I'd rather be happy than right: 2015 personal election review

Well that didn't go as expected.  With the luminous exception of Scotland, I was completely wrongfooted by how the election result turned out.  I could argue that so was everyone else but that would be a bit of a cop out.  It's my turn to eat some humble pie.

So what did I get right and what did I royally mess up?  The best place to start must be my end of year predictions:

http://newstonoone.blogspot.co.uk/2014/12/2015-may-and-everything-after.html

These were, in order:

1. There would be a hung Parliament
2. It will be neck and neck between the Lib Dems and the SNP which would be the third party.
3. UKIP would get a good poll rating and few seats to show for it.
4. The Greens would take precisely one seat: Brighton Pavilion.
5. The debates would take place, basically in the format put forward by David Cameron.
6. The election campaign wouldn't change very much.
7. There would be a Labour minority government.
8. All of the leaders (except Nicola Sturgeon) would be in peril of losing their jobs.

Not good, are they?  I was right about UKIP and the Greens, half right about the debates (the format changed considerably) and right about the pressure on the leaders apart from David Cameron - though Nigel Farage's bouncing resignation was an innovation in British politics that I hadn't expected.  But for the rest, I was hopelessly wrong. 

There's no point sugaring the pill, this was a terrible set of predictions.  What went wrong?

Despite being very aware that polling was an unreliable friend and likely to be particularly unreliable this time with the rise of UKIP and the SNP and the crash of the Lib Dems, I still relied heavily on the opinion polls.  I think I acted correctly when I put together these predictions - you have to make use of the evidence that you have got in the absence of concrete reasons to doubt that evidence.  At the turn of the year, I had no such concrete reasons.

In fact, my year-end predictions were largely what I expected right up to 10pm on 7 May 2015 (but by that stage I expected the SNP to far outstrip the Lib Dems, though even then not by as much as in fact happened).  The polls had remained static, even though the mood music from both main parties and the campaign itineraries of the two main party leaders suggested something different from the polls.  That failure to revise my expectations was less forgivable.  Ed Miliband would not have been campaigning in North Warwickshire and David Cameron would not have been campaigning in Twickenham and Bath if they didn't believe that the election was far adrift of where the polls suggested.  One of them might have been wrong.  Both of them were most unlikely to be.  I should have taken more notice.

So I wasn't right.  But I am still happy.  Why?  Because my betting produced the returns I was looking for from them.  How did I turn a profit when I had been such a poor prophet?  I need to break this down into asset allocation, stock selection and timing.

As of 10pm on Thursday night, my portfolio of fixed price bets was comprised in the following categories:

1. Scottish constituency bets, almost all on the SNP
2. English constituency bets, the great majority on the Conservatives, but not exclusively if I saw bets that I thought offered disproportionate value on other parties
3. Hung parliament (I'll come back to this)
4. Labour most seats
5. Labour minority government
6. Laying David Cameron as Prime Minister after the election and backing Ed Miliband as Prime Minister after the next election
7. Miscellaneous others

I had placed some spread bets, but this time I didn't put huge sums at stake that way.  My focus was elsewhere.

You can see from the bets that I placed that these were partly indirectly hedging.  Relative longshot Conservative constituency bets were inconsistent with bets in categories 4, 5 and 6 in particular.  By selecting constituency bets that I felt were likely to outperform each party's run-of-the-mill seats for the same odds, I hoped to maximise returns.

I did so well on my Scottish constituency bets that everything else pales into insignificance.  They comprised roughly 80% of my winnings.  This was a triumph of asset allocation - all those hours I spent poring over the likeliest SNP wins proved entirely superfluous in the end (indeed, at an early stage I had managed to back the SNP to win Edinburgh South, the one seat that they did not take, though having obtained odds of 25/1 on that possibility I can scarcely be too upset).

But I did OK elsewhere also.  On Thursday morning, I thought that most of my English constituency bets on the Conservatives were heading for the virtual ashtray.  But on Friday morning, they largely came up trumps.  My strategy of seeking out value on both sides of the fence left me ahead.  Here, stock selection was more important.  And I did OK, particularly as regards the battle between Labour and the Conservatives.  

