UK General Election - May 7th 2015

[Clyde1998]

My mistake. I had thought YouGov was systemically more positive than others.

YouGov were the pollster I criticised the most throughout the entire campaign, as I felt they were under representing support for 'yes' - until the late surge near the end of the campaign [mainly as they had never predicted a Scottish election correctly with their final poll]. In hindsight, their polls seemed to be pretty accurate...

[YL]

Ashcroft is finally going to release his Scottish constituency polls tomorrow, and is ramping them on Twitter.

FWIW (possibly not very much) in the YouGov poll the SNP voters are noticeably more confident that they will actually vote that way than the unionist parties' voters.  (The Lib Dem sample is tiny, of course, so is even less informative than the others.)

[Clyde1998]

Seeing that the Ashcroft polls are suggesting that Margaret Curran could lose her Glasgow East seat - which would be at least a 19% swing. The record swing in a general election in the UK is 21%.

EDIT: Just seen this on Twitter - don't know if it's real or not...

[Clyde1998]

You'd think that a billionaire would be able to protect opinion polls, but obviously not. Anyone is easily able to access the poll results.

Nearly every poll shows swings of over 15% to the SNP, with most showing swings of over 20% - some over 25%.

[Hifly]

You'd think that a billionaire would be able to protect opinion polls, but obviously not. Anyone is easily able to access the poll results.

Nearly every poll shows swings of over 15% to the SNP, with most showing swings of over 20% - some over 25%.

Can you post the link?

[Boston Bread]

Seeing that the Ashcroft polls are suggesting that Margaret Curran could lose her Glasgow East seat - which would be at least a 19% swing. The record swing in a general election in the UK is 21%.

Really? For a record swing that is spectacularly low.
http://en.wikipedia.org/wiki/St._John's_East
2006: Conservative beats Liberal by 5
2008: NDP beats Liberal by 62 (Conservatives in third)

[Phony Moderate]

Seeing that the Ashcroft polls are suggesting that Margaret Curran could lose her Glasgow East seat - which would be at least a 19% swing. The record swing in a general election in the UK is 21%.

Really? For a record swing that is spectacularly low.
http://en.wikipedia.org/wiki/St._John's_East
2006: Conservative beats Liberal by 5
2008: NDP beats Liberal by 62 (Conservatives in third)

We are much less fickle than you guys.

[Clyde1998]

You'd think that a billionaire would be able to protect opinion polls, but obviously not. Anyone is easily able to access the poll results.

Nearly every poll shows swings of over 15% to the SNP, with most showing swings of over 20% - some over 25%.

Can you post the link?

http://lordashcroftpolls.com/wp-content/uploads/2015/02/

The polls appear as images - which will show you the headline polls and the swing. The data tables are released tomorrow.

Ashcroft ask two voting intention questions - one asking how they would vote and a second asking how they plan to vote after thinking about the candidates in their consistency. It'll be very interesting to see how the two question results line up.

[Clyde1998]

Seeing that the Ashcroft polls are suggesting that Margaret Curran could lose her Glasgow East seat - which would be at least a 19% swing. The record swing in a general election in the UK is 21%.

Really? For a record swing that is spectacularly low.
http://en.wikipedia.org/wiki/St._John's_East
2006: Conservative beats Liberal by 5
2008: NDP beats Liberal by 62 (Conservatives in third)

I've found a bigger swing than 21% - it seems the list I was using was a little out of date:

https://en.wikipedia.org/wiki/Blaenau_Gwent_(UK_Parliament_constituency)#Elections_in_the_2010s

29.2% in 2010. You can argue that in 2005 - there's a 49% swing (but the independent didn't run the previous election).

Swings over 15% are very, very rare though in the UK - especially in Scotland (where the large swings this year are likely to happen).

[afleitch]

If correct it's worth noting that Ashcroft is predicting that the SNP will win in areas they didn't win in 2011 (Glasgow, Lanarkshire etc)

[Harry Hayfield]

Here are the biggest swings at every general election in the UK since 1987 where the swing was between two parties who contested the previous election (and excludes by-elections)

1987: Western Isles (Lab GAIN from SNP on a swing of 19.33% from SNP to Lab)
1992: Edinburgh, Leith (Lab HOLD on a swing of 13.76% from Lab to SNP)
1997: Brent North (Lab GAIN from Con on a swing of 18.83% from Con to Lab)
2001: Kingston and Surbiton (Lib Dem HOLD on a swing of 15.92% from Con to Lib Dem)
2005: Dunbartonshire East (Lib Dem GAIN from Lab on a swing of 20.62% from Lab to Lib Dem)
2010: Blaenau Gwent (Lab GAIN from Ind on a swing of 29.17% from Ind to Lab)

[You kip if you want to...]

