Here’s something interesting. Normally the best prediction (or at least, the least flawed prediction) of how votes translate into seats is a uniform swing projection – that is, if a party has increased it’s national share of the vote by 5 percentage points, you add five percentage points to the share it gets in each seat, and vice-versa if it loses support.

Broadly speaking, uniform national swing has been a pretty good predictor of elections – or at least, a good starting point to analyse elections and where parties have done better or worse than average. However, it is mathematically inelegant, particularly when it comes to the extremes. Say a party loses 5 percentage points in its national support – what happens in a seat where they only had 4% at the last election? On a UNS model they’d get minus 1% support, which is clearly impossible. Generally speaking though UNS doing odd things to parties with minimal support doesn’t matter, as it is all about the marginal seats and predicting overall seat numbers, how well it models changes in support in safe seats is irrelevant.

However, it has been playing on my mind how it will work with the Liberal Democrats at the next election if they maintain their present low levels of support. How would it cope with a drastic collapse in support for a party? It varies from pollster to pollster, but roughly speaking the Lib Dems have lost about half their 2010 vote, about 12 points or so. To start with there were 57 seats where the Lib Dems got less than 12% support in 2010, they can’t lose 12% in those. Equally, if their vote did collapse to what extent would their sitting MPs be insulated from the fall?

The Scottish Parliamentary election gives us a chance to see. Below is a scatter chart of the Lib Dem performance in 2011 – plotting the change in the Lib Dem share of the vote in each seat against the share of the vote they recieved in 2007. Gold dots are those seats that were notionally or actually held by the Lib Dems in 2007, blue dots are non-LD seats.

The green line is what we would expect to see if there was a uniform swing – the Lib Dem vote falling by 8% in each seat. The red line is what we’d get if the Lib Dem vote fell proportionally to their support in 2007 – basically if they lost half their support in each seat. The actual distribution of dots is clearly closer to the proportional line than the uniform swing. If this was repeated at a GB general election then the Liberal Democrats would do even worse than a uniform swing would predict.

On the other hand, look at the distribution of the blue and gold dots – in seats where the Lib Dems had incumbency the Lib Dems did better than a proportional loss would have suggested (and they do worse than than this in seats without incumbency) – while the Lib Dems did end up losing all their mainland seats in Scotland, they did actually perform somewhat better in the seats they held… just not by enough to save them.

I wouldn’t presume to make models or predictions about what would happen to the Lib Dems at a general election on this basis… just that UNS national swing may not be a very good predictor if the Lib Dem vote does remain in dire straights at the next general election.


183 Responses to “The Lib Dem collapse in Scotland”

1 2 3 4
  1. Anthony,

    May I deduce from your post that YouGov have already put together somewhere a complete set of the election results?

    This is something the Scottish Parliament have been promising to prove “soon” for two full weeks, whilst the EC simply link to the BBC website.

    Any hopes of making yours public?

  2. Have to admit that I assumed that seat prediction websites used proportionality rather than applying UNS as a single number to all constituencies.

    I thought that was what UNS meant!

    I’m amazed that anybody could have applied the same swing in every constituency regardless of reality.

  3. Lab 42
    Con 38
    LD 9
    UKIP 4

    Labour’s lead seems to be firming up again from around 2 to 4 points. LDs back into single figures.

    Effective Lab majority of 60 on UNS. LDs at 10 seats (or less if UNS turns out not to be a good predictor in LD meltdown).

  4. The reason for bigger than UNS falls in seats they held is because even if vote loss is totally even any seat where the vote share is more than twice the drop in share will mathematically show this.
    Whilst UNS may be good for marginals proportionality is more accurate across the piece and it is clear that for most LDs incumbency is worth some votes and the same will apply at UK GE.
    BTW – Good piece of work

  5. On the economic front, April saw the highest public borrowing for the month of April on record, with borrowing reaching £10b – 50% higher than the expected £6.5b.

    Signs here that the Coalition’s approach to the public finances might be in trouble, especially set against weak growth figures.
    If the Conservative Party’s fortunes are linked to the economy, we could see the Con VI slipping again.

