Thursday 19 February 2015

Purples reined? The current polling and the betting implications for UKIP in 2015

By their usually noisy standards, UKIP have been quiet since the turn of the New Year.  Labour and the Conservatives have been knocking chunks out of each other on the NHS and tax avoidance, and their paroxysms over immigration have been forgotten for now.  What does that mean for UKIP's chances?
 
The polling
 
Is this period of subdued activity having any effect on UKIP?
 
We need to take this pollster by pollster, so that we don't get confused by differing methodologies.  But on this occasion, I'm afraid you're not going to get a clear answer.

Populus - since mid-November, UKIP have tallied between 12% and 16%.  Populus changed their methodology at the beginning of the month to prompt for UKIP.  Since then, UKIP have not polled below 14% with Populus, but we should be wary about drawing too many conclusions on a relatively small sample of five polls.

ICM - there is little pattern to UKIP's polling with ICM.  In the last year they have polled between 9% and 16%.  In the two polls since the turn of the year, UKIP have tallied 9% and 11%, but a sample of two is small.

YouGov - UKIP have tallied between 13% and 18% in 2015 so far with YouGov.  In the ten polls so far in February they have stayed in a tighter range of 13 to 16%. 
 
Survation - Survation are every kipper's favourite polling company.  Since the Euro-elections last year, UKIP have dropped below 20% with Survation just once and have hit 25% on one occasion.  The only poll so far this year from Survation has UKIP on a very healthy 23%.
 
Ipsos-MORI - Kippers are much less keen on Ipsos-MORI.  Since the Euro-elections, UKIP have polled over 15% with Ipsos-MORI just once and so far this year have polled 11% and 9%.
ComRes - ComRes show a more consistent picture than many pollsters, with UKIP registering between 15% and 19% in every poll since the Euro-elections.  In the two polls this year, UKIP have tallied 17% and 18%.
Lord Ashcroft - Lord Ashcroft has found UKIP in a 14% to 19% range ever since last May.  In 2015, this range has been 14% to 17%.
Opinium - Since the Euro-elections, UKIP have polled in a range between 15% and 21% with Opinium.  So far this year, UKIP have polled 20%, 18% and 15%.
TNS-BRMB - Aside from one poll where UKIP registered 23% in June, UKIP have tallied between 16% and 19% with TNS-BRMB since the Euro-elections.  The three TNS-BRMB polls this year have found UKIP polling 16% and, twice, 18%.

So in 2015, depending on which pollster you believe, UKIP could be as low as 9% or as high as 23%.  They could be seeing their support firm up if you believe Populus, or seeing it trend down if you believe ICM or Ipsos-MORI.

It is tempting to use an average or to try to proceed by consensus.  On this occasion, I believe that temptation should be firmly resisted.  We should not attempt to reconcile that which has been produced by methods which are incapable of reconciliation.  Ultimately, one approach is going to be right and others are going to be wrong.  We have four broad camps:

1) UKIP bears - ICM and Ipsos-MORI
2) UKIP glass half empty - Populus, YouGov, Lord Ashcroft
3) UKIP glass half full - ComRes, Opinium, TNS-BRMB
4) UKIP bulls - Survation

It is possible for all of these camps to be wrong and it is possible on the boundaries between these camps that two camps can claim to be right (camps 2 and 3 are not very far apart, for example).  But unless we see polling convergence in the last three months, at least two of these camps are going to be shown to have got UKIP wrong well beyond the margin of error.  Reputations are on the line.

In stark contrast to the Lib Dems, the pollsters have no consensus as to how to measure how UKIP are currently doing and whether UKIP are maintaining the progress that they made in the second half of last year. 

This is not the place to go into detailed analysis of the difference between different methodologies and their virtues (not least because the detail of that is well outside my knowledge).  I note only that good and reputable polling companies take very different views of how UKIP are doing right now.  If you are wise, you will keep your mind open to all possibilities.  We are in uncharted territory.

When we don't know where we are right now, it's much harder trying to work out where we might ultimately end up in May.  Even if you decide whether UKIP are going to hold steady, rise or fall, that information is of limited use if you don't know your starting point.

Constituency polling

There has been a lot of constituency polling.  Surprisingly little of it has focussed on seats of particular interest to UKIP.  Survation, at the behest of Alan Bown, have investigated the following seats: South Thanet, Eastleigh, Great Grimsby, Dudley North, Folkestone & Hythe, Bognor Regis & Littlehampton, Great Yarmouth, Rotherham, North Thanet and Boston & Skegness.  None of the Survation polls are particularly recent - the most recent ones are from last September.

Lord Ashcroft has surveyed the following seats (some of them twice) where UKIP were in contention:  Camborne & Redruth, Cannock Chase, Dudley North, Eastleigh, Great Grimsby, Great Yarmouth, St Austell & Newquay, South Thanet, Thurrock, Walsall North and Wyre Forest (and others where UKIP had hopes but which have not yet shown up particularly in the polling, such as Portsmouth South).  And today he unveiled results from South Basildon and East Thurrock, Boston & Skegness, Castle Point and Cambridgeshire North East.
 
UKIP have done well in many of these polls, but have a clear lead in only one: Boston & Skegness as conducted by Survation, where they registered a 20 point lead.  This contrasts with Lord Ashcroft's poll in the same constituency released today, which shows a small lead for the Conservatives.  We can't assume that there has been any kind of swing since September in this seat.  It may simply be differences in polling techniques. 
 
I have previously noted some peculiarities with the detail of the Survation poll in Boston & Skegness:

http://newstonoone.blogspot.co.uk/2014/10/surfing-purple-wave-ukip-general.html

In any case should not rely too much on any one poll in isolation and try wherever possible to look at polls in aggregate to guide us. 