The biggest swing in the UK this time will surely be a Scottish seat, but where do we think for an English/Welsh seat? Surely a LibDem one...

[Clyde1998]

Here are the biggest swings at every general election in the UK since 1987 where the swing was between two parties who contested the previous election (and excludes by-elections)

1987: Western Isles (Lab GAIN from SNP on a swing of 19.33% from SNP to Lab)
1992: Edinburgh, Leith (Lab HOLD on a swing of 13.76% from Lab to SNP)
1997: Brent North (Lab GAIN from Con on a swing of 18.83% from Con to Lab)
2001: Kingston and Surbiton (Lib Dem HOLD on a swing of 15.92% from Con to Lib Dem)
2005: Dunbartonshire East (Lib Dem GAIN from Lab on a swing of 20.62% from Lab to Lib Dem)
2010: Blaenau Gwent (Lab GAIN from Ind on a swing of 29.17% from Ind to Lab)

So, we could see the largest swing for Scotland and there's an outside chance of the largest UK swing. Could be in quite a few seats too...

[You kip if you want to...]

Here are the biggest swings at every general election in the UK since 1987 where the swing was between two parties who contested the previous election (and excludes by-elections)

1987: Western Isles (Lab GAIN from SNP on a swing of 19.33% from SNP to Lab)
1992: Edinburgh, Leith (Lab HOLD on a swing of 13.76% from Lab to SNP)
1997: Brent North (Lab GAIN from Con on a swing of 18.83% from Con to Lab)
2001: Kingston and Surbiton (Lib Dem HOLD on a swing of 15.92% from Con to Lib Dem)
2005: Dunbartonshire East (Lib Dem GAIN from Lab on a swing of 20.62% from Lab to Lib Dem)
2010: Blaenau Gwent (Lab GAIN from Ind on a swing of 29.17% from Ind to Lab)

So, we could see the largest swing for Scotland and there's an outside chance of the largest UK swing. Could be in quite a few seats too...

With a smirk, Hallam.

[Filuwaúrdjan]

The record swing in a general election in the UK is 21%.

Where does that figure come from? If we're using swing in the sense the media does, then Southwark & Bermondsey in 1983 would have been over 30pts, for instance. If you're talking in a strict psephological hard-ass sense (which would make no sense given the context, but whatever) then it only applies to the Labour/Tory swing (and then the postwar record it is about 21pts; Merthyr in 1970).

[Clyde1998]

The record swing in a general election in the UK is 21%.

Where does that figure come from? If we're using swing in the sense the media does, then Southwark & Bermondsey in 1983 would have been over 30pts, for instance. If you're talking in a strict psephological hard-ass sense (which would make no sense given the context, but whatever) then it only applies to the Labour/Tory swing (and then the postwar record it is about 21pts; Merthyr in 1970).

(Change in party 1 vote + change in party 2 vote) / 2

I have a document that had a list of largest swings in each election since the 70s (I think) - I'll check it tomorrow.

[Phony Moderate]

The record swing in a general election in the UK is 21%.

Where does that figure come from? If we're using swing in the sense the media does, then Southwark & Bermondsey in 1983 would have been over 30pts, for instance. If you're talking in a strict psephological hard-ass sense (which would make no sense given the context, but whatever) then it only applies to the Labour/Tory swing (and then the postwar record it is about 21pts; Merthyr in 1970).

(Change in party 1 vote + change in party 2 vote) / 2

Anyone who knows the word 'psephological' is going to know how swing is measured, I imagine.

[Clyde1998]

The record swing in a general election in the UK is 21%.

Where does that figure come from? If we're using swing in the sense the media does, then Southwark & Bermondsey in 1983 would have been over 30pts, for instance. If you're talking in a strict psephological hard-ass sense (which would make no sense given the context, but whatever) then it only applies to the Labour/Tory swing (and then the postwar record it is about 21pts; Merthyr in 1970).