Labour took ten seats from the Conservatives.  Of these, I had placed losing bets on the Conservatives holding City of Chester, Hove and Ilford North.  But conversely I did bet on Labour taking Brentford & Isleworth and Ealing Central & Acton.  I avoided betting on the other five either way.  Meanwhile, I successfully backed the Conservatives (at fairly long odds) to keep many of their other most marginal seats, and while I'd backed Labour to take some other seats, I am not embarrassed by my choices there either.  Taken as a whole, I think I did a pretty fair job of identifying which Conservative seats might underperform and where Labour might outperform against the Conservatives.

In other battlefields, the story was much closer to neutral overall (I did conspicuously badly with UKIP).  But I did best where I'd bet most.  

I made one other critical decision early on Friday morning which was worth a four figure sum to me.  After seeing the exit poll, I was aware that it was in doubt whether the Conservatives would merely be the largest party or whether they might get an overall majority.  Following the results in Nuneaton and Swindon North, where the Conservatives had outperformed the exit poll in each case, I reversed my previous position on no overall majority and backed the Conservatives getting an overall majority at fairly long odds.  Timing was everything on this. 

If this all sounds too smug, please return to the top and reread my serving of humble pie.  I will.  Overall, I made a complete hash of my predictions.  Fortunately, I got away with it.  This time.

Reality check: testing the betting markets against external data

Up to now I have been working on the basis that the various betting markets on politics reflect the balance of opinion among those who are willing to place money on the outcome, and thus represent the current conventional wisdom about what is most likely to happen.  But that begs the question who are these people who are willing to place money on political bets.  Are their instincts skewed in any ways?

I do not have inside information on those who actually place bets on the political markets, but we can make deductions from the prices as to what motivates them to bet.  So let's look at one set of constituency markets.

A straight choice

To make it easy, I'm going to focus just on the Conservative-held seats that are Labour targets.  I'm going to exclude all seats where any other party is priced at 5/1 or less (or shorter than Labour), to cut out triangular fights.  That leaves 91 of Labour's first 125 targets.  Here they are arranged in order of swing required for Labour to take them:

https://drive.google.com/file/d/0Bygi8eZw-4q1aTdyeVlzanNRSDQ/view?usp=sharing

The first thing to notice is that the prices are driven by considerably more than swing.  The first seat on the list, North Warwickshire, is rated by punters as a less likely Labour gain than the 23rd seat (Brentford & Isleworth).  The 87th seat on the list, Portsmouth North, is rated a more likely Labour gain than the 50th seat, Vale of Glamorgan.  There is clearly a correlation with swing, but it's far from perfect.

Swing vs odds

Let's look at these seats reorganised by odds (I've used the Conservative prices):

https://drive.google.com/file/d/0Bygi8eZw-4q1X0dpN2g2bEdkaWc/view?usp=sharing

I've also changed the numbering so I you can easily see the movements between these two lists.  On average, there is a difference of over six places for each seat between these two lists.  Note, I have deliberately constructed a set of comparable seats with minimal third party interference (Dudley South, Crawley and Brigg & Goole being the only seats on the list where any other party is significantly affecting the two main parties' odds).  Yet punters seem sure that uniform national swing is going to be varied from fairly markedly even within this most typical of sets of seats.

19 seats shift 10 or more places, as follows:

20 or more places

Portsmouth North

Stockton South

Enfield North

15 or more places 

Dudley South

Calder Valley

Morecambe & Lunesdale

Ealing Central & Acton

Bristol North West

10 or more places

Erewash

Plymouth Sutton & Devonport

Gloucester

Stafford

Corby

Brentford & Isleworth

Preseli Pembrokeshire

Pudsey

Crawley

Milton Keynes South

Rugby

The bigger the shift, the more compelling a reason we should be looking for to justify that shift.  In some cases, the reason will become apparent.  In others, I can't see why there are such big movements.

