Monday, 28 April 2014

The untactical right: Conservative voters in Lib Dem/Labour marginals

I have mainly concentrated on the Labour/Tory battleground. I make no apology for this: this is where the election will be decided and this is where most of the constituency betting opportunities are. But for completeness, I ought to put up a brief post on the Lib Dem/Labour marginals.

I have already looked at the likely behaviour of 2010 Lib Dems in these seats. I came to the conclusion that those voters who have defected to Labour have made their minds up and will not be returning. This leaves the question whether the Lib Dems can encourage more Conservative supporters to lend them their votes.

In favour of this notion, Tories who had previously thought that the Lib Dems were surrogate socialists may have changed their minds following five years of coalition. Some prominent Lib Dems have enthusiastically made the case for austerity - Danny Alexander for one. (Others, however, have been Hamletesque in their public anguishing over getting into bed with the evil Tories.)

Historically, the Lib Dems have not had much success at this (in stark contrast to their ability to harvest Labour tactical votes when they are competing with the Conservatives).  I attach a list of the Lib Dem seats where Labour are in contention:

The average Conservative vote share in these seats is nearly 19%.  Admittedly, four of these seats have Conservatives in second place at present, but even without those four, the average is still 17%.  The Conservative vote in these seats has remained resilient.

To date, the polls have betrayed no hint that such Conservative voters are amenable to the idea of tactical voting, but that is not necessarily that informative, since these voters have not thought about the idea before and they may not have processed the concept as yet this time either.  Will they do so this time around?

This is one of the unknowns of next election. My default expectation is that we will not in general see the Lib Dems tap into this potential voting resource. Too many Lib Dem MPs have been publicly too lukewarm about the coalition to inspire any kind of affinity from Conservative supporters. Before 2010, many Labour supporters saw the Lib Dems as a kid brother to their own party. The Lib Dems in general have done too little to inspire a similar light affection in the minds of Conservative supporters.

As an added disadvantage, with the heavy swings against the Lib Dems, in many seats Conservative supporters will reasonably conclude that the Conservatives stand at least as good a chance as the Lib Dems of getting elected. Why vote for ginger beer rather than champagne when champagne is also on the menu?

Some MPs may be able to appeal credibly for tactical Tory support. Anecdotally, Danny Alexander and Nick Clegg have both been seen as positive contributors to the coalition by Conservatives. Nick Clegg hardly needs the votes, but Danny Alexander does. But his is a constituency where the Conservative vote has already fallen to a relatively low 13.3%. I am doubtful whether he can get that all that much lower.

In summary, while this in theory looks like a tempting source of votes for Lib Dem incumbents, in practice I think they are going to find these voters tough going to convert. I'm very bearish about Lib Dem chances in such seats in general and so my betting approach is generally to be very cautious indeed about backing Lib Dem chances in such seats, and to consider strongly backing Labour challengers in these seats even where large swings are required. Having lost the casual affection of many left of centre voters, they have failed to gain in compensation the casual affection of right of centre voters. There is only so far that a personal vote can take you. If the Lib Dems are holding half of these seats this time next year, they will have done very well indeed from the current starting point.

Saturday, 26 April 2014

The hunt for 2010 Lib Dems. Part 4: conclusions

I have spent a long time looking at the 2010 Lib Dems because this group is going to play a central part in the next election. I am now going to summarise my conclusions, to put them all in one place.

1. The Lib Dems' decline has been so steep, uniform national swing is no longer a reliable tool. This is particularly apparent in Scotland (where the maths wouldn't work in more than half of all Scottish seats), but it is true in the rest of the UK also.

2. Such polling evidence as we have suggests that in seats that the Lib Dems will not be seriously contesting, they are doing better than uniform national swing would suggest. This is unsurprising, since there will always be a residual level of support and these seats had fewer Lib Dem voters in the first place. The proportion staying behind seems to be in proportion to the national opinion polls - though it should be stressed that we have very limited and not hugely reliable data.

3. The corollary of this is that at present the Lib Dems are doing worse in the seats they are seriously contesting than uniform national swing would suggest.

4. In the seats that the Lib Dems are not seriously contesting, their 2010 voters who have deserted them have so far decisively broken for Labour.  I have made some assumptions about the proportion of voters in each seat that has shifted to Labour, which gives a very different impression of the order of marginality of seats.

5. As a consequence, Labour have secured a substantial strategic shift in the marginals. It will be much easier for them to secure most seats or an overall majority than the results alone suggest (though the effect seems a bit more muted than some have suggested owing to the relative outperformance by the Lib Dems in these seats).  This accentuates the already substantial advantages that Labour has I the electoral system.

6. The Lib Dems,on the other hand, can take no particular comfort from their past outperformance of uniform national swing. The scale of the drop in support will make this much tougher this time around.

That's the position now. What might change these voters' minds over the next year?

I suggest the answer is different in different types of constituency. Those 2010 Lib Dems who have taken a leftward turn are normally going to stick with their current choice. They've decided what they think of the current government and it's not positive (to put it mildly). In Labour/Conservative marginals, that means Labour. In Labour/Lib Dem marginals, that means Labour.

There are two interesting cases. In Scotland, disaffected left-leaning Lib Dems have the additional option of the SNP. Many of them will take it. Some of these voters will rule out the other left of centre option, while many will be open to both choices. In seats where both Labour and the SNP are viable options, there remains the possibility of swing between these parties.

This could aid the Lib Dems in constituencies where there is confusion about who is the best-placed left of centre challenger. In constituencies where it is clear who out of Labour and the SNP is best placed to win, these votes will mostly end up with that challenger.

Similar thought processes are presumably also at play in Wales with Plaid Cymru and in those few English seats where the Greens are credible challengers. 

What of those seats where there is no other left of centre challenger capable of winning? The Lib Dems have traded heavily on tactical votes from the left in the past. But can they repeat the feat after making five years of the perceived betrayals and compromises of coalition government with the Conservatives? 

Rationality would suggest yes. This is the sensible, if unappetising choice, for voters who see the Lib Dems as better than the Conservatives, even if the difference is not as great as they previously understood. 