Both Survation and Lord Ashcroft also conducted polls in the four by-elections since the Euro-elections in May, which enables us to perform a rough sense-check on the likely accuracy of these polls.  I set Heywood & Middleton to one side, since it appears fairly clear that there was a late sharp swing to UKIP - as both Survation and Lord Ashcroft are at pains to point out regularly, opinion polls are snapshots not predictions.

Survation overstated UKIP in their final poll in all three of the other by-elections - by 1% in Newark, by 4% in Clacton and by 6% in Rochester & Strood.  Lord Ashcroft overstated UKIP in his final poll in Newark by 1%, understated UKIP by 4% in Clacton and overstated UKIP by 2% in Rochester & Strood.  On this very limited sample it seems that in by-elections both do fairly well in estimating UKIP's vote share in by-elections, but that Survation's house style may consistently tend to overestimate UKIP by a little bit.  (Both Lord Ashcroft and Survation seemed to underestimate the Conservative vote share in these by-elections by a bit too.  This meant that the margin of victory was greater for the Conservatives than anticipated in Newark and lower for UKIP than anticipated in Rochester & Strood.  But in all three cases they both predicted the right result.)

It's very dangerous drawing too many inferences from a sample of three, and we don't know how this  effect will read across to general election results or whether we have similar house effects to watch out for. Nor do we know how reliably the national polls relate to the by-election polls.  But I very tentatively draw the conclusions that Lord Ashcroft's constituency polls may well be there or thereabouts for the UKIP share of votes at least, while the Survation polls may be a bit on the high side for them.

Placing assumption on top of assumption

I feel as if I am placing a chair on top of a table that is balanced on a beachball now.  I have tentatively assumed that Lord Ashcroft's constituency polls are there or thereabouts on UKIP's share of votes.  I also am assuming that the Conservative vote share in Conservative/UKIP marginals may be understated by a bit.  With these assumptions in mind, I guess that UKIP are in the lead in only one seat right now: Clacton.  Possibly it is ahead in Thurrock and Rochester & Strood also: we don't know.  UKIP will hope that South Thanet will fall to them with Nigel Farage's high profile, though it remains moot whether that will be an advantage.

Even if it is not leading in many seats, UKIP seems to be close to the Conservatives in a handful of other seats.  (We have no external information to judge how UKIP's polling results in Labour/UKIP marginals reflect the underlying position, even on the most tentative of bases.)

So while UKIP may on a best guess be a bit behind the Conservatives in its key targets, it is breathing down the Conservatives' necks hard.  Two things may help them overtake the Conservatives from here: a good ground game and any improvement in the polls between now and 7 May.

In general, we should expect the Conservatives to have a better ground game than UKIP.  They are longer established, should have better systems in place and are better resourced.  Set against that, what are now Conservative/UKIP marginals have previously been safe Conservative seats, so the Conservatives may not have anything like the information available in these seats that they would have in previously-identified marginals. 

Castle Point looks likely to be an exception to the rule.  It was the first seat ever to be represented by an MP under the UKIP banner and UKIP have the local support of a long-established and well-organised group, the Canvey Island independents.  Lord Ashcroft's poll shows a statistical dead heat and that currently UKIP are well ahead on the contact rate.  The Conservatives are going to have to pull out the stops to hold that seat.

For UKIP to get more than four or five seats, though, my current guess is that they are going to need to make progress nationally.  But I must stress that this is a guess based on some fairly flimsy assumptions.  There are many alternative readings of the current state of play that are entirely plausible.  If you vehemently disagree with me in one direction or another, I will not be matching you in my vehemence.  Unless there's such a thing as vehement diffidence.

What we don't know

There's a contrast to be drawn here with the position of the Lib Dems.  There is much that we don't know about how the Lib Dems are going to perform, but the general battle lines are understood.  The Lib Dems are going to poll nationally nothing like what they polled in 2010, but will be trying to rely on their local reputation in constituencies where they are strong to salvage what they can.  The national polling from each pollster tells an internally consistent picture over the last few months and the various pollsters don't differ all that much from each other.  We have had a lot of polling from Lord Ashcroft in specific constituencies and the uncertainty is about how to interpret this information, not in the sense of how does it read across but in the sense of how accurate the polling actually is.  In the broadest terms we expect that the Lib Dems are going to do very badly against the SNP and Labour, can hope to hold the line a bit better against the Conservatives and the uncertainty is how far along each of these fronts the Lib Dems will be able to hold up.

We have none of these certainties with UKIP.  The pollsters are poles apart in their picture of how UKIP is doing right now.  We have limited constituency polling of seats in which UKIP have an interest and because the various national pollsters have such a radically different view of how UKIP is doing right now, we can't be confident about how to put this constituency information into an overarching picture.  Worse, we have no track record of how UKIP performs in general elections rather than by-elections, so we don't know whether UKIP will benefit from or suffer from tactical voting or whether its vote will be squeezed.  We have quite a lot of information and all of it is confusing.

How to use that lack of knowledge

We are used to betting by interpreting the knowledge that we have.  But on this occasion we can say fairly definitively that we don't have a clear picture of what is going on at a national level, that we don't have a clear picture of how that might change by May and we don't have a  particularly clear picture of how that will map onto individual constituencies.


Should we give up?  No.  The right decision may ultimately be no-bet, but just because we don't know anything doesn't mean that's automatically the right conclusion without further investigation.  Betting is a process of managing uncertainty.  This is just a different kind of uncertainty. 

