led to a huge failure, therefore all future promises on immigration, are believed by Tories alone, and not all of them. If Dave thinks that one kipper will be turned by his excellent speech the other day, he is mistaken.
    He has promised to much to often and fallen flat on his face. We Tories might very well give thanks for Miliband, but Farage gives thanks for Cameron.

    Cameron has the typical PR man’s tendency to say whatever he thinks will please his immediate audience, without really bothering whether it makes much sense or will work in practice. He then couples this with the typical Old Etonian belief that the servants will somehow clear up the resulting mess[1]. This can’t always be done and results in all sorts of U-turns and ‘explanations’ and briefings saying different things for different markets when reality bites. Even with a compliant media to help cover the cracks, a lot of the public tend to notice.

    There’s no point in the Conservative Party complaining about it though, because that’s why you elected him. He promised that he would produce whatever you wanted, even if it made no sense. And he did exactly the same thing with promises over a referendum over Lisbon (which was clear at the time couldn’t be fulfilled) which meant that no one now believes him on his promises of a referendum over leaving.

    Though I do actually have a bit of sympathy with Cameron on the immigration issue. Because current Conservative attitudes are so incoherent that any immigration strategy, now matter how harsh or liberal, would be bound to offend. They want Brits to have freedom to do whatever they like in Europe and everyone else to be restricted in what they can do in the UK. They want to stop the flood of foreign workers and to be able to recruit anyone they want for their own business. To have foreigners kept under tight surveillance and themselves not be stopped in the street for ID all the time.

    And UKIP believe exactly the same contradictions only more so[2], but aren’t in power and so can continue to agree to impossible demands. As Cameron always has, hoping that charm and rhetoric will be enough to see him through in a job which he “thought he would be good at”.

    [1] Not that he’d ever dream of using the word ‘servants’, he’d say ‘staff’. Unfortunately because those working for politicians are often called ‘staff’ as well this makes it easy to continue this attitude into the real world.

    [2] We should probably open a book on, during the GE, how many UKIP candidates are discovered to be employing illegal workers/paying Romanian migrants under the minimum wage/etc.

  48. @OldNat

    I hope the op went well.

    As for Griffen and the referendum while he may have shared a specific view on a binary question with many others, the others were not seen to be agreeing with him.

    Griffen’s support for Ukip on the other hand could cost Ukip dear, as Ukip under Farage have made a huge effort to publically distance themselves from the BNP (with whom Griffen even today is inextricably linked).

    How many times have we heard people say they will consider viting Ukip but they would never vote for the BNP because the latter is perceived as a racist party and the former not. And no-one (or very very few people) want to consider themselves to be racist, let alone racist across a range of issues.

    The argument holds (in a slightly modified form).

  49. RaF – ON knows that he is just being … how can I put it yes I know – he is just being Old Nat.

1 7 8 9 10 11