But voters are not entirely or even mainly rational. Some will make the rational choice while some will want to register their unhappiness by voting with their heart or by simply not voting at all. The unanswered question is how angry these voters remain. 

It is reasonable for the Lib Dems to hope to woo these voters back. In the next year, they need to find the positioning and message to reach those of these voters who are unreasonable. They have it all still to do. 

Friday, 25 April 2014

The hunt for 2010 Lib Dems. Part 3: Conservative/Labour marginals

In 2010, Britain was swept by Cleggmania.  Off the back of a highly successful air campaign, the Lib Dems attracted new voters up and down the country.  Unfortunately for them, they mainly did so in seats where they had no prospect of success and they actually lost seats to the Conservatives, falling back from 62 seats to 57 seats.

So the Lib Dems had a substantial vote share in the Conservative/Labour marginals where the next election will be decided.  Much attention has been placed on this group.  The Lib Dems have declined sharply in the polls and these votes will play a big part in the outcome of the next election.

First things first, how many of them are available to be snaffled by rivals?  As I noted yesterday, Lord Ashcroft found that they had declined in the 32 most marginal Conservative/Labour seats by just over half, matching the decline in the national opinion polls.  In seats where the Lib Dems are not seriously campaigning, it seems that they may lose votes in proportion to their national loss of vote share.  This isn't going to be precisely accurate, but uniform national swing clearly doesn't work when we're looking at such dramatic falls in support, so I propose using that.

In those same 32 most marginal Conservative/Labour seats, the Lib Dems averaged 17.5% of the vote.  So on average this puts 9 or 10% of the vote in play, regardless of other voter movements.  That's a big chunk.   In some constituencies, the numbers in play will be substantially bigger.

How these voters have divided between other parties is unclear - different polls have different findings.  Those that have expressed a preference are dividing 2 or 3 to 1 in favour of Labour over the Conservatives, with a non-trivial number going to UKIP.  Anything up to 40% of these voters still don't know how they are going to vote.

Putting these movements into a typical seat, we find that of our 17.5% Lib Dem vote share, Lib Dem don't knows comprise 7% of the electorate, Lib Dems comprise 5%, Labour 3 or 4% and the Conservatives 1% or so.  So there has been a swing to Labour on these voters alone of 1 to 1.5% of all voters (probably closer to 2 to 2.5% of decided voters).  And that's without taking account of any other improvements in Labour support or declines in Conservative support from the last election.  This is a big bonus.

Because of this effect, much attention has been paid to the Lib Dem to Labour switchers.  Are they likely to return to the fold?  In short, no.  All the polling (and anecdotal evidence) suggests that this group comprises the most zealous supporters of Ed Miliband, matched only by an intense dislike of the Conservatives in particular.  I shall look separately at vulnerabilities in Labour's vote, but there's no vulnerability here.  These voters are going to be trooping out next May to vote Labour and will be in danger of pushing the pencil through the ballot paper in their vehemence when they mark the cross.  If these voters do not vote Labour, it will to inflict maximum damage on the Conservatives. A decline in the number of Lib Dem to Labour switchers would not be good news for the Tories: it would mean that Lib Dem incumbents were likely to do better.

There's a big betting point here.  The bigger the Lib Dem tally at the last election, the more of an in-built swing we can expect to see to Labour in Conservative/Labour marginals.  I have prepared the following table of Conservative defence seats where Labour are in contention.  I have added a column where I have notionally allocated a net 25% of the Lib Dem vote tally to Labour, and ordered the seats in order of the notional majority on that basis.  This equates to half the Lib Dem voters switching, and doing so in the ratio 3:1 Labour: Conservative.  I'm open to debate on the exact figures, but this seems a fair starting point to me:

Where I have put "MAJORITY" in block capitals, Labour would have taken the seat.  So Labour would have won a further 26 seats on this basis.  That would have been enough to have made them largest party at the last election.

It will be apparent from this table that some seats are seriously mispriced.  It repays very careful study.  On this basis, I would rather back the Conservatives to keep Enfield North where Labour need only a 1.91% swing but only 12.2% of the electorate voted Lib Dem in 2010 than backing them to keep Northampton North where Labour need a 2.41% swing but 27.9% of the electorate voted Lib Dem in 2010.  Yet the Conservatives are 11/4 in Enfield North and 5/4 in Northampton North.  These odds are in the wrong order.  There are many similar examples.

And of course, this is just from the starting point of the last election.  Labour will hope to gain other supporters and to see the Conservatives fall back.  However, I suggest this table makes a more reliable guide to the starting point for the next election than looking at the unadjusted results last time around.

In my next post, I shall draw the threads together and try to establish some principles for handling the sharp drop in the Lib Dem vote when considering these markets.

Thursday, 24 April 2014

The hunt for 2010 Lib Dems. Part 2: the Lib Dem held seats

Yesterday I looked at the Lib Dem position in Scotland and found it was truly dire. But more important, I found that it did not fit standard electoral modelling methods. Now to look at what might happen in England.

The Lib Dems are not doing as badly in England and Wales as they are in Scotland, but the picture still isn't a happy one. In Scotland they've lost two thirds of their support since the last election, if the polls are to be believed. In Great Britain as a whole, they've lost half their support, maybe a bit more. 

The same technical question arises: is uniform national swing a useful tool? And while the position is not quite as stark as in Scotland, for me the answer remains no, basically for the same reasons: the drop in support at these levels is too steep and it takes insufficient account of the local campaigns that the Lib Dems will fight. 

If the Lib Dems tally half the votes they got in 2010, then the maths of uniform national swing will not work in any constituency where they got under 11.5% of the vote. And it also implies that in seats where the Lib Dems tallied only slightly more than that in 2010 they will be reduced to trace elements. 

I just don't buy this. In almost all constituencies there will be a residual number who will vote Lib Dem come what may, regardless of how little campaigning the Lib Dems put into the area. Bear in mind that in 2010 the Lib Dem vote was very evenly spread. In the 125 Labour target seats held by the Conservatives, the Lib Dems' average vote share was over 10% in every constituency and over 15% in 94 of these:

Some of these Lib Dem voters will stay loyal. 