From the betting reaction to the most recent polls from Lord Ashcroft, which has seen UKIP prices lengthen and Conservative prices shorten in Castle Point and Boston & Skegness in particular, it appears that punters had got ahead of UKIP's actual progress.  There may be a ripple effect into other constituencies where UKIP are short priced, though I will not be following that bandwagon if it happens.  UKIP seem to be decentralised as a party and it seems dangerous to extrapolate too much from constituency to constituency.

If you're very bearish on UKIP's chances, Ladbrokes' over/under markets are worth considering.  For example, you can get 11/2 on UKIP getting under 1.5 seats.  That doesn't appeal to me at all because of the decentralised nature of UKIP's operation, but if you take the view that both Survation and Lord Ashcroft are presenting a bit too rosy a view of UKIP's chances, there are worse bets.  You're taking one view of the uncertainty around UKIP's chances that stems from the polling chaos if you place this bet.  So long as you appreciate that view might very well be wrong.

At the other extreme, at the end of November I was able to back UKIP getting under 19.5 seats at 1/3.  The price for that is now 1/10 and I could if I wished close out this bet for a profit by backing the other side of the bet at 6/1.  I'm not remotely inclined to though, because UKIP's route to 20 seats looks at least as problematic as it looked then, and time is running out for the kippers to find that route.  7/2 for over 9.5 seats looks a little more tempting, but I'm still not tempted.

I don't feel the need to back UKIP in any constituency just now, but equally I don't yet feel the need to close off my current bets.  My expectation is that UKIP will hit the headlines again during the election campaign and that they will be backed in accordingly in seats where they are thought to have half a chance.  Punters have already got a bit ahead of themselves once in this direction, so I hope that they will do so again.

I hope to see more of that type of enthusiasm in the election period and if so I will be aiming to take advantage of that, but will probably ultimately be closing off most of my bets where UKIP are in the mix.  I'd need to be a lot surer of my ground than I am at present to leave some of the short priced bets open.

So my current position is indeed no-bet as regards UKIP, but I hope that it's a positive no-bet.  My position may change soon enough.

Tuesday 10 February 2015

Adding and subtracting; the implications for the seat bands markets

Prompted by politicalbetting poster Pulpstar, I've been looking at the seat band markets today.  These are not markets I usually like playing, because there are usually substantial overrounds and the bands are usually fairly narrow.  But I've noticed something a bit odd about these markets.  To understand what's odd, I need to start elsewhere.
 
Here are the current spreads on Sporting Index's seat markets:


The mid-points do not differ markedly from the assumptions in the over/under seat markets on the conventional bookies.  I haven't recently updated my list of assumed seat winners from the constituency markets at best prices, but Lord Ashcroft put this up on twitter today and I have no reason to doubt its accuracy:



Anyway, my point is this.  Whichever market you look at, the Conservatives and Labour are currently expected to get something of the order of 550 to 560 seats between them.  This is not a zero sum game, because either could do better or worse against UKIP, the Lib Dems or the SNP than is currently anticipated, but assuming this share of seats as a starting point seems reasonable enough.

If we look at the seat band markets, however, we get a different picture.  Here are the current best prices:

Con
0-200                50/1 (Ladbrokes)
201-225             25/1 (Ladbrokes)
226-250             16/1 (Ladbrokes)
251-275             7/2 (Ladbrokes, SkyBet)
276-300             15/8 (SkyBet)
301-325             11/4 (Ladbrokes, SkyBet)
326-350             7/1 (SkyBet)
351-375             16/1 (Ladbrokes, SkyBet)
376-400             25/1 (Ladbrokes, SkyBet)
401+                 50/1 (Ladbrokes, SkyBet)

351+                 8/1 (William Hill)

Lab
0-200                33/1 (Ladbrokes, SkyBet)
201-225             20/1 (Ladbrokes, SkyBet)
226-250             12/1 (Ladbrokes)
251-275             11/4 (William Hill)
276-300             15/8 (Ladbrokes, William Hill, SkyBet)
301-325             3/1 (Ladbrokes, William Hill, SkyBet)
326-350             8/1 (Ladbrokes)
351-375             33/1 (Ladbrokes)
376-400             50/1 (Ladbrokes)
401+                 66/1 (SkyBet)

351+                 7/1 (William Hill)
The pivot is not around the 550-560 seats that every other market assumes.  For starters, the favourite for both is 276-300.  That's possible if they share 560 seats, but only if both are at the bottom end of that band.  That really is like landing on a pinhead.  More likely, one would get in that band and one would get in the 251-275 band.

The pivot on these markets seems to be something closer to 570 seats shared between Labour and the Conservatives, or maybe a little higher.  But with the rise of the SNP, that looks about a minimum of 10 seats too high.  Note that chart published by Lord Ashcroft - the SNP are expected to take 35 seats.  Yet that still lags a bit behind the preponderance of the polling of the last few months.

The pricing gets odder when you go outside that band.  If Labour are getting 301-325 seats, the chances are that the Conservatives are getting somewhere around 250 seats.  But while Labour getting 301-325 seats is a 3/1 shot and getting 326-350 is an 8/1 shot, you can get 16/1 on the Conservatives getting 226-250.  Something is wrong somewhere in the pricing there.

Clearly it's possible for Labour to get c305 seats and the Conservatives to get c255 seats, taking the Conservatives outside the 226-250 band.  But c245 seats for the Conservatives is also perfectly possible on a seat tally for Labour of c305 seats. And if Labour are in this band, they might be getting 315 to 320 seats.  In those circumstances, you would definitely expect the Conservatives to be in the 226-250 seat band. 

You can make the prices for the Labour 301-325 band and the Conservative 226-250 band work if you take the view that Labour will probably get into the 301-325 band only if they claw back some of their current losses from the SNP (that doesn't explain the short price for 326-350 seats, but I'll leave that be).  That's an entirely valid way of looking at the position, though I don't share that view.