Uniform national swing

Leaving aside the internal movements, let's think further about the implications of uniform national swing.  In the last few YouGov polls, Labour have posted 33% (give or take a point either way) and the Conservatives have posted 32%.  That is a swing across the UK of 4%.  That would mean that Labour could reasonably expect to take every seat up to Stevenage, the 47th seat on the list.  But Labour are in fact odds against or second favourite  in every seat bar one after Keighley, the 39th seat on this list, where only a swing of 3.1% is required.

If that is true of national swing, it is still more true of the position discounting Scotland.  Labour are racking up a swing against the Conservatives of 5% or more outside Scotland, suggesting still greater gains.  Evidently this does not faze punters on the constituency markets. 

And all this is minor compared with the "most seats" markets.  On Betfair, Labour and Conservative are both just about evens to get most seats.  This implies a swing to Labour in these constituencies of under 2%, even allowing for the SNP taking a big bite out of Labour's Scottish seats.  If you take this view of affairs, you should instead be betting on the Conservatives at odds against on the constituency markets in seats like Hastings & Rye and Nuneaton.

Clearly punters are expecting the swing to Labour to subside a bit from present levels.  You need to decide for yourself whether that is reasonable.

Geography

Geography is destiny, so they say, but punters don't seem convinced.  These 91 seats are spread across nine regions: there are no Scottish seats.  If we leave the one seat in the north east to one side as representing too small a sample (I will not be expecting a major Conservative renaissance in the north east on the basis of Stockton South), the regional pattern is quite muted.  In London and the South East, the Conservatives are rated by punters at 3.6 and 3.7 places per seat worse than uniform national swing would suggest.  In Wales, the Conservatives are rated by punters at 2.5 places per seat better than uniform national swing would suggest and in the East the Conservatives are rated by punters at 3 places per seat better than uniform national swing would suggest.  But otherwise, the effects of geography aren't worth noting.  I should record that the result in the South East is almost entirely attributable to one seat: Portsmouth North (and that we would see the Conservatives rated at 3.7 places per seat better in the West Midlands than uniform national swing would suggest if the unusual seat of Dudley South is excluded).

I think that punters are probably underestimating the effects of geography at the next election.  In particular, as we shall see, there is quite a bit of evidence that Labour are outperforming in London and I'm not convinced that this is adequately reflected in the markets yet.

Polls

Unlike previous general elections, we have had a plethora of constituency opinion polls, particularly for this section of constituencies.  Survation have polled in Stockton South and Crewe & Nantwich, while Lord Ashcroft has polled in no fewer than 50 of these seats.  Both Lord Ashcroft and Survation stress that opinion polls are snapshots not predictions, but in the absence of more direct methods of getting information for predictions, opinion polls are inevitably going to be used to some extent to help punters make predictions.

Here are all the polls for constituencies on this list, in decreasing order of Labour lead:

https://drive.google.com/file/d/0Bygi8eZw-4q1SUs3OGJJSXpuUzQ/view?usp=sharing

I've also included a column in the table of seats organised by odds (in the link given further up) which includes the headline poll figure.

Only 11 of Lord Ashcroft's constituency polls show a lead for the Conservatives, and only one of the Survation polls.  But punters don't really seem to be taking all that much notice of these constituency polls.  Only 12 of the constituencies noted above that have moved ten or more places out of order from uniform national swing on the odds table have constituency polls.  Labour are shown just on the seats polled to be presently ready to take 40 of the 51 seats polled.  Either the punters don't believe the polls or they are expecting swingback to the Conservatives.

As can be seen from the table above, there are some anomalies.  The Conservatives are rated less likely to take Hove than Carlisle, despite recording only a 3% deficit in Hove and an 11% deficit in Carlisle.  The Survation poll in Crewe & Nantwich seems to have been essentially ignored.

Should punters be taking more notice of the constituency polls?  It's hard to know how much weight to put on them, because in Britain we have limited experience of their use and this has been drastically increased in advance of this election.  I have reservations about them, since they involve the compilation of non-standard polling bases for each constituency.  The risk of mistakes is much greater and there is always the risk that a poll in a particular constituency is an outlier.  A surprising result may just be a quirky sample - these things are to be expected from time to time.