As I noted yesterday, there is no easy mathematical model for dealing with this to identify a seat count for the Lib Dems. My suggested approach is to assume a national level of polling, assume a base level of support in seats where the Lib Dems will not in practice campaign and see what that leaves you with in the rest. 

What might the base level of support be in untargeted seats?  This is unclear, but we can make some guesses from Lord Ashcroft's marginals polling:

Last year he found 8% support for the Lib Dems in the Conservative-held Labour targets, and I have no particular reason to disbelieve this figure. Lord Ashcroft tells us that the constituencies surveyed were the 40 most marginal, and the average Lib Dem vote share in the 32 of these seats that Labour are targeting in 2010 was 17.5%.  The national comparison opinion poll had the Lib Dems on 11%, so the Lib Dem support in the national poll was 47% of the 2010 performance and that in the marginals poll was 45% of the 2010 performance.  This looks much closer to proportional swing than uniform national swing in these constituencies.

If this is correct, this is bad news for the Lib Dems because it implies that they are currently doing worse than uniform national swing in the seats that they are seriously contesting. (It also implies that the value bet on Lib Dem lost deposits remains at the bottom end of the spectrum.) Of course, current polling does not take into account how the Lib Dems will in fact perform on the ground in the campaign. But the Lib Dems will have it all to do if they are to retain the seats they would hope to keep on a uniform national swing basis.

Let's have another look at the list of Lib Dem seats arranged by swing:

Based on the last ICM poll, on a uniform national swing basis the Lib Dems have seen a 3% swing to the Conservatives and a 9.5% swing to Labour. If they are going to struggle to meet that, we should look with disfavour on their chances of holding on at the margins of this swing. particularly in those seats where Labour are the challengers, because the defectors from the Lib Dems to Labour have no tactical reason to return.  Perhaps Simon Hughes won't get by on a recount after all.  Labour don't look like a 9/4 shot in Southwark & Old Bermondsey, anyway - they must be at worst an evens shot. I'm on.

But how will 2010 Lib Dem voters react when the Lib Dems come calling for their vote? We know that substantial numbers have deserted them, principally for Labour but in some numbers for both the Conservatives and UKIP.  Different polls show different proportions, but Labour are poaching of the order of 30% of the 2010 Lib Dem voters.  The Conservatives have also skimmed off maybe 10 to 15% of these voters.

I shall look next at how likely it is at a national level that these voters will return to the Lib Dems (spoiler: not very). For now, I want to look at how vulnerable the Lib Dems are in their own seats to losing previous Labour-inclined tactical voters.

Here's a table I prepared just after the last election of the Lib Dem held seats with Conservative challengers:

The first thing to note here is just how low the Labour vote shares are. This is quite out of line with the Lib Dem vote share when they are in third (as shown by the earlier table) and the Conservative vote shares when they are in third are much like the Lib Dems'. This strongly indicates past tactical voting by Labour supporters in the Lib Dems' favour.

The next thing to note is that Labour have been squeezed in varying degrees of severity in different seats. That must be a hint as to how vulnerable the different Lib Dem incumbents are to untactical voting.  Unless incumbents have convincingly pitched themselves as internal opponents of the coalition, many Lib Dem MPs will have alienated this part of their support.

The Lib Dems will continue to pitch for tactical votes from Labour supporters of course.  And they will continue to have some success.  But it seems unlikely that they will gain tactical votes to the same extent as before.  Labour supporters in such constituencies have an invidious choice between casting a tactical vote in favour of Judases or casting a useless vote in favour of their preferred party.  Many will prefer the latter.

I would expect the average Labour vote share in these constituencies to look much more like the typical vote share that the Lib Dems and the Conservatives managed at the last election in constituencies where they were third placed - getting up towards the 15% mark or so.  This has serious implications for those who have to date been highly successful at gathering those tactical votes.  Unless the Lib Dems pick up a bit in the polls, they might easily find themselves losing up to half their seats.

Jeremy Browne in Taunton Deane has managed the unusual feat for a Lib Dem of appearing to oppose the coalition from the right.  Given his slender majority and that he has apparently benefited from substantial Labour tactical voting in the past, this seems adventurous.  Take the evens on the Conservatives.  Cheltenham might be a worthwhile bet on the Conservatives at 7/4.  Conversely, the Lib Dems may be harder for the Conservatives to crack in Berwick-upon-Tweed than they look on swing alone.

I have danced around the subject, but next I will look at the main action: what will the 2010 Lib Dem voters do in Conservative/Labour marginals?  This is very important for the outcome of the next election.

Wednesday, 23 April 2014

The hunt for 2010 Lib Dems. Part 1: Scotland

One of the central questions for the next general election is how those voters who voted Lib Dem in 2010 will vote next time round, and how this will differ between different categories of voter.  This is a big subject, too big to be dealt with in a single post.  So I propose dividing it as follows:

1) Scotland
2) Lib Dem held seats
3) Conservative/Labour marginals

Let's start with some basics.  The Lib Dems have undeniably lost support since 2010.  At the last general election, they polled 23% of the national vote share.  In the most recent ICM poll (by no means the least favourable to them), they polled 12%.  They have lost about half their voters in four years, if the polls are to be believed, and possibly more depending on which pollster you choose to believe.

The scale of this drop raises some important technical questions.  Normally when we assess the likely impact on seat numbers, we look at uniform national swing.  Past experience has shown that usually is the best rough and ready guide.  Is it a reliable tool here?  In short, my answer is no, but it's probably still the best starting point in England at least, and if we choose to deviate from it we should need some kind of reason why.  I'm very doubtful whether it's any use at all in Scotland.

I want to start looking at the Lib Dem seats in Scotland because they give an extreme illustration of some of the problems we face when trying to work out what is happening in the electorally more important English marginals.

Why is Scotland different?  Simply, the Lib Dems have performed very differently electorally in Scotland from south of the border and are now apparently suffering even greater woes in the polls.