But that doesn't really explain the reverse - you can get 11/4 on the Conservatives getting 301-325 seats, but 12/1 on Labour getting 226-250 seats.  You can imagine Labour doing relatively better against the SNP than against the Conservatives as a matter of theory, but as a matter of practicality, it sounds unlikely.  It's just as possible that the SNP could do better against Labour than currently expected, making 226-250 more likely if the Conservatives outperform.

What are the odds in practice that either Labour or the Conservatives will get more than 300 seats?  For Labour, there are two routes from the current position: to win more seats from the Conservatives than is currently expected or to hold onto more from the SNP than is currently expected (a bit of both is allowed, of course).  For the Conservatives, there's only really one route: to win seats more from Labour than is currently expected (though they are allowed to pick up more than currently expected from UKIP and the Lib Dems, these are unlikely to get them to the 300 seat mark by themselves).

In both cases, it has to be odds against, but it's not out of the question, and most of those possibilities in both cases will be concentrated in the 301-325 band. 

With that in mind, backing Labour at 12/1 in 226-250 and the Conservatives at 16/1 in 226-250 both seem like good bets to me as proxies for the possibility of either Conservative or Labour outperformance against current market expectations.  Clearly both are odds against bets.  But the combined probability of 13/2 looks too low to me - they're probably twice as long as they should be.  I'm on both.

And more generally, if you insist on playing the Labour and Conservative seat bands, I suggest that you go low on the opposite party.  It seems to offer better value.

Saturday 7 February 2015

Blessed are the kingmakers: who will the Lib Dems be after May?

Conventional wisdom now has it that we are going to have a hung Parliament.  And it is entirely possible that the Lib Dems will have a central part to play in the formation of the next government.  A lot of thought has been given to how many Lib Dems will survive.  Not much thought has been given to what the Lib Dems will look like in the next Parliament. 
 
The Lib Dems span a wide range from those who place most emphasis on the social aspects of liberal democracy, like Charles Kennedy and Tim Farron, to those who place most emphasis on the small state aspects of liberal democracy, like Danny Alexander and Jeremy Browne.  If the Lib Dems are to lose a lot of seats, as we currently expect, the balance of the Parliamentary party could be crucial to the shape of negotiations.

Who's going?

I'm going to assess this in the main by strict reference to the odds.  I'll divide this into four groups: those who look very likely to stay (1/2 or shorter); those who look very likely to go (2/1 or longer); those who are more likely than not to stay but no sure thing (between 1/2 and evens); and those who aren't expected to stay but who have a decent chance of overcoming the odds (evens to 2/1).  Here is the current state of play in the Lib Dem held seats:

https://drive.google.com/file/d/0Bygi8eZw-4q1ZTQ1djlSWjIyZUE/view?usp=sharing

As you can see, this is not a pretty picture for the Lib Dems.  They are rated at 1/2 or better in only 18 seats and are odds-on in only 11 more, while they are at 2/1 or worse in 21 seats and evens or worse in a further seven seats.  They look set for a spanking and the only question is how sore they're going to be by the end.

Doomed (2/1 or longer)
Safe (1/2 or shorter)

Danny Alexander
Gordon Birtwistle
Annette Brooke (standing down, replacement Vikki Slade)
Sir Malcolm Bruce (standing down, replacement Christine Jardine)
Lorely Burt
Sir Menzies Campbell (standing down, replacement Tim Brett)
Michael Crockart
Lynne Featherstone
Stephen Gilbert
David Heath (standing down, replacement David Rendel)
John Leech
Tessa Munt
Alan Reid
Sir Robert Smith, Bt.
Ian Swales
Jo Swinson
Sarah Teather (standing down, replacement Ibrahim Taguri)
Viscount John Thurso
David Ward
Jenny Willott
Simon Wright

Norman Baker
Tom Brake
Vince Cable
Alistair Carmichael
Edward Davey
Tim Farron
Don Foster (standing down, replacement Steve Bradley)
Martin Horwood
Simon Hughes
Charles Kennedy
Norman Lamb
David Laws
Greg Mulholland
John Pugh
Bob Russell
Sir Andrew Stunell (standing down, replacement Lisa Smart)
Steve Webb
Roger Williams

Up against it (evens to 2/1)
At risk (1/2 to evens)

Sir Alan Beith (standing down, replacement Julie Pรถrksen)
Jeremy Browne (standing down, replacement Rachel Gilmour)
Andrew George
Duncan Hames
Mike Hancock (expelled from party, replacement Gerald Vernon-Jackson)
Sir Nick Harvey
Michael Moore

Paul Burstow
Nick Clegg
John Hemming
Julian Huppert
Mark Hunter
Stephen Lloyd
Dan Rogerson
Adrian Sanders
Mike Thornton
Mark Williams
Stephen Williams


The politics of all this

If we want to see what the Lib Dems will look like after the next election, we should focus on the right hand column.  Obviously there may be upsets, but if the bookies are even approximately right, the Parliamentary party will be drawn largely from those names.  But let's look at some different cases.

The full-on hammering

If the Lib Dems take only their safe seats, Nick Clegg would be gone.  Three leading rightwingers will remain: Ed Davey, David Laws and Norman Lamb.  But the influence of leading leftwingers will be much greater than at present, especially with Nick Clegg out of Parliament.  Vince Cable, Charles Kennedy, Tim Farron and Steve Webb are generally thought to be much more comfortable looking leftwards (though Steve Webb has worked effectively in the coalition government, making his policy area very much his own).  