    Govt net approval has dropped over the last 2 weeks from -18 to -23.

  6. SWebb,

    The Labour lead was never as low as 2 points on Yougov (on average). I’d say there has been little, if any, change since the elections. If you take the other pollsters as well, I’d say Labour are around 2-3% ahead at the moment.

    However, I do think we will see the lead start to widen very soon. I expect Labour to open up a healthy lead by the end of this year. They need to really before the better economic news (and tax cuts/sweeteners) start to happen in 2014-2015. I am still confident that Labour can win in 2015 if they start to gain momentum now.

  7. Agreed a good piece of work partly because it illustrates what most of us knew..

    1) UNS has its limits.
    2) Proportional predictions have limits and
    3) Incumbency has an effect.

    A bit of fuzzy logic could come up with a formula that took a position between the UNS and proportional predictions and shifted it towards one or the other depending on incumbency.

    So what were the UNS and proportional figures for the 72 Scottish LibDem candidates (I think they didn’t stand in one) and the actual results to see what would work.

    Peter.

  8. The real line bisects proportional loss and uniform swing… the “golden mean”?

  9. AmbivalentSuppOrter

    EM has been developing a higher profile recently – “national mission”, attacks on Clarke, etc. This is critical to start boosting his personal ratings.

    I think Lab is preparing for a 2012 election following a wave of strikes and further gloomy news on the economy.

  10. The Tories will always be happy in the next few years as long as they are hitting 35+.

    Labour really needs to eat away at the high Tory numbers (like it did with the Libs), especially as the elections showed that Labour’s numbers (and the Tories, in all likelihood) are pretty soft at the moment. They need to gain momentum whilst the going is at its very toughest. Otherwise, things could get dreary and depressing in 2014.

    I remain optimistic.

  11. @Swebb,

    Yep, I agree. Labour should start planning for an imminent election because one never knows how tough things will get with the economy.

    Ed is still only building up steam. I think the next 12 months could be where he really demonstrates his leadership qualities.

  12. Anthony

    YouGov have the data to show whether the simplistic “LD votes went to SNP” is true, or whether the more complex scewnario, that canvass returns here suggested, is more accurate – ie Westminster ID LD votes wnt to a number of parties, but that there was a corresponding shift of voters from those parties to voting SNP.

    That would be a really interesting analysis.

  13. BBZ
    You expressed an interest (unless I have my snp posters mixed up again) about the by election in my Aberdeen ward. I have seen the figures for transfers (un-needed of course) and the main points were
    1 Labour and SNP dominated in second, third etc preferences more than anyone would have thought.
    2 Tories were excluded from everyone’s plans with very few transfers.
    3 Of paticular relevence to this thread, the Lib Dems did very poorly in atracting transfers, a previous strength. This bodes extremely badly for the Lib Dems in the looming Scottish local elections next May. It may not reflect movements of real voters, but again the losses of the Lib Dems in first preferences exactly mirrored the gains of the SNP. For comparison, last time in this 4 member ward, the SNP had the largest vote but stood only one candidate who was elected. Labour also stood one (me!) who was elected with quota. The Lib Dems stood two candidates, one with few first preferences but both were elected because of transfers. The Tory was fourth as an individual but was not elected because of a lack of transfers.

  14. swebb

    EM has been developing a higher profile recently – “national mission”

    Interesting. That term may resonate differently in areas where the “nation” concerned may be interpreted differently from what Miliband intended.

    Did he mean the Brits, and if so was that restricted to the relevant policy areas?

    Did he mean the English, and if so was that restricted to the relevant policy areas?

    No doubt a party whose NEC all but one live within the noose of the M25 may not even recognise the questions!

  15. @AmbivalentSupporter
    How on earth are Labour meant to eat away at the Tory vote like they did with the Liberals? The Liberals’ vote had many voters sympathetic to Labour, who’d probably voted Labour before, and considered themselves left-of-centre. Labour are not going to be recipients of disillusioned Tory voters.