In 2010, the Lib Dems took 11 seats in Scotland.  Here they are:

They did this on 18.9% of the vote, so they got exactly their proportionate share of the seats.  If they had done equally well across the UK in converting votes into seats, the Lib Dems would have had 150 MPs rather than the 57 they actually tallied.  So the Scottish Lib Dems far outperformed their southern colleagues on a lower share of the vote.

It should also be noted that, unlike the rest of Britain, the Scottish Lib Dems saw their vote share decline by 3.7% (a sixth of their 2005 votes), but kept all the seats that they had won in 2005.  So they already have practice in husbanding their seats against a background of electoral decline.

If the polls are to be believed, however, that was just the hors d'oeuvres for what is coming next year.  Few Scottish polls have been taken on Westminster voting intention - for some reason the pollsters are more interested in the independence referendum.  So far as I can see, only Survation have conducted surveys on the subject this year.  In their two most recent polls this month, the Lib Dems registered 6% both times.  Depressingly for the Lib Dems, this is not that particularly out of line with the polling of the last three years:

If we look at Lib Dem prospects on the basis of uniform national swing, then using that last Survation opinion poll there has been a 14.5% swing from the Lib Dems to the SNP, a 3.5% swing from the Lib Dems to Labour and a swing of just over 3% from the Lib Dems to the Conservatives.  Uniform national swing would leave the Lib Dems with four seats: Berwickshire Roxburgh & Selkirk, North East Fife, Ross Skye & Lochaber and Orkney & Shetland (some of the seats would be lost to the SNP from third or fourth place).

The problem with this approach is that the maths simply doesn't work for uniform national swing in seats where the Lib Dems polled 13% or less in 2010 if the Lib Dems indeed only secure a national vote share of 6%. In 2010 the Lib Dems polled 13% or less in 33 out of the 59 Scottish constituencies. That is far too many to overlook as a quirk. The decline in vote share is simply too steep for uniform national swing, which is a model designed for smaller swings.  Even if the maths works, there's the question of plausibility: is it credible that the Lib Dems will poll as low as 1 or 2% in numerous constituencies? I can't see it. So I would not regard uniform national swing as very helpful unless the Lib Dem vote share in Scotland looks set to exceed 10% by a bit. Right now, that really doesn't look very likely.

What other methods are there? We could work on the basis of proportionate swing - votes are gained or lost in each seat in proportion to the national vote share. But this takes no account of the concept of regional variation, the popularity of a local MP or differential campaigning effort, the last two of which in particular are relevant to the Lib Dems. So I'm not attracted to that.

Alternatively, can the Lib Dems repeat the trick they pulled off in 2010 of defying uniform national swing completely?  Not at 6% they can't.  In the 11 seats that they won, they tallied 178,534 votes.  On the same turnout, the Lib Dems would tally only 148,000 votes on a 6% vote share across the whole of Scotland.  Since negative votes have not yet been put into operation in general elections, something has to give.  If the Lib Dems poll 6% across Scotland, a lot will give.

We need to start from first principles.  Neither uniform national swing nor proportionate swing are helpful in Scotland because the Lib Dems will in effect be fighting a handful of seats and in practice ignoring the rest.  But equally, the Lib Dems will get some votes even in the seats that they ignore.  They polled nearly 300,000 votes in 2010 in seats which they did not win.  Some of them will remain loyal.

The current opinion polling is likely to understate local constituency effects because most voters don't think about them between elections.  Some will, of course.

I have no magic answer to this problem. I suggest tackling it in reverse.  If the Lib Dems are indeed to tally 6% at a general election, we need to assign a given average vote in seats that the Lib Dems will not seriously contest.  I am uncomfortable going below 3% for this, given their past significant voting strength right across Scotland.  That implies that the Lib Dems would get around 45,000 votes in the 48 Scottish seats that the Lib Dems do not currently hold.

At 6% across Scotland overall, that leaves around 100,000 votes to be shared among the remaining 11 constituencies.  Charlie Kennedy and Alistair Carmichael would presumably account for around 25,000 of these votes.  Which leaves nowhere near enough votes to go round for the other MPs.  The Lib Dems would be lucky if they salvaged any of the rest.

Now you might say that the Lib Dems would in practice get more than 6% of the vote nationally because of the power of incumbency of some of these MPs.  You might well be right - in fact, I expect that you are.  But you can't decide to fix the maths to suit your desires.  Bear in mind I've set the floor for votes in seats that the Lib Dems will not seriously contest at a very low level.  Without any incumbents, the Lib Dems on this basis would be heading for an asterisk in the polling tables.  If you are going to assume that the Lib Dems would do better than 6%, you need to assign a percentage, make some assumptions as outlined above and then consider the consequences constituency by constituency.

Where left of centre former Lib Dem voters have a clear alternative to the Lib Dems, it seems reasonable to suppose that many of them will be inclined to take it.  Given the strength of feeling on the left, many of them may take such an option even if they don't have a clear alternative.  That leaves current MPs relying heavily on personal votes.  With Malcolm Bruce and Menzies Campbell stepping down, the Lib Dems will lose even that in two constituencies.

Where might the Lib Dems hang on?  They're best placed when they're most closely challenged by Conservatives without a plausible alternative on the left.  Enough left of centre voters might hold their noses and choose the lesser of two evils to get them home again.  For this reason, the 4/6 on the Lib Dems keeping Berwickshire Roxburgh & Selkirk looks an OK bet, even with everything else going against the Lib Dems.  But it's not compelling.

Far better is the 1/2 on Labour taking East Dunbartonshire.  These odds seem bafflingly long to me, no matter how good Jo Swinson might be as an MP.

Tuesday, 22 April 2014

And where might UKIP's support go to?

At present UKIP are polling upwards of 12% in the general election polls. Can they maintain or even improve on this? Perhaps. But there is a widespread assumption that they will not do this well next May. This gives rise to an important question: who are the voters who might change allegiance from UKIP and where are they based?