Actually, though, the internal composition of the Lib Dems in these circumstances wouldn't make much difference in a hung Parliament: the Lib Dems' 18 seats would be unlikely to be decisive, except in circumstances where they had only one choice. Also, after such a hammering, it's unlikely that the Lib Dems would want to be in government with anyone.  They'd want to regroup.

The heavy pounding

If the Lib Dems take their safe seats and those that they are narrowly odds-on to take, they would tally a more respectable 29 seats.  At that point, they could start having influence in hung Parliament discussions.

And this group would be dominated by the left.  14 of the MPs who would remain are in the Beveridge Group, who represent the socially liberal wing of the party:
+ http://en.wikipedia.org/wiki/Beveridge_Group

This group is much keener on the public sector than market provision.  These 14 MPs do not include Vince Cable, Steve Webb or Charles Kennedy.  Other MPs who would survive the cull, such as Paul Burstow, have made no secret of their hostility to the Conservatives.

Even if Nick Clegg makes it back, it's hard to imagine a grouping like this cosying up to the Conservatives for another five years unless they are driven to it by numerical necessity.  And it appears that Nick Clegg is going to have a hard fight now.  Without him, I don't see any Lib Dem voice who is going to be arguing hard for a link-up with the Conservatives again.

The light dust-up

If the Lib Dems, Houdini-like, manage to keep those seats which they are currently up against it to keep, they will have 36 seats.  At this point, they start to become credible coalition partners and not just confidence and supply material.   Note that the seven seats in the "up against it" category are all Conservative targets. 

The extra seven MPs would not obviously alter the weight of the party decisively.  Duncan Hames is an acolyte of Nick Clegg and Andrew George is in the Beveridge Group.  Sir Nick Harvey and Michael Moore are both one-offs who are not too easy to categorise.  We can infer from his long friendship with Mike Hancock that Gerald Vernon-Jackson is likely to be in tune with the Beveridge Group.  But otherwise we can't easily tell where prospective MPs fit on the spectrum: their campaigns are very local.  

Overall, we should still expect the Lib Dems to want to look left on this scenario also.  For a contrary view, I attach an article from the New Statesman's site May2015:

http://may2015.com/featured/the-lib-dems-will-be-a-centre-right-party-after-the-election/

I think that this article is utterly wrong, since it ignores the politics of the individuals who will make up the post-2015 Lib Dem Parliamentary party.  But make your own mind up.

Betting implications

If I am right, it will not be enough for the Conservatives to gain most seats to stay in power.  They will need to get a sufficient number of seats so as to make it practically impossible for any other stable government to be formed.  To form a stable government, the Conservatives with the Lib Dems are going to need something of the order of 330 seats as a minimum.  That means that on either the "hard pounding" or the "light dust-up" scenarios, the Conservatives will need 300 seats or near enough to form the next government.  If they have fewer, the Lib Dems will find the lure of the left irresistible.  295 looks the absolute minimum for a continuation of the Conservatives in power.

The midpoint with the various bookies on their over/under bets on the Conservatives is 281.5 for Ladbrokes, 282.5 for Paddy Power and Coral, 283.5 for SkyBet and 285.5 for Bet365, and with all of these bookies you can back either option at the same price.  The Sporting Index spread at present is 281-287.  At these levels, the Conservatives would not continue in office.

So lay David Cameron on Betfair to be Prime Minister after the next election (available on Betfair at prices between 4/6 and 4/5).  This should be somewhere around 6/4.  It is certainly not close to an evens bet at present, even if you believe that the Conservatives will get most seats, which itself is far from clear at present.  This price is a steal.  You can get 13/8 with William Hill that David Cameron will remain Prime Minister to 2016, and that seems more realistic.

Alternatively, back Ed Miliband to be Prime Minister on 1 August 2015 at 11/8 with William Hill.  This is taking the risk that he might not survive the mayhem of hung Parliament negotiations, which will be a topic for discussion if Labour has considerably fewer seats than the Conservatives, so it's a bit less satisfactory.  But it remains a good bet.

And generally, place bets on post-election negotiations bearing in mind that the Lib Dems who return will have a firm political preference for swinging to the left.  If the Conservatives are to stay in power, they're going to have to do most of the work all by themselves.

Friday 6 February 2015

Timber! The great SNP price crash

It's time to take stock in the wake of Lord Ashcroft's Scottish constituency polls.  It's important to note that not a single vote has been cast in the coming Westminster election yet.  But its contours now seem so much sharper.  Shit just got real. 
How we got to where we are now
In 2010, the story of the Scottish Westminster election was simple: no change.  Not a single seat changed hands.  And for a long time, the expectation was for something similar in May.  This is what I wrote in April last year:
"Two things immediately stand out.  First, the SNP is fighting a battle on three fronts.  And secondly, the size of the swings required to make even limited progress is daunting.  In many of the more promising seats, the SNP would need to come from third or fourth.  Not only would they need to dislodge the incumbent, they would need to muscle past existing established challengers."
I looked at this again at the end of August just before the referendum, with a particular strategy in mind:
By this stage we had a full set of markets, but the SNP were still not expected to make any gains.  Indeed, they were shorter than 5/1 in only 14 constituencies.

That, remember, is less than six months ago.

