UK Post-Election Analysis

[afleitch]

For bits and pieces.

First off: how one pollster almost got it right

http://survation.com/snatching-defeat-from-the-jaws-of-victory/

'As in the Scottish Independence Referendum, where Survation shared “joint honours” with Ipsos Mori for accuracy, Survation conducted a voting intention telephone poll the day before the election (Wednesday) with three specific attributes:

Naming candidates through a ballot prompt specific to the respondents’ constituency based on their postcode.

Carefully balancing our sample based on age, region, sex, and past vote prior to weighting, from a nationally representative sample frame

And crucially, Speaking only to the named person from the dataset and calling mobile and landline telephone numbers to maximise the “reach” of the poll.

This was conducted over the afternoon and evening of Wednesday 6th May, as close as possible to the election to capture any late “swing” to any party – the same method we used in our telephone polls during the Independence Referendum that produced a 54% and a 53% figure for “no”.

This poll produced figures of:


CON 37%
LAB 31%
LD 10
UKIP 11
GRE 5
Others (including the SNP) 6%


We had flagged that we were conducting this poll to the Daily Mirror as something we might share as an interesting check on our online vs our telephone methodology, but the results seemed so “out of line” with all the polling conducted by ourselves and our peers – what poll commentators would term an “outlier” – that I “chickened out” of publishing the figures – something I’m sure I’ll always regret.'

[Phony Moderate]

In terms of published polls, Opinium, ComRes and Ipsos-MORI were technically joint 'best', which says a lot in itself.

[Silent Hunter]

The poll blow-up probably influenced Labour strategy; they thought they were more popular than they were.

[Lief 🗽]

Ah yes, now begins the "but our secret, never-released numbers with questionable methodology were right, we were actually super accurate!" phase.

It is one thing to publish a poll that ends up being inaccurate. It is totally another to fudge your methodology and your numbers at the last minute so that your numbers match the consensus and you can all sink or swim together. Bunch of cowards and frauds, the entire polling industry.

[Simfan34]

The senior ministers are all returned!

[Torie]

i am surprised that the parties don't have competent internal polling, which is typically better than the published stuff we see, as in the US.

[ChrisDR68]

Quite a shocking general election overall really.

The SNP now owns Scotland which is a major strategic problem for Labour as fear of SNP influence over a minority Labour government without doubt helped the Conservatives. That will continue to be a problem for Labour moving forward.

Labour support is still solid in the north, south Wales and London. In the midlands and the south of England outside of London the party's appeal is desperately weak. Only the very singular personality of Tony Blair has been anything approaching popular in this part of the UK over the past 40 years. Another strategic issue for the party that isn't going away anytime soon.

The crushing of the Liberal Democrats will take decades for them to recover from. Weirdly blamed for the policies the Conservatives pushed through during the coalition while given little to no credit for the policies they championed and help implement. Clegg was right imo when he said the election results were crushing and unkind to his party.

It looks to me that the break up of the UK is probably inevitable after this election. Possibly within the next 10 years.

[Gallium]

This is a good first analysis of why the polls were so wrong.

Regarding the Survation Poll:

You must be logged in to read this quote.

For all Atlas jokes about junk polls, at least there's several American polling firms that are willing to buck consensus and post unusual results. It's a sad indictment on British polling that companies were too concerned about potentially embarrassing themselves to question orthodoxy.

[Lief 🗽]

Yet another ruinous consequence of Nate Silver's rise to fame. The sooner that man is thrown into the dustbin of history, the better.

[Хahar 🤔]

Reposting this because it seems appropriate:

It seemed like the issue that hurt the Liberal Democrats most was tuition fees, in the sense that that was where they were perceived to have sold out. The raise in tuition fees was passed in December 2010 with the support of a majority of LD MPs. The questions I have:

1. Could the Liberal Democrats have stopped the government from implementing this plan?
2. If the answer to the previous question was no and the Liberal Democrats had withdrawn their support for the government over tuition fees, how would they have done in the following election?

The crushing of the Liberal Democrats will take decades for them to recover from. Weirdly blamed for the policies the Conservatives pushed through during the coalition while given little to no credit for the policies they championed and help implement. Clegg was right imo when he said the election results were crushing and unkind to his party.

It seemed to me that if a voter supported the government, then they would vote Conservative, since a Tory victory would definitely lead to a continuation of the government, whereas if they opposed the government, they would vote for some other party. There was no compelling reason to vote Liberal Democrat.

[Lurker]

Yet another ruinous consequence of Nate Silver's rise to fame. The sooner that man is thrown into the dustbin of history, the better.

While Nate Silver is overrated, he shouldn't be blamed for the British polling industry being utterly terrible at its job, tbf.

[Torie]

Yet another ruinous consequence of Nate Silver's rise to fame. The sooner that man is thrown into the dustbin of history, the better.

As Silver himself would say, his work is only as good as the poll numbers upon which it depends. He's a number cruncher, not a savant.

[Lief 🗽]

Sure, but I think it's undeniable that he's had a negative influence on the polling industry (mostly in the US yes, but we are living in a globalized and inter-connected world, and, like Big Brother or Masterchef, Nate Silver imitators have popped up in droves from country to country). "Herding" was much less of a problem before he poked his nose into things.