Historically UKIP supporters have been seen as Tories on holiday, with their public image largely set by the stripes blazers of the Maaastricht rebels in the 1990s. But as I showed yesterday, this image is out of date. UKIP has more working class supporters than middle class supporters. More than half its current support did not vote for the Conservatives in 2010 (though 45% did). So far as we can tell, their supporters are disproportionately based in rural eastern England, the Thames estuary and along the south coast east of Bournemouth, but have significant strength through large parts of England.

That is now. But what might change in the future? What follows is speculation. But it is speculation based on what I hope is logic.

1. UKIP supporters in safe seats will be put under no significant pressure to change their votes. So relatively few will (unless they would have done anyway).

This is a mixed blessing for the Conservatives in particular. They can count on any Tories on holiday to return in their safe seats, but that won't make any difference to the outcome.

2. In marginals where UKIP are not perceived to be in contention,their supporters will be put under a lot of pressure to choose between one of the major parties. Most UKIP voters who have a decided preference between the parties on offer will be tempted to switch.

So UKIP should fade most in marginals where they are out of contention. But this effect may be limited. In 2010, the Lib Dems' support rose in most seats where they finished third (in a general election where their overall support went up only marginally); their voters proved resistant to squeeze. Many Kippers may still prefer to register their antipathy to the major parties.

There is some polling evidence to support the resilience of the Kipper vote:

It is unclear how these views would evolve in an election campaign and this polling was not specific to marginals.  Each gambler will need to form his or her own view on this.

On the one hand, the largest part of these voters were 2010 Conservatives, so that's bad for them. On the other hand, these disaffected voters aren't actively supporting Labour, so the damage done to the Conservatives is half what these voters could have inflicted in Labour/Conservative marginals.

3. Most Kippers who do switch between now and the election will probably do so to the Conservatives.

The question to ask, regardless of the UKIP voter's previous voting record, is why he or she is not supporting Labour now, while they are in opposition and not making the compromises of government. If they aren't tempted by Labour now, why should they be tempted in a year's time?  These are voters who have chosen to protest at the direction that the government has taken but have decided so far not to do that through the main opposition party. Given that to date UKIP has picked up relatively few 2010 Labour voters, few have a track record of voting Labour either. Whatever their views about the Conservatives and the Lib Dems, they seem to be turned off Labour.

Where the Lib Dems are in contention, they can hope to pick up some of those of their 2010 voters who are currently choosing to protest via UKIP. But the fact that UKIP supporters voted disproportionately Tory in 2010 suggests that they are much more likely to be receptive to a Conservative message than a Lib Dem pitch, and the 2010 Lib Dem voters who have simply transferred their party of protest will not easily be won back after five years of coalition.

It should also be noted that UKIP is particularly strong among readers of the Express, the Mail and the Telegraph:

UKIP poll above 20% with the readership of all three (any visit to the websites of the Telegraph or the Mail will leave you marvelling that the percentages are so low). Even in today's declining newspaper market, that's a large audience. The Conservatives will hope to be able to use those newspapers' traditional loyalties to reach these voters. It's a reasonable hope - more reasonable than hoping that the Mail might change many Labour voters' hearts and minds.

4. There must be serious doubts whether those voters who didn't vote last time will vote next time.

11% of UKIP's supporters did not vote in 2010, according to YouGov. I'm doubtful how many of those will in practice turn out next year.

5. In constituencies where UKIP are perceived to be in serious contention, their voters will not easily be squeezed.

UKIP will seek, in homage to the Lib Dems, to maximise this using dodgy bar charts constructed  from the local and EU elections and from the constituency opinion polls that Survation has carried out for the party. We can work out well in advance which constituencies that tactic is likely to work best in.

In those constituencies where Labour is in contention Labour is likely to do disproportionately well because of the differential way in which UKIP has drawn its support from the Conservatives and Labour. The Conservatives need to see UKIP decline substantially in the national opinion polls if they are to hope to counteract this effect.

This, however, is a small set of constituencies and this effect seems already to have been fully priced in. 


The rise of UKIP has been bad for the Conservatives, harming them disproportionately. That effect, however, is fully shown in their current poll rating. UKIP seem, so far as we can tell, to be performing particularly well in the east of England, the south coast and the Thames estuary. With the exception of the last of these, these contain few of the marginals that will decide the next election.

UKIP's vote is likely to be squeezed in Labour/Conservative marginals and where it is the Conservatives are likely to benefit (or rather, recoup some of the disadvantage that UKIP's rise has caused them).  It is unclear how successfully UKIP's vote will be squeezed and each gambler will have to form his or her own judgement on this. My view is that the squeeze will be most effective where UKIP is structurally weakest, so betting on Conservatives in northern marginals may offer better value than betting on the Conservatives in the Thames estuary seats (thus accentuating the pre-existing regional differentiation still further). We might reasonably expect to see the Conservatives pick up two or three points in the opinion polls before the next election off the back of any inroads into the UKIP vote lost.  I am doubtful that the Conservatives will squeeze UKIP much further than that.

Overall, it is my judgement that while UKIP's increased presence since the last election will be of some advantage to Labour, the effect is likely to be quite muted, at least in seat numbers.  

Monday, 21 April 2014

Where is UKIP support coming from?

As I said yesterday, I'm now pursuing a slightly different tack. We could proceed on the basis that support had moved evenly between the parties in all of the different constituencies by uniform national swing. That is certainly a handy tool and in the absence of any evidence that it's wrong, it should always be the fallback position.

But it is apparent that in some cases at least support has not moved uniformly. And there we should try to do better - to identify where it has moved differently and how that might affect different constituencies. It will help us decide whether the bookies have got their odds right.

The single most dramatic development of the last few years has been the rise of UKIP. In 2010 UKIP polled 3.1% of votes and came third in only four constituencies. It is now routinely polling in double figures and in some polls well into the teens. While as I noted on Saturday UKIP is odds against to win any seats and its chances of winning significant numbers of seats are slim, if it gets voters in the numbers currently being polled, its voters could swing the outcome of many constituencies. So where have UKIP's voters come from? And will they stick around for the general election?