But having disparaged them, I also recognise that they are the best information that we have in most seats.  We ignore them at our peril.

How best to use them?  It would be unwise to rely on single polls uncritically.  Where we have more than one poll in a single constituency, as in Stockton South, we can be much more confident about the underlying picture.  Unfortunately, that is the extent of the seats on this list with two polls.

As with any opinion poll, the headline figure of each poll should be treated as in the rough area of the actual lead.  So a finding of a 5% lead might easily be 2% or 8% - 5% is merely the single most likely lead on the data available.  Where the two parties are found to be dead level, it is equally likely that each was in reality in the lead in the constituency at the time the poll was taken.  But a finding of a 10% lead is unlikely to mask a reality where both parties are neck and neck. 

Again, we should always remember that a poll is a snapshot not a prediction.  Things may change.

I am more confident when we can look at comparable constituencies together.  So with that in mind, I have prepared a list of the seats by region:

https://drive.google.com/file/d/0Bygi8eZw-4q1MWZMMXpFOFBrLTA/view?usp=sharing

In the north west, the Conservatives are apparently outperforming in both Morecambe & Lunesdale and Blackpool North & Cleveleys, given the size of their majorities in those seats.  The fact that we have two Lancashire coastal constituencies telling the same story strengthens the credibility of both polls.  We don't have a poll yet for South Ribble, but might that outperformance extend down the coast?

In the east midlands, punters seem not to believe that Lincoln or Amber Valley are quite close, despite Lord Ashcroft's polls suggesting that the Conservatives are only a bit adrift, apparently relying more on the small swing required to take them.  If you do take the view that the Conservatives might revive a bit against Labour, both of these seats are worth considering for a bet on the Conservatives.

There are substantial swings to Labour in every London Conservative/Labour marginal so far polled except Harrow East.  That makes that particular poll a little suspect in my eyes (particularly since the pattern of a strong Labour performance is also reflected in polls in Hampstead & Kilburn and various Lib Dem/Labour marginals).  At present, neither Finchley & Golders Green nor Enfield Southgate are seen as disproportionately likely to fall to Labour.  If the trend is replicated in those constituencies once polled, the odds on Labour are likely to be cut sharply.  I'm already on Labour in Finchley & Golders Green from last year and I've now invested a small sum in Labour in Enfield Southgate too.

In the west midlands, the Conservatives seem to be bearing up fairly well in general.  The Wolverhampton South West poll will disappoint them, and the poor odds on the blue team in Dudley South reflects an extraordinary poll from Lord Ashcroft in Dudley North which shows UKIP a whisker behind Labour.  But the odds on the Conservatives holding Stourbridge, which is adjacent to Dudley South, have so far been completely unaffected by this deduction.  This may be a seat worth a few quid on Labour too.  But take your own view of the matter - Halesowen & Rowley Regis is adjacent to Stourbridge and the results of Lord Ashcroft's poll in that constituency were pretty good from a Conservative perspective, given the size of their majority.

There have been four polls in the eastern region and all four have been fairly poor for the Conservatives.  However, this does not seem to have been noticed.  Perhaps the betting public assume that East Anglia is predestined to be blue.  None of the prices on the Conservatives look attractive and the evens on Labour in Stevenage still looks decent to me, given that Labour recorded a lead of 5% in a recent poll.

Mysterious seats

Even after taking all these points into consideration, some of the large movements look odd. Why are the Conservatives so poorly thought of in Portsmouth North?  Why are Labour thought so likely to win Bristol North West from third?  Why is Rugby seen as so safe for the Tories?  None of the information available seems strong enough to justify large deviations from the norm.  Betting against these anomalies seems sound.

Summary

Punters do seem to be taking a sunnier view of the Conservatives' chances on the betting markets than the current information would suggest if no further polling movements take place, apparently on the assumption that in the run-up to the election, the polls will swing back in the direction of the Tories.  The movement being anticipated is not all that large on the constituency markets, so far as one can judge from the constituency polls so far conducted.  The movement expected on the general "most seats" markets is larger, and must be correspondingly less likely.