    You can’t transplant local election VI over to G.E. VI because the political makeup of the locals is completely different. There were many areas that Labour didn’t stand, and the Liberals – and consequently their projected vote – were the natural beneficiaries. There’s also councils, like mine, where Independents make up the main opposition – whose voters will vote differently come G.E.

  16. Barney

    STV certainly provides an interesting situation where the elected are based not only on voter preferences, but on accurate guesses from the parties as to what level of support thay have and how transferable that vote is between candidates.

    While I have long been a supporter of STV, I have increasingly come round to BZ’s idea that to claim the status of “major party”, the party should require to put up as many candidates as there are seats available.

  17. “How on earth are Labour meant to eat away at the Tory vote like they did with the Liberals?”

    I’m talking in relative terms. They are not going to pulverise the Conservative vote to the same degree as the Libs – very obviously. But they can – and do need – to start eating away at the Tory support. I think they will achieve this in the next year. The Tories could well be hitting 33s or 34s (on average) by the end of the year IMO.

  18. For another bit of fun- Electoral calculus has a Scottish Westminster prediction tool here:

    http://www.electoralcalculus.co.uk/userpoll_scot.html

    Tonight’s miniscule YG Scottish sample (for illustration only: weighted sub- sample is 212 people…) =

    Con 17%
    Lab 38%
    LD 6%
    Nats 34%

    which on this tool leads to

    Con 2 seats (+1)
    Lab 40 seats (-1)
    LD 2 seats (-9)
    Nats 15 seats (+9)

    The table lists all the seats- Charlie K holds his seat: but Menzies, Malcolm Brucie and Danny A fall to Nats as do Eric Joyce and Gordon Banks on the red side.

  19. SWEBB

    My feelings exactly – Labour needs to be prepared for a snap election anytime after September this year OR be prepared to get the Lib Dems to realise that joining a rainbow coalition is their only route to salvation.

  20. Yeah, whatever..

  21. Craig

    The percentages quoted (thanks to Rob S for the HoC reference) are national equivalents. That means such things as not every seat being contested by a political Party or not every area of the country being contested are taken into consideration.

    That said, you’re right that local and national election are not the same. This is partly because some people will actually vote differently based on their assessment of the competence or otherwise of their local councillor(s) and local authority. It is also because turnouts are typically much lower and different Party supporters may turn out with differing degrees of enthusiasm.

    Normally one of the results of this is a ‘local premium’ for the Lib Dems. I commented on this just before the local elections (there are some figures for it there) and it showed that the Lib Dems normally got 6-7 percentage points more in local elections (in non-general election years) than they were currently getting in the polls.

    Interestingly, despite the Lib Dems having only around half the poll rating at the moment than they did in the years I looked at (10% rather than 18-21%) they still got a similar premium this year of +6% rather than half that.

  22. Neil A

    :-) Do I detect a slight note of cynicism as to Labour supporters postulating some future success?

  23. @ Rob – thanks for those interesting stats.

    More dire news for the LDs – they are in such big trouble and there’s nothing they seem to be able to do about it.

    Even Chris Huhne seems to be disppearing at break-neck speed … (although apparently his wife is driving).

  24. I am right in thinking that the Lib Dems would lose out more, when the constituencies are equalled out at around 75,000 electors. Surely in Scotland, this would make the current Lib Dem seats more vulnerable to Labour and SNP.

  25. Incumbency may help in some seats, though many LD “safe seats” only become safe in the last decade or so.

    There are examples where, when their vote is seen to be in collapse, a constituency specific opinion poll has appeared to accelerate the decline… giving the electorate a stark Labour/Tory choice.

  26. “More dire news for the LDs – they are in such big trouble and there’s nothing they seem to be able to do about it.”

    On the contrary. Only today Tim Farron says the May results herald the greatest opportunity in the history of the Lib Dems and wisely concludes that Nick Clegg is not unpopular.

    Crisis, what crisis ? :-D

  27. BARNEY CROCKETT
    You expressed an interest (unless I have my snp posters mixed up again) about the by election in my Aberdeen ward.