UKIP's voters have been little-studied, and the first major academic study is set out in "The Revolt On The Right":

From analysis of opinion polls, the authors conclude: “Contrary to those who argue that Ukip’s voters are middle class Tories, we actually find that its base is more working class than any of the main parties”.

Blue-collar Ukip voters outnumber their white-collar counterparts by 42 to 30 per cent. In contrast, 44 per cent of Conservative supporters are middle class; while for Labour, the middle classes have a narrow 36-35 per cent lead.  57 per cent of Ukip supporters are over the age of 54, while just over one in 10 are under 35.  55 per cent left school at 16 or earlier, while only 24 per cent went to university.

However, Ukip suffers from “a persistent failure to appeal” to three key groups of voters – women (put off by the “chauvinistic and anti-feminist views” of Ukip members and politicians); young people (who find the party “almost comically out of touch” with their own worldview) and the ethnic minorities (because of its “strident and often emotive language about immigration.”)

 If you want to go into more detail, I can strongly recommend the excellent election data site:

We know something of UKIP voters' political background, thanks to YouGov:

And it's clear that the Conservatives are hit hardest by this.  Fully 45% of UKIP supporters were 2010 Conservative voters - a far greater percentage than any other single segment of UKIP support.  That's over 5% of the electorate as a whole.  So in constituencies where UKIP is likely to be disproportionately strong, the Conservatives are likely to be disproportionately affected.  But it shouldn't be forgotten that in seats where UKIP is likely to perform poorly, the Conservatives can expect to do better than otherwise expected.

It seems that UKIP's rise has favoured Labour. It has taken a disproportionate number of Conservative voters, resulting in Labour being relatively better off by about 4% than it otherwise would be. Where we have no particular evidence to the contrary, we should assume for now that relative swing will be uniform.

This uniform swing is of course already factored into the polls, so we do not need to make an additional adjustment. What we need to do is determine which seats may see a disproportionate swing to UKIP and adjust our betting strategy accordingly by tending to favour Labour and write down the Tory chances. Oh, and to identify which seats may be resistant to UKIP's charms, where we should do the reverse.

It seems from local election results that UKIP's new support is geographically concentrated.  This image from after the last local elections tells the story (click on the map to see it in detail):

It seems a safe bet that the deeper the colour, the more impact UKIP is going to have on the outcome, even if it doesn't win.  Note also that large areas have proved resistant to the purple tide so far (though this year will be the first test of their newfound popularity in metropolitan areas).  These include seats with UKIP at relatively short prices to win, like North Devon and South Suffolk.  And, obviously, UKIP is basically an English affair.  I suppose EIP would be harder to pronounce.

Let's look at this map a while.  Even bearing in mind that the metropolitan areas are not on the same footing, UKIP still look as if they are a rural south eastern party.  UKIP seem to be particularly strong in East Anglia, and particularly the fenlands.  But even here they have no county council seats in Norwich, Ipswich, Cambridge or Peterborough.  There are other areas of strength - notably Kent and Sussex, but with the exception of south Essex, UKIP councillors are mainly elected in rural areas.

Now obviously council seats are awarded on a winner-takes-all basis, and healthy vote shares don't always register in such tables.  We know that northern English urban areas have potential UKIP voters, because UKIP gathered respectable vote shares in the Rotherham, Middlesbrough, South Shields and Wythenshawe & Sale East by-elections (the contrast with Manchester Central and Cardiff South & Penarth is marked).  But in the absence of enough data to judge reliably, we should not make any allowance either way as to how UKIP will fare in these areas in the general election.  As and when we get more data, we can come back to that.  The coming local elections in May are going to be a very important data point on this critical question.

If UKIP's support is disproportionately in rural south east England, it's going to be disproportionately in Conservative safe seats.  That's a sideshow from the main Labour/Conservative fight (and arguably reflects Labour's inability to speak to those areas of the country).

Leaving aside other regional variations, we should tend to look more kindly on Conservative prospects in more northern marginals where UKIP have so far failed to impress and less kindly on Conservative prospects in marginals on the eastern coast and in the Thames estuary. All other things being equal, this consideration would make you think that Weaver Vale would be a more likely Tory hold than Waveney, and Rossendale & Darwen a more likely Tory hold than Cleethorpes.  To give an example, if UKIP's vote is 24% in Waveney and 8% in Weaver Vale, that would disproportionately benefit Labour relative to the Conservatives in Waveney to the tune of 5% if UKIP's voters' previous allegiances are comprised in the ratios identified by YouGov in both seats (and conversely give the Tories a better chance in Weaver Vale).

Of course all other things are not equal. There are other geographical considerations at play. And it is far from clear how much of its current support UKIP will hold onto for the next election. So next we need to look at where UKIP's current support might go to.