Some of the seats seem to be irrationally regarded as safer for the Conservatives than the information available suggests, and some the reverse.  There remain betting opportunities, even in the simpler problems posed by a straight fight between the two main parties.

Reading the entrails: a few polling observations

So far, I have made only limited attempts to try to work out how the polls might move in the run-up to the general election.  That is deliberate.  I am highly sceptical that anyone can foretell the future with any great confidence.  Things happen.

But we do have a lot of polling data, and while I absolutely agree with the pollsters like Lord Ashcroft who repeatedly advise us that an opinion poll is a snapshot and not a prediction, buried not all that deep in the data we have information that may help us predict at least some of the movements in opinion polls.

Understanding the present

Before we can work out what might happen in the future, we need to understand what is happening right now.  And unfortunately that's far from clear.  We have eight pollsters who regularly take national opinion polls: ICM, Ipsos-MORI, YouGov, Populus, Opinium, Survation, ComRes and Lord Ashcroft.  They differ quite markedly.  Are Labour on 36%, as in the most recent poll from Populus, or 29%, as in Ipsos-MORI's most recent poll?  Are the Conservatives on 35%, as Populus found today, or 29% as Opinium found in their most recent poll?  The Lib Dems tallied 6% with Survation and 11% with ICM in their most recent polls.  And UKIP recorded 11% with Populus this morning and 23% in the last Survation poll.  These are not minor differences.

How can you tell which poll is most accurate?  You can't definitively, unfortunately.  All we have to go on is past results.  Past performance is not necessarily a guide to the future and the value of pollsters can go down as well as up.

Pondering the future

But perhaps we can understand a little better the differences between the various pollsters and identify as a result how polls might move in the future.  Others are far more knowledgeable than me about the intricacies of past vote weighting, the use of certainty to vote, the differences between telephone polls and online polls, and so on.  Different pollsters use different methods to get a balanced sample.  These explain the differences in present results.

I want to look at two aspects that may affect how polls move in the future: how parties are prompted for in opinion polls and how voters adjust their thinking in their specific constituency.  We have illuminating information about both of these.

Party prompting

It would be natural to think that all pollsters would ask the same question when trying to get their headline polling figures.  But they don't.  Six of the pollsters follow essentially the same format.  Survation and ICM each have their own house style.

UK Polling Report have a handy explanation of the standard approach here:

http://ukpollingreport.co.uk/faq-wording

The bit I want to look at now concerns "the issue of whether or not the question prompts by party name. While in the questions above only three of the pollsters include party names, in fact they all prompt by party name. Since Yougov survey’s are mostly selecting onscreen options, obviously the party names are there for people to tick. MORI would have done the same when they used to do their surveys face-to-face (interviewers use laptops to display the questions and record answers), I can only assume they now read out the party names afterwards.

Historically prompting by party name was a very significant difference between pollsters, as it gives a significant boost to the Liberal Democrats. A poll that includes the Lib Dems in the prompt tends to show their support about 2 points higher than one which doesn’t. These days that is no longer an issue as everyone does it, but it is still a consideration for minor parties like the Greens and UKIP. These are not normally prompted for, and prompting for them gives them significantly more support. The problem is, when they are compared to election results, polls that prompt for minor parties invariably overestimate their support. So while it seems unfair, the most accurate way of polling seems to be to prompt using the names of the three main parties (plus the SNP and Plaid Cymru in Scotland and Wales), but ignoring minor parties."

You will note that the standard approach excludes UKIP at the first stage.  This is a point not lost on ardent kippers.  Does it remain the case that the most accurate way of polling is to prompt without mentioning UKIP (or indeed the Greens)?

Uniquely at a national level, Survation's prompts include UKIP.  Its methodology is explained here:

http://survation.com/methodology/

Survation also record the highest level of support for UKIP.  Is this cause and effect?

It is important to appreciate that the other aspects of the composition of opinion polls also affect the party shares.  Opinium do not prompt for UKIP but consistently record high ratings for UKIP.  There are other things in the mix.