    Yes, I did express an interest in the Dyce/Bucksburn/Danestone by-election, but you’re certainly “mixed up” regarding SNP posters, of whom I am not one. If you look at my posts, you’ll see they’re in the colours Anthony provided for Plaid Cymru and have no logo. As an old, un-merged home-rule Liberal, they’re the nearest thing I could find to the old Liberal Party colours, the L-Ds having appropriated the orangey ones.

    But thanks for your interesting take on the transfers. It will be interesting to see how that relates to the Airyhall/Broomhill/Garthdee ward for the June 23 by-election, which clearly had a very different profile in 2007, with the 1st prefs going:
        L-D 37.6%, Con 27.4%, SNP 17.3%, Lab 12.8%, Ind 4.8%

    71% is in the Aberdeen S and Kincardine N plurality seat, with the remaining 29% in the Aberdeen Central plurality seat. The former was notionally a 7.3% L-D plurality ahead of SNP in 2007 whilst I’m sure you’re well aware that Aberdeen Central was a notion SNP plurality of 1.4% ahead of Labour. On May 5, both were taken by the SNP with pluralities of 22.1% and 2.5% ahead of Labour in both cases.

    Which of those results better mirrors the mood in the ward may determin whether the Con can take the ward from 2nd or the SNP from 3rd place.

  28. oldnat

    “I’m amazed that anybody could have applied the same swing in every constituency regardless of reality.”

    I said it was all rubbish often enough. The fact that all Scotland’s parties (except maybe the Greens) are, or at least were, very regional must enhance the effect.

    Ranking constituencies by proportional gains and losses would show up some interesting extremes: Argyll, Moray, Dumbarton, Skye, Caithness where there were known local factors.

    In some professions you would be disciplined or struck off if you produced work as unrofessional as that.

  29. The following will clearly turn out to be a very good predictive model of the fall in LD vote share in the Scottish results:
    1. A proportional swing model based on the seats where the LDs were not incumbent.
    2. Where the LDs were incumbent, reduce the prediction in 1 by around a half or perhaps a touch less than that.

    But the bottom line for the LDs is that, despite strong incumbency effects, the proportional effect meant that their vote share still fell by more than UNS in all but one of the Scottish seats that they held. If an eventual UK general election mirrors this pattern, they’ll be in even direr straits than the polls suggest.

  30. OLDNAT
    Have to admit that I assumed that seat prediction websites used proportionality rather than applying UNS as a single number to all constituencies.

    My proportionality calculator does show that it’s not as simple as that. The “official” results still aren’t available, but based on the BBC ones, plugging their plurality and list vote counts into my calculator gives:

      Party   BZ Actual Error
        SNP   70     69   +1
        Lab   38     37   +1
        Con   14     15   -1
        L-D    3      5   -2
        Grn    2      2   nc
        Ind    2      1   +1

    But the apparent accuracy is spurious, by hugely overstating the SNP plurality seats but giving full compensation on the list.

    BILLY BOB
    The real line bisects proportional loss and uniform swing… the “golden mean”?

    Spot on, I think, but how to apply it, especially just for plurality seats as Westminster seems destined to remain?

  31. As the Lib Dems were barely a factor in much of Scotland in any case, how different would the figures look if we stripped out all the constituencies where the Lib Dems got less than 10% in 2007?

    Areas where they had next to no support to begin with, your Thurrocks and your Barnsleys, are less of a factor. You assume they’re reduced to their core vote their, but that they never got far beyond that anyway. It’s the distribution between the seats they won, the seats that were marginals involving them, and the seats they were hoping to compete in in an election or two that will decide how the seats fall.

  32. Rob Sheffield

    It’s not the LibDem collapse that is the most significant feature of the Scottish election results: it’s the loss of Labour’s FPTP dominance in the central belt, something which I for one could not imagine I would ever see though I expected the SNP to be ahead by enough elsewhere to be the largest party.

    The Scottish Election Constituency vote was a FPTP vote with NO shy tories, 100% certainty to vote, a selection of local issues included and a huge 100% “sample”.