Things changed rapidly after the referendum.  In the immediate wake of the referendum, the SNP polled well.  The effect was to result in a shortening in the prices of long shots, but betters were reluctant to back the SNP down to odds-on in any constituencies.  So the SNP dropped in just over a month from 33/1 to 5/1 in Glasgow East, for example, but were no shorter than 11/8 in Ochil & South Perthshire, having previously been 7/4:

 
The SNP then gained and kept a steady large lead in the polls of 20% or more, but the constituency markets did not move anywhere near as much as you would have expected (or as I expected).  This was the position just before Christmas in the Labour/SNP seats:
Despite the SNP benefiting from what pollsters were recording as a 20+% swing against Labour, the SNP were favourites to take only three seats from them:
And the slow shortening in SNP prices continued throughout January.  This was the position on Tuesday night, on the eve of release of Lord Ashcroft's constituency polls:
By this time the SNP were odds-on in 25 seats and favourites or joint favourites in four more.  Not bad, but still far below what you would expect of a party that was 20% ahead in the opinion polls.
Where we are now
What a difference a day makes.  24 little hours later, the SNP saw a price drop worthy of a shopping channel special:
 https://drive.google.com/file/d/0Bygi8eZw-4q1d05oazlrcGpYc0E/view?usp=sharing

The SNP are now odds-on in 39 constituencies and favourites in one more.  They are no longer than 5/1 in any constituency in Scotland, and longer than 2/1 in only eight constituencies.

This sudden catching-up with the opinion polls is inexplicable on any rational basis.  Lord Ashcroft's polls told us little that we could not have inferred from the Scotland-wide polls.  As Lord Ashcroft is at pains to point out, his opinion polls are snapshots and not predictions.  No votes have been cast as yet.  And the fundamentals remain the same.

But somehow seeing inferences confirmed by hard data has impressed on the betting public the reality of the current state of play that has been evident since at least Christmas.  The markets have finally caught up.

Interestingly, the markets have not slavishly followed the constituency polling.  Lord Ashcroft found the SNP to be ahead in Coatbridge, Chryston & Bellshill, but the SNP is still rated as an odds against proposition in this seat.  It's the overall picture that's had the impact, not the detail.

What does Lord Ashcroft's polling actually tell us?

There are few examples of required reading in political betting, but Lord Ashcroft's summary of his Scottish constituency polls is definitely in that category:

http://lordashcroftpolls.com/2015/02/scottish-battleground/

All the things to be concerned about in constituency polling continue to apply.  These polls are more reliable in aggregate than individually - there is always the risk of an outlier, and getting accurate samples on a constituency basis is a greater challenge than balancing samples on a broader basis.  For that reason, I agree with the markets in having a polite mild scepticism about the SNP's chances of taking Coatbridge, Chryston & Bellshill - a small SNP lead in one constituency poll is not something to bet the house on.

This polling is not especially good for the SNP by recent standards.  The SNP averaged 48% in the 16 constituencies polled, which is not so far away from the average tally in recent Scotland-wide polls (for example, the last two YouGovs have recorded them at 47%, the last two Ipsos-MORIs have had the SNP at 52% and 53%, the last three Survations have had the SNP at 46% twice and 48%).  But these constituencies were chosen specifically because they were in areas that were particularly Yes-friendly.  So we might expect the SNP to be recording lower percentages in other areas.  It is noteworthy that the one constituency polled where Yes did not do particularly well, Gordon, shows the SNP benefiting from a markedly smaller (though still hefty 15.5%) swing.  These straws suggest that Lord Ashcroft's polling is perhaps more consistent with the 43% vote share that ICM found in December or even the 41% vote share that Panelbase found in January.

Set against that, there are good technical reasons for thinking that this polling may understate the SNP's position.  Lord Ashcroft weights his polls by reference to respondents' recall of their vote in the 2010 election.  That makes sense in England, but in Scotland the voting pattern in the 2011 Holyrood election was very different from 2010, and there is evidence that some respondents get confused.  Since the SNP did far better in 2011 than 2010, their vote may be being weighted down in these polls.

Leaving aside Gordon, the SNP swing is remarkably uniform in the seats polled (considering the sizes of the swings), ranging from 21% in Airdrie & Shotts to 27% in Dundee West and Motherwell & Wishaw, with tight clustering around the 24% mark in most seats. There is no obvious reason that I can see not to apply this type of percentage across the board in Yes-friendly seats as a starting point.

So where might things go from here?

What are Labour's chances of winning back the defectors?  Quite a bit has been made of Lord Ashcroft's observation that "Just over two thirds (68%) of switchers from Labour to the SNP say they definitely rule out voting Labour again in 2015 – which means nearly one third are at least open to the idea of returning".  This straw has been quite firmly clutched in some quarters.

This looks to me like teasing on the part of his Lordship.  He might equally have pointed out that only 65% of current Labour supporters have ruled out the option of voting for the SNP in 2015.  In other words, things look marginally more likely to get worse for Labour from voter movements than better.  And Labour would in any case have to convert large numbers of people who have changed their allegiance but who haven't finally excluded an idea in three months.  That sounds like a tough challenge to me.

Can Labour pull it back still?  There's no shortage of people making suggestions.  Isabel Hardman of the Spectator suggests that Labour need to pound the pavements:

http://blogs.spectator.co.uk/coffeehouse/2015/02/ashcroft-poll-proof-labour-needs-to-get-moving-across-the-pavement/

This is a good idea with one major flaw - Labour doesn't have the members to do this.  Still more worryingly for Labour, the SNP do.  The SNP is much more likely than Labour to be able to make the voter contact in these constituencies (though like Labour, it seems from Lord Ashcroft's polling, they have yet to get in gear on this in most seats), and I expect it to do so, firming up the nationalist vote rather than seeing it fade away.

One weakness for the SNP is that they seem to be selecting some surprising candidates to fight what are now very winnable seats.  Douglas Alexander is projected to lose his seat in this poll.  But his challenger is a 19 year old.  [EDITED TO REMOVE AN ERROR OF FACT] It might make the difference between success and failure for D Alexander (Labour) in a much tighter race.