(Change in party 1 vote + change in party 2 vote) / 2

Anyone who knows the word 'psephological' is going to know how swing is measured, I imagine.

Just for good measure

[YL]

If correct it's worth noting that Ashcroft is predicting that the SNP will win in areas they didn't win in 2011 (Glasgow, Lanarkshire etc)

Given the referendum results and the polls I don't find this surprising.  In fact, Ashcroft has polled mainly "Yes" areas: his 16 seats include Dundee West, all seven Glasgow seats, West Dunbartonshire, and the four North Lanarkshire seats [1].  The SNP are ahead in all of these except Glasgow NE.  It would have been more interesting if he'd polled some Labour-held "No" areas too.

He also polled two Lib Dem held seats, Salmond's Gordon, where Salmond is well ahead but not by quite as much as I thought he might be, and Inverness et al, where, unless the poll is totally wrong, Danny Alexander is toast.

[1] Coatbridge, Chryston & Bellshill; Airdrie & Shotts; Motherwell & Wishaw; Cumbernauld, Kilsyth & Kirkintilloch East.  (What happened to Kirkintilloch West?)

[YL]

Here are the figures for the parties over 10% in each seat.  These are for the first question ("standard" voting intention); the constituency question boosts the LDs by 5 points each in their two held seats but doesn't make much difference in most cases.

Airdrie & Shotts: SNP 46 Lab 38
Coatbridge et al: SNP 46 Lab 41
Cumbernauld et al: SNP 48 Lab 36
Dundee W: SNP 58 Lab 25
Glasgow C: SNP 46 Lab 36
Glasgow E: SNP 46 Lab 39
Glasgow N: SNP 43 Lab 33 Green 10
Glasgow NE: Lab 46 SNP 39
Glasgow NW: SNP 43 Lab 38
Glasgow S: SNP 46 Lab 33 Con 10
Glasgow SW: SNP 42 Lab 41
Gordon: SNP 41 LD 21 Lab 17 Con 15
Inverness et al: SNP 48 LD 16 Con 14 Lab 14
Motherwell & Wishaw: SNP 48 Lab 38
Paisley & Renfrewshire S: SNP 45 Lab 39
West Dunbartonshire: SNP 45 Lab 36

[Joe Republic]

What is the likelihood that, if the numbers add up sufficiently, Labour and SNP form a coalition?

[afleitch]

What is the likelihood that, if the numbers add up sufficiently, Labour and SNP form a coalition?

It might be the only viable option in terms of arithmetic if the current polls and seat 'predictions' are correct. I don't think Labour wants it though. They have an election to fight in Scotland in 2016.

As for Ashcroft, first off 'lol Airdrie'

Anyway, I think Ashcroft has another wave coming but the change in the share of the vote within all these seats based on 2010 is Conservatives down 2.5, Labour down 14.9, Lib Dems down 13.6%, SNP up 26.4% and others up 3.6%. That would, nationally give the following:

SNP 46.3
LAB 27.1
CON 14.2
LDEM 5.3
OTH 7.1

So not out of line with the polls, but it does suggest that the swing to the SNP is high enough to actually oust Labour from the few seats they would mathematically hold under a swing of that degree. Can they in turn hold on in other seats?

We haven’t had results from the more interesting competitions actually and nothing from outside Lanarkshire except those lone polls for Gordon, Inverness and Dundee West. Gordon is interesting, but only on the basis that it’s the only discernable seat where Labour and the Tories drop significantly and the Lib Dems rise as a result on the constituency question, but the SNP also go up a little too. Inverness is terrible for the Lib Dems and is one in fact where I thought they would do alright.

Nothing from Ochil, Falkirk, Dumfries etc which will probably be the more interesting results in terms of what the effect is on Labour.

[jaichind]

Would not polls like this from Scotland cause LD and Labor voters to tactically vote to defeat SNP and keep the number of seats won by the SNP lower than what the polls imply ?

[Phony Moderate]

Would not polls like this from Scotland cause LD and Labor voters to tactically vote to defeat SNP and keep the number of seats won by the SNP lower than what the polls imply ?

According to the details of the poll, only 33% of voters in these seats said they would never vote SNP, compared to Labour in the 40s and the Lib Dems in the 70s.

[You kip if you want to...]

Why has a turnaround like this happened? I just don't get it.