    We know that up till 2010 there were many who voted Labour for Westminster, and SNP for the Scottish Parliament. These are some of the anti-Cons, steadfast in that principle, but loyal to no party.

    Westminster voting intention is misleading for the Scottish Parliament and the converse is also true, but if, by the next UK election, these people see the SNP as the more effective anti-Con force, perhaps in support of a minority Labour government, they won’t need any persuasion to use the SNP as the vehicle for their hatred and contempt for the party of Thatcher.

    See what would happen in the Scotland Votes seat predictor using the constituency percentages:

    Lab 13 -28
    SNP 44 +38
    LibDem 2 -9
    Con 0 -1

    The SNP would have 64% of the Scottish seats on 45% of the votes, well past the FPTP tipping point.

    Before devolution and later referenda, that would have been seen as a clear mandate for the Scottish MP’s to convene elsewhere.

    All Scottish Labour seats in both parliaments must now be considered marginal.

  33. Phil

    An incumbency factor that affected two Highland constituencies was the retiral of two long serving, local and popular MSP’s. One had a reputation as an independent minded rebel and was far more popular than his party on the list. Added to that was a strong list MSP challenger.

    Independent interventions were also a feature elsewhere.

  34. John B Dick

    “See what would happen in the Scotland Votes seat predictor using the constituency percentages:”

    With the greatest respect- that’s total fanciful wishful thinking.

    Its been a fact of Scottish polling that a large body of voters split their VI differentially between Westminster and Holyrood.

    You Nats- prior to the recent election- were saying that Scottish Westminster subsets were of no use in predicting the Holyrood election because voting patterns were different in the different elections.

    Yet here you are trying it on with the reverse scenario ;-)

    The most recent *Westminster* Scottish voting intentions (albeit a miniscule sub-sample) are above.

    Labour lose one seat on Tuesdays poll :D

  35. @ Old Nat

    “YouGov have the data to show whether the simplistic “LD votes went to SNP” is true, or whether the more complex scewnario, that canvass returns here suggested, is more accurate – ie Westminster ID LD votes wnt to a number of parties, but that there was a corresponding shift of voters from those parties to voting SNP.”

    My gut instinct would be that the the most simple LD voters went over to Labour and the most politically attuned LD voters went to the SNP. I think this impact would probably be greater considering the relatively low turnout of the Holyrood elections compared to the Westminster election.

  36. @ Neil A

    “Yeah, whatever..”

    Maybe you were responding to something else but I think I share your sentiments. It would make zero sense politically for the Lib Dems to leave the Coalition and join some rainbow coalition. They’re taking a beating in the polls and they took it on the chin in the local election but they’re getting through a lot of their policy goals. And that’s really the whole point of politics. Making the positive changes that you actually want to make and using the power you have when you have it to get done what you want. The Lib Dems wouldn’t have this kind of power under a rainbow coalition.

  37. My quick run through of this suggests that it’s bad news for the Conservatives.

    Let’s assume that the Libdems have a big incumbency premium, but not one that prevents them from succumbing. Now, who will benefit? Obviously the party best positioned to pick off not only the most, but the most with a large polling margin.

    The Libdems are better placed to defend against the party which has made the least gain in polling, and right now that’s the Conservatives. So the benefit from generating incumbency based holds is going to come out of any gains the Conservatives had hoped to make on libdem seats, while Labour have sufficient polling margin to steam-roller their target libdem seats.

    Without the libdem gains to offset their losses to labour, the Conservatives would have a much larger net-loss than currently expected.

  38. Since you guys often post election results including those of by-elections, I thought I’d post my own special election results tonight in the second Congressional special election since last year’s midterms. In New York’s 26th Congressional District, long a Republican stronghold, a special election (by-election) was held today to replace Congressman Chris Lee (R-NY) who resigned earlier this year after a scandal.