[Thomas D]

There are going to be TWO Iowa caucuses between now and the next election.

 That feels like such a long time.

[Torie]

Sure, but I think it's undeniable that he's had a negative influence on the polling industry (mostly in the US yes, but we are living in a globalized and inter-connected world, and, like Big Brother or Masterchef, Nate Silver imitators have popped up in droves from country to country). "Herding" was much less of a problem before he poked his nose into things.

The pollsters are fixing their numbers because they are afraid of Nate Silver?   Really?  Man, the guy is powerful.

[Lief 🗽]

Sure, but I think it's undeniable that he's had a negative influence on the polling industry (mostly in the US yes, but we are living in a globalized and inter-connected world, and, like Big Brother or Masterchef, Nate Silver imitators have popped up in droves from country to country). "Herding" was much less of a problem before he poked his nose into things.

The pollsters are fixing their numbers because they are afraid of Nate Silver?   Really?  Man, the guy is powerful.

I'm sure the polling firms that Nate Silver has criticized have lost a considerable amount of business as a result. Like it or not, he's by far the most respected and well-known name when it comes to this sort of thing. Why risk that when it's so much easier to nudge your numbers towards realclearpolitics.com's numbers?

[Torie]

Sure, but I think it's undeniable that he's had a negative influence on the polling industry (mostly in the US yes, but we are living in a globalized and inter-connected world, and, like Big Brother or Masterchef, Nate Silver imitators have popped up in droves from country to country). "Herding" was much less of a problem before he poked his nose into things.

The pollsters are fixing their numbers because they are afraid of Nate Silver?   Really?  Man, the guy is powerful.

I'm sure the polling firms that Nate Silver has criticized have lost a considerable amount of business as a result. Like it or not, he's by far the most respected and well-known name when it comes to this sort of thing. Why risk that when it's so much easier to nudge your numbers towards realclearpolitics.com's numbers?

That supposition makes about as much sense to me, as your suppositions as to what goes on in The Cave.

[The Mikado]

There are going to be TWO Iowa caucuses between now and the next election.

 That feels like such a long time.

There'll be an EU Brexit referendum to tide us over

Maybe in 2018 or 2019 another Scotland referendum, too, depending on how much Cameron alienates Holyrood.

[Lumine]

I suppose Osborne is "Deputy Prime Minister" in all but name now, right? (I wonder if Hague would have gained that title had he run for another term)

[King of Kensington]

I'm surprised how many people here have defended the Lib Dems forming a coalition with the Tories (provided stability when the country needed it, etc.)  Obviously voters didn't agree.

[afleitch]

The verdict on the constituency polls is that when taken together (as the 538 model did), they were more accurate if you simply used the 1st question (who would you vote for) rather than the second (thinking about your constituency)

The Tories were still underestimated and that's because the polls were wrong.

[Clyde1998]

There are going to be TWO Iowa caucuses between now and the next election.

 That feels like such a long time.

There'll be an EU Brexit referendum to tide us over

Maybe in 2018 or 2019 another Scotland referendum, too, depending on how much Cameron alienates Holyrood.

Scottish, Welsh and Northern Irish elections as well next year.

[jaichind]

In a what if, I wonder what would have happen if in the second Salmond versus Darling debate last year had Darling creaming Salmond instead of the other way around.  This is because most likely the polls would not have turned which triggered this gang up by CON LAB and LD on SNP with promises that was viewed as not kept.  If Scotland Independence vote lost by a mile without massive intervention by the Westminster system then most likely SNP would have made gains in 2015 but not massive ones.  This in turn would mean that the English fear of LAB-SNP government would not have given CON edge.  I wonder how would the election turn out in England/Wales in such a case.  I figure in such a scenario LAB would have been the biggest party although not a majority as it would have avoided/minimized losses in both Scotland and England/Wales.

[ChrisDR68]

Looking at this general election with a little perspective it's clear to me that voting UKIP is in effect voting Conservative in all practical senses.

The voters who switched from Lib Dem to UKIP were decisive in sweeping away all but 2 Lib Dems MP's in southern England which gave the Tories their majority in this election.

If the Lib Dems are to make a comeback in the near future they need to drum into UKIP voters that they're voting Conservative by proxy. Some may be happy with that but I suspect a good many will not like to hear it.

They'll have to say it again and again and again until the message hits home. It's classic and clever divide and rule tactics from the cynical Tories.

The other thing the Libs need to do is reconsider their all encompassing love affair with the EU. It's toxic for them with the public. Be moderately in favour of remaining in the EU but adamantly and stridently against the setting up of a federal European superstate and the UK's absorption within it.

[jaichind]

http://www.bbc.com/news/uk-politics-32696505

Farage stays as UKIP leader after resignation rejected

Very hypocritical of him to do this.  I actually like Farage a lot and if I were in the UK I would vote UKIP (although vote CON tactically if it makes a difference.)  But this stunt will cost him a lot of good will and make him out to be yet another politician which he claims he is not.