Sunday, 20 April 2014

End of part one

So far I have looked at the individual constituency betting odds from the following perspectives: the chances of SNP gains, Labour gains and Conservative gains, the chances of the Conservatives and the Lib Dems defending their seats, and comparing the odds in the general overall majority and most seats markets with the odds in the individual constituency markets.
I have done so on the assumption that in aggregate the constituency odds are our best estimate of what's going on, while noting individual anomalies that I have spotted along the way (more commonly known as betting opportunities). I have further assumed that the individual constituency odds - for Labour and the Conservatives at least - are perfectly correlated contingencies (which of course they are not) to establish a handy tool for judging the chances of success for each of the main parties overall. The general idea is that looking exclusively at marginality takes insufficient account of the different nature of the seats (who's second, is there a relevant third player, where is the seat). The constituency odds factor those matters in to the best judgement of bookies and punters. By arranging constituencies by order of odds rather than majority, we can see how many seats gamblers expect the parties to take.
So all my posts so far have been based on the implicit assumption that there is some particular insight to be gained from the constituency markets in aggregate, even if some individual constituency markets are wrongly priced.  Is this assumption safe?
In a word, no.  Any kind of market can be wrong if the participants are in aggregate mistaken.  Political betting markets, which are small markets by any conventional standard, have relatively little analysis put into them and many who bet on them are influenced by heart rather than head.  They are to be read with care.
For a summary of the risks in such markets, I can do no better than to refer back to a post I made in the wake of the last general election:
I draw attention to the following points in particular.
"There is no special wisdom in the market
Money talks and money should be respected, but money can still be hopelessly wrong. Most political gamblers don't do as much research and testing as professional financial analysts. So there is probably more scope for uncovering errors to profit by than in the financial markets. It's always good to start with the belief that others are on the right track - assuming that your opponents are stupid is always a recipe for losing. But look for where assumptions may have been made. Those assumptions might be incorrect or out of date."
"Think about what's driving prices: it might not be the underlying odds
It is easy to assume that betting prices are driven by underlying probabilities. Not so. What drives betting prices is the money that is placed. Betfair automatically works in this way and conventional bookies are going to want to keep their books balanced. The price ends up as the balance between two competing flows of money.
This is particularly important in something as emotional as politics. A lot of gamblers want to back their own horse. This means that prices can be quite seriously askew, particularly in constituency markets.
The particular danger arises where a party has considerable support but relatively few exciting betting opportunities. The odds on the Lib Dems and the SNP taking new seats were way too short in most cases. This was observable in advance (at least, I observed it). For every Redcar there was a St Albans, a Sheffield Central, a Newport East and a Bedford. The SNP didn't even have those meagre satisfactions.
The same effect could be seen in a localised way for the Tories. In Scotland, the Conservatives came nowhere close to justifying the odds in some seats. The challenges in Morley & Outwood and Exeter failed, though Antony Calvert did very impressively in Morley & Outwood. The emotional satisfaction of backing an upset against a particularly unpopular (with Tories) Labour minister drove prices out of kilter.
It is no surprise that many of the longer priced bets that came home were on Labour candidates. There's something very unheroic about backing a candidate to retain a seat. It's much more fun backing a candidate to win a seat. But fun doesn't equate to money."
How might this affect betting for next time?  I have already noted that prices on the SNP and UKIP in many constituencies seem way too short.  The swings that the SNP would require to make even a few gains are large.  UKIP have no track record and are not even that clear on their best target seats.  Yet both have avid punters willing to back them at unappetising odds.  It would be unwise to join either of these herds.  If you must bet on either of these parties, bet on longshots.  It is unlikely that either will benefit from a huge wave of popular support, but not impossible and some of the 50/1 shots are probably too long priced.
What about Labour, the Conservatives and the Lib Dems?  These are harder to judge.   We can see that the odds of the general markets are a bit more Tory-friendly than the odds of the constituency markets - but which of these is closer to the mark?
I need to do more analysis on their chances to judge properly how their odds stack up.  How are specific groups of voters going to move at the next election?  So my next few posts will look at the likely impact of UKIP supporters, voters in Lib Dem/Conservative marginals, 2010 Lib Dem voters in Labour/Conservative marginals and  Conservative voters in Lib Dem/Labour marginals.  I then ought to look at geographical considerations.  And then I shall take stock.

Saturday, 19 April 2014

UKIPalypse now? Where might UKIP win a seat?

With UKIP polling well at present, it is reasonable to ask whether they might win seats, and if so where. Because UKIP are such a new party, presenting their target seats by required swing is pointless. So let's go straight to the list of seats by odds:
Three things stand out. First, UKIP are serious contenders in very few seats. I have highlighted those where their odds are at 20/1 or less. There are 24 seats even with such an extended list of targets. Admittedly we have yet to get prices on some seats where UKIP will hope to prosper, but the shortness of this list should be sobering. I've included an extended list of longer shots up to 50/1 for diehard Kippers to plough through.
Secondly, this is a multicoloured list. Six out of seven of their top targets are Conservative-held, but after that we see more or less equal quantities of red and blue. They have almost as much potential to mess up Labour's plans as the Conservatives'.
Thirdly, if one leaves the longshots alone, UKIP have a very few seats where the punters think they have a decent chance. By way of comparison, UKIP are thought more likely to take Eastleigh (a seat they couldn't win in a by-election) than the Conservatives are thought likely to take Sherwood, a seat the Tories already hold.
And this should set alarm bells ringing. UKIP's prices must be too short in these seats if that's how the odds match up.  UKIP could conceivably win one of these short priced seats, but the value is generally lousy.  It seems that the bookies are catering to the enthusiasm of the purple punters by relieving them of their money at poor odds.
I'm going to be slow to back any UKIP shots, but what should we look for when deciding where they might be worth a flutter?  I'm looking for constituencies which have seen better days, where the constituents compete for jobs with immigrants and where two parties are both looking to win the seat (reducing the total number of votes that UKIP need to reach).  Of the relatively short priced seats, Great Yarmouth and Thurrock look most promising on these tests.   Once we know where Nigel Farage is going to stand, that seat price will shorten also - so if you have a firm view on that, consider backing UKIP in that constituency as a trading bet.
But why back UKIP at 5/1 in Great Yarmouth when you can back them at 50/1 in the adjoining constituency of Waveney?  UKIP kept its deposit in Waveney at the last election, showing that it is relatively fertile Kipper territory.  And it's highly marginal - the Conservatives hold the seat by 769 votes from Labour (making that retention of deposit by UKIP look still more impressive).  This price looks plain wrong to me.  I'm on.

Friday, 18 April 2014

The Conservatives' overall chances: putting together the pieces

It is important to understand that Labour's chances of making gains and the Conservatives' chances of defending their seats, while related, are not identical. We can infer Labour's chances of securing an overall majority or most seats with some confidence from the list of its targets. But we cannot simply look at the list of seats that the Conservatives are defending and do the same thing. It is obvious that a party that holds seats already does not just aspire to gain seats, they must also defend what they have.
In the case of Labour, that's a largely theoretical distinction, because only a handful of seats that they hold are more likely to be lost than seats that they covet are likely to be won. But for the Conservatives, the position is a lot more complicated. The complicating factor is the likely decline of the Lib Dems. The Conservatives are currently expected not just to lose seats to Labour, but also to gain seats from the Lib Dems.