    I’ll give the results as you guys give them:

    NY-26 (with 97% of precincts reporting):

    Kathy Hochul (Democratic Party and Working Families Party): 47% (48530 votes)
    Jane Corwin (Republican Party, Conservative Party, and Independence Party): 43% (43836 votes)
    Jack Davis (Tea Party): 9% (9495 votes)
    Ian Murphy (Green Party): 1% (1,130 votes)

    Dem gain from GOP :)

    I can’t really do an accurate swing calculation (I think you would calculate it as 26.6% from GOP to Dem) but for reference in the 2010 midterms, the Republican Chris Lee defeated his Democratic opponent 73.6%-26.4%. In 2008, Lee defeated his Democratic opponent, Alice Kryzan, in an open seat race 55%-40.5%.

    I know I’m biased but this is a stunning upset victory for the Democratic Party. We had attempted to pick up this seat in both 2006 when the incumbent was linked with the Mark Foley scandal and lost. We had failed again in 2008 despite having an open seat. The Democrats have almost no bench in the area (I think Kathy Hochul is a local town clerk). By contrast, Jane Corwin is a New York Assemblywoman. Hochul led the race wire to wire.

    Four final sidenotes:

    1. GOP Registration in this District overwhelmingly outnumbers Democratic registration. Obama lost this District to McCain in 2008. Republican Carl Paladino won this District against Mario Cuomo in last year’s gubernatorial race despite Cuomo drubbing Paladino statewide.

    2. The Tea Party candidate, Jack Davis, was the Democratic nominee in 2006. He also ran for the Democratic nomination in 2008. It appears that he took roughly equally from both Hochul and Corwin.

    3. The race was projected for Hochul by MSNBC and the AP with 71% of precincts reporting within about an hour of polls closing but I wanted to report it like you guys do so I waited until near 100% of precincts had reported (this was a very fast vote count until the very end).

    4. Hochul also received the ballot line of the Working Families Party. Corwin received the the ballot line of the Conservative Party and the Independence Party. In case you’re confused, New York has a number of smaller political parties (they used to have a Liberal Party until very recently) who sometimes run candidates of their own but sometimes nominate Republican and Democratic nominees to their ballot lines. You have in the past had Democratic candidates with both the Democratic ballot line and the Conservative Party ballot line and Republican candidates with both the Republican ballot line and the Liberal Party ballot line.

  39. My guess is most of the Lib Dem deserters to the SNP will return to the LDs in a GE and thus save many of the Lib Dem MPs in Scotland.

    What’s the point of voting SNP for Westminster? They might as well keep their strong local Lib Dem MP than waste their vote on the Scots Nats in that instance. Though, as a Lib, i’m very fond of the Scots Nats actually

  40. Btw….this makes me feel all warm inside though my local news informs me that the President once again had some royal courtesy snafus (apparently toasting when he should have bowed or bowing when he should have toasted)

    http://www.msnbc.msn.com/id/43147754/ns/world_news-europe/

    And though I think this article is meant to be sarcastic, this is a really good looking picture of your Prime Minister.

    h ttp://www.guardian.co.uk/world/2011/may/24/obama-patience-pomp-ping-pong

  41. And btw….just noticed this article

    http://www.guardian.co.uk/technology/2011/may/24/david-cameron-resist-internet-regulation

    Good for David Cameron.

    @ Ashley

    “My guess is most of the Lib Dem deserters to the SNP will return to the LDs in a GE and thus save many of the Lib Dem MPs in Scotland.

    What’s the point of voting SNP for Westminster? They might as well keep their strong local Lib Dem MP than waste their vote on the Scots Nats in that instance. Though, as a Lib, i’m very fond of the Scots Nats actually”

    I think Lib Dem deserters have shown that they’re up for grabs. They’re not automatic votes for Labour and can be wooed by other parties. The next GE is a long way off (unless Cameron gets mischeivous) but it’s entirely possible that current Lib Dem deserters could return to the fold.

  42. @Ashley

    “What’s the point of voting SNP for Westminster?”

    Simply put, if you’re an SNP supporter, you will be keen to see 59 SNP MPs in Westminster. Even 30 would be a tremendous coup since the SNP could declare that the majority of Scots favour the party of Independence.