But we have recently had an experiment which shows the potential impact of prompting for UKIP.  ComRes conducted an opinion poll on its standard basis recording UKIP at 19%.  It simultaneously conducted another where UKIP was prompted for, which increased UKIP's rating to 24%, a jump of 5% (almost entirely among men, interestingly):

http://ukgeneralelection2015.blogspot.co.uk/2014/10/comres-prompted-v-not-prompted-for-ukip.html

Now it is important to appreciate that you can't just add 5% onto the polling of every other pollster to get a "with prompting" figure for UKIP.  Polling methodologies are fully interlocking and no doubt those pollsters that do not prompt for UKIP would claim to account for UKIP through different means.  But it is an illustration of how prompting for UKIP could make a substantial difference.

We must envisage the possibility of a point being reached where UKIP are sufficiently established that to fail to prompt for them would risk under-recording their ultimate vote share.  We may already have reached that point.  If so, we should expect UKIP to do better than they otherwise might be expected to do from the results of most pollsters. 

One observation at this point: UKIP draw a disproportionate amount of their support from 2010 non-voters.  I suggest that if such supporters need prompting before remembering to support UKIP, their support needs to be downrated to a considerable degree.  The likelihood of them getting to a polling booth to have the opportunity to be pleasantly surprised by a purple option is substantially diminished.

Local constituency dynamics

Seven of the eight pollsters who regularly conduct national polls do not direct the respondent to consider their own constituency.  The eighth is ICM, which asks the following question:

"The Conservatives, Labour, the Liberal Democrats and other parties would fight a new election in your area. If there were a general election tomorrow which party do you think you would vote for?"

So as you can see, ICM do not prompt for UKIP, but uniquely prompt respondents to think about their constituency.

What difference might that make?

Lord Ashcroft's constituency polling

We now have quite a large body of constituency opinion poll evidence, thanks primarily to Lord Ashcroft. 
He has conducted 102 constituency polls, with more no doubt to come.  One aspect of that polling is potentially very helpful in understanding how voters will react when they focus on the question of constituency polling. 

When he conducts national opinion polls, Lord Ashcroft does not prompt for UKIP.  But when he conducts constituency polls in England, he does so asking two questions in succession, as follows:

"1. If there were a general election tomorrow, which party would you vote for? Would it be Conservative, Labour, Liberal Democrat, UKIP, or another party? If 'another party’: Would that be, the Green Party, the British National Party (BNP), or some other party – or do you not know how you would vote?

2. And thinking specifically about your own constituency and the candidates who are likely to stand there, which party's candidate do you think you will vote for in your own constituency at the next general election? [Prompts as at Q1]"

As you can see, he actively prompts for UKIP, and then gets the respondent to think about the specifics of the constituency.  We can therefore see, through a mechanism that kippers will be happy with, how respondents factor in the dynamics of their own constituency.

I have therefore looked at these polls and analysed how each party performs in each constituency, tallying up the difference between their poll ratings between the two questions:

https://drive.google.com/file/d/0Bygi8eZw-4q1Z2l1Z3NzcW54NjlYakVPTV81NW5sVy12MFZv/view?usp=sharing

Note, I am not taking a view at present on the absolute accuracy of Lord Ashcroft's constituency polls (I am wary of constituency polls), but rather I am relying on their ability to measure movements between generic questions about voting and considering constituency-specific considerations.  I see no reason to doubt the aggregate picture that these constituency polls demonstrate.

Labour and the Conservatives

In the Conservatives and Labour battlegrounds, the effect is relatively minor.  While there is a lot of variation by constituency, overall the Conservatives put on an average of just over 1/2 % in constituencies that they hold when voters are asked to think about the local considerations.  However, they are outgunned by Labour, who put on over 1% on average in such constituencies.  (I'm aware, incidentally, that some constituencies have been polled twice and that there is the theoretical risk that they would unduly skew the sample, but in practice they don't particularly).  Labour look as though they will get a small lift in most of the national polls when voters in Conservative/Labour marginals start thinking about their own constituencies.