We still don't have that clear a picture about what is happening in the areas where No was stronger.  Bearing in mind that Gordon showed a sharply lower swing to the SNP (though still very impressive at 15.5%) and that was no doubt increased by the high profile of Alex Salmond, who is standing in that seat, I intend treading carefully in such areas until we have more polling.  Lord Ashcroft has said that he is looking at more Scottish constituency polling this month - I can't wait. 

Betting opportunities

The party is over for those of us who have spent the last few months betting on the SNP in the constituency markets.  With the SNP now favourites or joint favourites in 40 seats, the markets have largely caught up with the pollsters now.

In areas with a strong Yes vote, I am taking that average 24% swing at face value, rating that swing down a bit in areas where Yes did fairly well rather than amazingly.  There may be some return to Labour ranks, but equally as I have noted, this may actually understate the swing to the SNP by a bit.  So using a 24% swing as a starting point seems fair enough.

With that in mind, the following continue to look like good bets on the SNP: Lanark & Hamilton East at 5/6 and Inverclyde at 5/6.  If you're feeling adventurous, Rutherglen & Hamilton West at 11/4 is interesting, being a seat that borders the Yes fortress of Glasgow.  I've put more on all three of these.

Some of the shorter priced bets also look good, especially in the constituencies polled where the SNP have a double digit lead.  The question you have to keep asking yourself is how Labour are going to make up the ground.  And the answer to that is not at all obvious.  How, for example, are Labour going to make up an 18% deficit in Cumbernauld, Kilsyth & Kirkintilloch East?  Yet the SNP remain priced at 1/2.  That is surely too long.  I'm not putting more on this though, simply because I have a fair amount at much longer odds.

What about in areas where Yes did less well?  I'm much more cautious there.  So far we have a constituency sample of one seat, Gordon.  We should not be extrapolating from single examples.  But equally we should not be ignoring them.  This is not the time to start ploughing in at short odds on SNP victories in such seats.  There may be betting opportunities on Labour in some of these seats, but I'm prepared to let them slip by, given the tenuous evidence so far.  If you're feeling brave, the 11/10 on Labour in Midlothian might be such an opportunity.  I'm tempted, but I'm going to pass for now.

What of your own model?

Fair question.  This is how I'd modelled a possible outcome on a constituency basis before we had more detailed results:

http://newstonoone.blogspot.co.uk/2015/01/bedtime-stories-extending-my-thoughts.html

Using Lord Ashcroft's polls as the benchmark, it stands up quite well, all things considered.  Lord Ashcroft agreed with it that the SNP would take six out of seven of the Glasgow seats.  Lord Ashcroft found that the SNP would take Coatbridge, Chryston & Bellshill and Paisley & Renfrewshire South, which my model did not predict.  Otherwise, the same results are expected.  Lord Ashcroft finds that the SNP are doing even better than my model suggested, and agrees that the swing to the SNP is apparently less marked in Gordon than elsewhere.

It remains to be seen what will ultimately happen.  But on the evidence that we have so far it seems that my model wasn't too bad a guide to what might happen.  Now we need more polls from less favourable terrain for the SNP to see how they are doing there. 

Tuesday 3 February 2015

Missing, presumed red? The 2010 Lib Dems

I have repeatedly turned to the question of the Lib Dems' impending fate.  It is one of the central questions of the next election: how far will the Lib Dems fall from the 23% they recorded in 2010 and where will those losses be felt most?  Where will their erstwhile voters put their cross if they do desert the Lib Dems?  The next election will be decided in large part by the answers to those questions.

Many electrons have been spilt over these questions, but we are not much closer to an answer, except chronologically.  Perhaps we can refine some of the questions a little further.In this post, I intend to look at the following:

1. What are the Lib Dems currently polling nationally?
2. Can that be reconciled with the constituency polls that Lord Ashcroft has been conducting?
3. If not, what might be causing the discrepancy?

What are the Lib Dems polling nationally?

I like questions of fact.  We need to take this pollster by pollster, so that we don't get confused by differing methodologies.

Populus - the Lib Dems are scoring between 8 and 10%.  They last went outside that range (on the low side) in mid-November.
ICM - the Lib Dems are recording between 10 and 12%.  They recorded 14% in December, but otherwise have stayed in that range since last May.
YouGov - rather more grimly for the Lib Dems, YouGov very consistently find their support in the 6-8% band.  They last went outside this narrow band in October, which considering YouGov poll five times a week represents astonishing consistency.
Survation - Survation have a much wider scatter.  In the last few months they have found the Lib Dems at 6%, 11% and all points in between.  Tentatively I'd suggest the range is 6-9%.
Ipsos-MORI - belying its reputation as being very swingy, Ipsos-MORI also has the Lib Dems in a narrow band, consistently finding that they poll between 7 and 9%.  The last time they went outside this range was in March.
ComRes - Aside from one poll in December, you have to go back to 2013 to find an occasion when ComRes found the Lib Dems polling outside a 7-10% range.
Lord Ashcroft - Since July, the Lib Dems have polled between 7 and 9% with Lord Ashcroft in all bar two polls (one with them at 10%, one at 6%).
Opinium - until this week, the Lib Dems recorded between 6 and 9% since July (this week's poll had them at 5%).
TNS-BRMB - the Lib Dems have recorded in a wider range with TNS-BRMB, with scores between 5 and 9% in the last year.

So if we look at the position on a pollster by pollster basis, it seems fairly clear that the Lib Dems are becalmed in the polls.  If you believe ICM, they are in low double digits.  Otherwise, the consensus has them in high single figures, with an average somewhere around 8% for almost all pollsters other than ICM, with no meaningful variation over the last few months.

Isn't it nice when a factual question has a fairly clear answer?