    Another reason might be reducing the Labour majority, since Labour seems less inclined to compromise with the SNP than the Conservatives (remember Gordon Brown not even acknowledging Salmond’s 2007 win for about a fortnight?). While many Scots are inclined to the “anyone but the Tories” vote, I feel that they might be the better option for the SNP.

    Certainly, the past Labour domination of Scotland is something I don’t want. I could tolerate a broad mix, but not 2/3 of the seats to one party, election after election.

  43. Anthony

    Excellent post. I suspect that the ‘incumbency factor’ is actually a lot more complex though. I’d be suprised if there isn’t a strong correlation with a number of factors. If we had information on LD campaign effort, for example number of contacts, that might give us an even better measure.

    Why? Because my guess is that a lot of those seats had no more than a paper candidate… I would expect this to be collinear with incumbency – I’m sure more effort was spent in seats perceived as winnable…

  44. Perhaps a simplistic model could be created that predicts a 33% fall where the LibDems are the incumbents, but a 67% fall elsewhere? (This is based on the graph – of course the figures should be adjusted based on the current opinion polls.)

  45. @THESHEEP

    For what it’s worth, I’ve done a quick comparison on the two constituency Lib Dem seats from the Scottish Elections (2007 & 2011).

    I’m not sure quite how the Lib Dems have improved their share proportionally per Anthony’s comments, but I’m sure someone can explain.

    The graphs speak for themselves:

    http://imageshack.us/photo/my-images/11/shet.jpg/

    http://imageshack.us/photo/my-images/94/orkn.jpg/

  46. Mick Park
    “On the contrary. Only today Tim Farron says the May results herald the greatest opportunity in the history of the Lib Dems and wisely concludes that Nick Clegg is not unpopular.”

    I agree. ;-)

    The May elections do herald a great opportunity for the LDs – ie get rid NC
    And
    Clegg is not unpopular – it’s worse than that.

  47. Quick question –
    How is it to best weight the most recent and previous polls?
    Unweighted yougov 5 day average (versus recent poll)-
    41.6 (-0.4), 37.4 (-0.6), 9.2 (+0.2)
    Weighted average 1 –
    41.7 (-0.3), 37.6 (-0.4), 9.2 (+0.2)
    Weighted average 2 –
    41.65 (-0.35), 37.5 (-0.5), 9.2 (+0.2)

    Weightings for 1 – 1, 0.8, 0.6, 0.4, 0.2
    Weightings for 2- 1, 0.9, 0.8, 0.7, 0.6

    Also Q about judging polling figures vs historical figures –
    Since the ‘Others’ parties have burst in popularity in recent years, is it fair to judge current party performance vs historical performance or is it wise to also look at the figures weighted with ‘Others’ removed?

    This may seem like a pointless question – but if others start from a larger percentage, can it be completely fair to judge the big 3 on this if the big 3 did better when others started on a much smaller percentage?

    So when we had 2 party politics, then 3 party politics, now multi-party politics emerging, is it fair to judge the multi-party percent vs 2 party/3 party?

    Hope I’m clear there.
    Because if we remove ‘Others’ from the equation, LibDems have only had their worse showing since 1981 and Labour their best since 1997.

    Or should we judge performance on swings?

  48. This is a great piece of analysis.

    I’m reminded of NC saying that the coalition would break the mould of British politics…really quite prescient.

  49. Looks like I’m not the only pesssimist – h ttp://www.guardian.co.uk/commentisfree/2011/may/24/european-dream-single-currency

    Europe is unravelling and the business entranced European governments are doing pretty much everything wrong.

    My guess is that we are moving ever closer to the point at which banks will be forced to write down heavy losses and where the electorate in many countries reaches the point where they refuse to take all the pain for the mistakes of the financiers. I think we are about to enter Phase 2 of the financial crisis, and austerity will be proved to have been a waste of time and merely a delaying mechanism to avoid confronting the real issues.

  50. Mick Park & Mike N

    “Tim Farron says … that Nick Clegg is not unpopular”

    I believe he said Nick Clegg is more popular than Ed Miliband (40% to 36% according to MORI). You could have it that Ed is even more unpopular than Nick if you prefer.

1 2 3 4