The effect is more marked and still more in Labour's favour in constituencies that Labour hold, with Labour on average advancing by 1% and the Conservatives falling back by closer to 2% on average.  This will make the chances of many Conservative gains from Labour still harder than the polls currently suggest.

The Lib Dems

These movements pale into insignificance when one looks at the movements in constituencies where the Lib Dems are in contention.  In Lib Dem seats with a Conservative challenger, the Conservatives drop by over 4% on average when respondents are asked to think about their own constituency and the Lib Dems rise by nearly 12% on average.  In some seats, the movements are much greater: in Eastbourne, the Lib Dems rose 22% on the second question and in Sutton & Cheam the rise was 19%. 

Lib Dem seats with a Labour challenger show nearly as large movements: Labour drop by 3% on average while the Lib Dems rise by 8% on average. 

It should be noted that even such large movements in the Lib Dems' favour will not by themselves prevent the Lib Dems from having a torrid time.  Their poll ratings are so low that if Lord Ashcroft is in the right ballpark even such large movements in their favour on the constituency question would leave them with substantial projected losses.

As I have already noted, seven out of eight national pollsters do not direct respondents to think about their constituency when getting their view on how they would vote.  If those polling movements are replicated across just those seats that the Lib Dems hold, their poll ratings would rise by 1% nationally without any other movements in public opinion when voters come to consider their local options.  

But in fact, the Lib Dems do better not just in constituencies where they are in contention but also in constituencies where they are making up the numbers.  They put 1/2% on average in Conservative held constituencies with Labour challengers and over 1% on average in Labour held constituencies with Conservative challengers.  It seems that when voters are asked to think about their own constituencies, the Lib Dems benefit.  They appear to be the local party for local people.

All of this goes a long way towards explaining why the Lib Dems poll particularly well with ICM in the periods between general elections.  It also suggests that ICM's methodology on this point at least is superior to their rivals, given that all voters will be thinking about their local constituency at some point.

What is unclear is whether the local effect is yet fully factored in.  Lord Ashcroft's polls do not name candidates (in some cases they are not yet known), so we don't know what impact name recognition might have.  It seems reasonable to assume that may help the Lib Dems further (particularly where they are the incumbents), given that focussing on local factors already gives them a boost.

UKIP

UKIP on the other hand suffer when things get local.  When polling respondents are asked to focus on their constituencies, UKIP lose votes in every category of marginal constituency.  Overall, they average losses of 2% per constituency pretty much evenly across the board.  UKIP gained support in only six out of 102 constituency polls between the first and second question, and in each case by just one percentage point.  This looks most unlikely to be a coincidence.

UKIP presumably benefit under the first question from the prompting for them.  But when respondents consider their own constituency, UKIP's star appreciably dims.  It appears from the data tables to be a general dimming rather than specific to either sex, any age group or any class.

If we put together the ComRes experimental poll with this finding, it appears that prompting for UKIP encourages more men to select UKIP, but prompting respondents to think about their constituencies leads to a drop in UKIP support across all groups.  It isn't as simple as a single group of voters first being encouraged and then being discouraged.  It appears that different groups wash into and out of supporting UKIP by these two mechanisms.

We know that respondents will ultimately start thinking about their constituencies before the election, because they will have no choice.  What we don't yet know is whether the general election process will do the equivalent of prompting voters to think about supporting UKIP.

Conclusions

We shall see further polling movements caused by events, and possibly by voters starting to concentrate more seriously on the coming general election.  But we can infer from the above that we may see some polling movements caused solely by latent biases in the polls.  The Lib Dems are likely to rise with most pollsters at least towards the levels currently recorded by ICM because of their strength at local level.  Labour may also see their vote share bolstered by local strength in Labour/Conservative marginals.

UKIP, as so often, remain the big imponderable, with two competing forces alternately potentially boosting and dampening their chances.  My guess - and it is no more than a guess - is that UKIP will remain in the public eye right the way to polling day, having the effect of prompting voters to think of them.  If so, UKIP may have a small net boost to their national polling figures (independent of other voting movements).  But their weakness at the local level is likely to make their gains harder to make.