Can that be reconciled with the constituency polls that Lord Ashcroft has been conducting?

Gratifyingly, Lord Ashcroft's estimation of the Lib Dems' national polling is very mainstream at a stable 8%, which means that his constituency polls don't need calibration for changing national vote shares and can be read across without too much being required in the use of cold towels.

Lord Ashcroft, reasonably enough, has so far largely confined his constituency polling to marginals.  As a result, we have a skewed set of seats, and we'll have to fill in some blanks.
 
The constituency polling conducted so far falls into the following categories: Lib Dem/Conservative battlegrounds; Lib Dem/Labour battlegrounds; Conservative/Labour battlegrounds; Labour/UKIP battlegrounds; and Brighton Pavilion.  But that may be over-precise - a simple division between seats where the Lib Dems are in contention and the rest may suffice for now.  Obviously, the Lib Dems want to find as many Lib Dem voters as possible in the first set of seats, and if they have to shed any, they would much prefer to shed them in the second set of seats.

And they do seem be shedding votes in the second group of seats.  Across the 89 constituency polls (some are repeats in the same constituency) in such seats, the Lib Dems are averaging between 6 and 7% (closer to 7%, in fact).  This conceals substantial variations: the Lib Dems would lose their deposit in 23 of these polls and they are recorded at as low as 2% in North Warwickshire and Thurrock. 
 
They are also shedding votes in their own battlegrounds, but still averaging over 30% across the 57 opinion polls conducted in constituencies where they are in contention (again, there are some repeats).

All well and good, but does this add up?  To which the answer is, mostly.  If we reckon that there are say 70 seats where the Lib Dems would see themselves as in serious contention and use an average of 30%, and use 6.5% as an average for the other 561 mainland seats, we get a Lib Dem overall poll rating of just over 9% - a full percentage point ahead of the centre point of their national polling. 

So the constituency polling is in the same ballpark as the national polling.  It's broadly plausible.  But there is a bit of a difference, and the constituency polling suggests that the Lib Dems are doing a bit better than you'd expect from the national polls.

It may not sound like a big discrepancy, but in the context of such low national poll ratings, 1% is significant, given the sheer volume of national and constituency polls that we have had and their consistency.  It could make the difference of quite a lot of seats, depending on how the two are reconciled.  It needs explanation, or an acceptance that national polling and constituency polling are telling us slightly different things.

Why are the Lib Dems apparently overperforming a bit in the constituencies?

There are a range of possible explanations, not all of which exclude each other.  Here are the possibilities that spring to my mind.

1. Labour safe seats and Conservative safe seats may behave differently from marginals

It is theoretically possible that the Lib Dems will do worst in Labour and Tory safe seats. That doesn't sound desperately plausible to me. No one else will be squeezing their vote.

I suppose that the Lib Dems might be doing particularly badly in safe Labour and Conservative seats where UKIP are pressing.  I can't say that I'm overwhelmed with confidence in that theory.
 
2. The Lib Dems may be losing more support in their safest seats

Same theoretical possibility. Still less plausible: the Lib Dems will be squeezing out the last ounce in such seats. 

3. The Lib Dems may be losing more support in areas where Lord Ashcroft has not so far conducted much polling (Scotland especially, but perhaps Wales and north east England also)

There may be something in this.  The Lib Dems do seem to be doing particularly poorly in Scotland, if the Scottish only polls are to be believed.  However, these seats make up a relatively small proportion of the total, so the underperformance would need to be severe indeed to make the sums add up.

Lord Ashcroft is releasing some Scottish constituency polls tomorrow.  We shall see more then, I guess.

4. The Lib Dems will in fact do less well in one or both of these sets of constituencies 

This has to be a possibility.  The Lib Dems are going to make no effort in seats where they are not in contention, and Lord Ashcroft's polling is based on respondents being prompted to think about who to vote for in their own constituency.  Lib Dem supporters will need to be very committed (or very strongly believe in their civic duty) to turn out in practice.  I expect that some of those who would say that they are going to vote Lib Dem when prompted will be seduced by another party or will stay at home.

What of seats where the Lib Dems are in contention?  Lord Ashcroft derives his headline poll numbers from the second of his two part question, where voters have concentrated on who is standing in their constituency.  That relies on voters going through the same two stage process.  If the Lib Dems' electoral machine is running smoothly, that may be safe.  In seats where the Lib Dems are not so fluent, perhaps it isn't.

5. The national polling understates how well the Lib Dems will do in practice 

This also has to be a possibility.  It may be that voters will turn their minds to their own constituency and decide that the Lib Dems after all should be favoured with their cross.  It is noteworthy that ICM, the one pollster that gets respondents at a national level to think about their constituency, is the one pollster that finds the Lib Dems with a higher level of support.

The Lib Dems may in practice overperform in seats where they have the organisation to get out the vote.  Past performance is not always a guide to the future, but it is at least a guide to where the Lib Dems have the organisational skills.

Conclusions

The Lib Dems have been remarkably steady (or flat, to use a less friendly word) in the national polls.  There is no immediately obvious reason why they should start putting on support in the run-up to the election nationally, though they may do so in their local constituencies if their local MP has a particularly strong profile.

So the key question remains where they will get that support.  We can now expect a lot of lost deposits for the Lib Dems.  As noted above, I expect the Lib Dems to do best where they are organisationally strongest and that the identity of the local MP will be vital to their chances of survival.  For the Lib Dems, everything is local.  There is nothing in the offing that suggests that the Lib Dems are going either to suffer annihilation or to avoid substantial losses.   But the detail remains uncertain, and will depend in large part on which pollster's methods is closest to the mark.

I appreciate that none of this is particularly ground-breaking but sometimes important conclusions aren't.