Many who have looked at this market have looked only at the seats that the Conservatives might lose without considering the seats that the Conservatives might gain. This is a serious mistake. To set this right, here are the Conservative battleground seats, arranged by odds:

The big point to note here is that quite a few of the seats currently held by the Conservatives are rated as less probable Tory wins than seats that they are targeting. 37 Lib Dem or Labour seats come in this category, meaning that if the Conservatives achieve what we might call a uniform odds swing such that they retain all their existing seats up to and including Thurrock, they would have a comfortable overall majority. The magical 326th seat would be Yeovil, which they would win at odds of 10/3. This is more or less the same as the odds on Betfair on a Conservative overall majority, which can be backed at 4.2 (16/5). So there is no great advantage to backing the Conservatives in individual constituencies as opposed to backing the Conservatives on Betfair.
Similar observations can be made about the Conservative most seats market. It's unclear how many seats would in practice be required. The 305th seat would be Brighton Kemptown at 9/4. The 300th seat would be Bedford at 2/1. The 290th seat would be Elmet & Rothwell at 7/4. There's a bit of an advantage over the Betfair most seats prices (currently around 5/4), but it's not overwhelming.  That Betfair price equates to the 283rd seat by swing - which given the current seat division between Labour and the Conservatives would have been just enough to give them most seats.  Of course, 283 would not be enough this coming time on the implicit assumptions in this table, because the Lib Dems would have far fewer seats.
So by all means prefer to bet on the Conservatives in the constituency markets rather than the general "most seats" markets for better value, but don't kid yourself you're getting amazing value.  It's just a little better.  By the time you get to the levels required for an overall majority, there's not much additional value to be had at all.
Where is the crossover between Labour and the Conservatives?  For that we need to compare the Conservative battleground with Labour's targets:
If Labour take Halesowen & Rowley Regis (and every seat before that on their target list) and the Conservatives take Elmet & Rothwell (and every seat before that on their defence list), both will have exactly 290 seats.  So there's the pivot point, if the bookies' odds are right.

Thursday, 17 April 2014

In the blue corner: how do the Conservatives shape up to resist a Labour advance?

Yesterday I considered the prospects for Labour gains.  Today I want to look at the other side of the coin, how the Conservatives are placed to defend their current seat tally.  This is not quite the same question, because Labour have the prospects of gains from the Lib Dems, while the Conservatives are sitting on some tiny majorities over the Lib Dems also.  But the main action is blue on red.
Here are the Conservative seats ranked in order of increasing majority:

The first thing that leaps out is how well the Conservatives did in close battles.  They won 18 seats with a majority of less than 1,000 and 18 seats with a lead of 2% or less.  So even a 1% swing away from the Tories would be a real struggle for them.  And that's leaving aside the fact that voters seem to have moved between the parties in some pretty non-standard ways this time round, which may not be particularly to the advantage of the Conservatives.
Just a note on the colour coding.  The colour of the majority represents the nearest challenger: red for Labour, gold for Lib Dems and grey for the Wyre Forest hospital chap.  As you can see, six of the top 20 most marginal Conservative seats have a Lib Dem as the closest challenger last time around.

Note, the prices have moved on during the week, so the odds on the Conservatives in a few seats (eg Morecambe & Lunesdale) are not directly comparable.
Anyway, let's look at this in terms of bookies' odds:
And once again we see a transformation.  The bookies have more or less written off the Lib Dems' chances of making any gains against the Conservatives.  We see the Lib Dem seats slump to the bottom of the table.  The only gold on the first page of this table is Watford, and that's because Labour finished a close third.  And it's a similar story on the second page.  Both Colne Valley and Bristol North West have Labour in a not-too-distant third.  The first seat on this list in which the Lib Dems are second favourite is Camborne & Redruth, where the Conservatives have a micro-majority of 92.  Even there, the Conservatives are priced at a very short 8/15.
The opposite is the case for Labour.  The Conservatives are odds against in 44 seats on this list. If they lost all of those without making any other gains, their chances of staying in power would be remote.  They are quoted at odds of more than 2/1 in 25 seats that they currently hold.
The basic question is whether you think current polling is a reliable indicator of the result in May 2015.  It has to be the starting point for our thinking, even if we apply adjustments for how polling might move.  But where might we expect to see the Conservatives do disproportionately well?
In the red/blue contests where I am considering a bet on the Tories, I look for constituencies with a fairly low Lib Dem tally and where the Conservatives and Labour have a relatively high share of the vote.  Such constituencies can be expected to have fewer protest voters of the type that might be attracted to UKIP and fewer Lib Dem voters who are likely to be backing Labour.  On that basis, Morecambe & Lunesdale looks fair value even at 3/1 despite its tiny majority (compare and contrast Lancaster & Fleetwood), as does Stockton South at 7/2.  Both seats have first time incumbents (historically first time incumbents of any party tend to outperform their party at elections, having built up some personal loyalty), and indeed this is a consideration that should help the Conservatives in many of their marginals this time around.  Wolverhampton South West at 3/1 is worth considering for the same reason.  Cannock Chase, despite the best efforts of its MP, is worth nabbing at 11/5 with Paddy Power.  Ladbrokes rate this an evens shot and I think Ladbrokes is much closer to the mark.  This seat seems to be trending Conservative.
Generally I'm much less attracted by the shorter priced Conservative seats.  It's unclear how the Lib Dem vote in particular is going to distribute itself.  The short prices have hidden risks that are not fully priced in, in my view. I have, however, taken the 1/2 on the Conservatives in Tamworth.  The Lib Dem vote share there is relatively low.
Two seats, Thurrock and South Thanet, have a substantial intervention from UKIP.  I shall look at the possible impact of UKIP separately.  But I'll note now that betting on the Conservatives in both these seats at present prices needs to be carefully weighed.
What haven't I talked about in this thread?  I have studiously avoided talking about looking at the seat market for a proxy for a Conservative overall majority or Conservative most seats bet.  We need a separate table for that.  I'll look at that tomorrow.