2015 General Election - a record breaker

[iain]

I have tried to collate a list of records broken in the 2015 general election. There may be some errors, and the list is probably incomplete.

Biggest Swing
39.3% - Glasgow NE
The top 11 are from 2015

Biggest vote increase:
43.9% - Glasgow North East
The top 11 are from 2015

Lowest winning vote share:
24.5% - Belfast South

Lowest major party vote share:
0.1% - Belfast West
The top 10 are from 2015 - 8 Tory NI candidates

Youngest elected MP:
20 years, 237 days - Mhairi Black

Biggest fall in position for incumbent (joint):
To 4th - Simon Wright

Gain from lowest position (joint):
From 4th - SNP in 10

Highest number of candidates (excluding PM's constituency):
13 - Uxbridge & South Ruislip





Biggest LD vote fall:
35.8% - Brent Central (second biggest ever)

Biggest Labour vote increase:
26.9% - Birmingham Hall Green

Lowest Labour winning vote share:
31.1% - Ynys Mon

Biggest SNP vote increase:
43.9% - Glasgow North East

Biggest Green vote:
41.8% - Brighton Pavilion

Biggest Green vote increase:
23.0% - Bristol West

Biggest UKIP vote:
44.4% - Clacton

Biggest UKIP vote increase:
29.6% - Heywood & Middleton

[Davıd Boothroyd]

No, you're wrong. The biggest swing in 2015 was 17.0% in Glasgow North East, and it did not break the record which was 20.8% in Merthyr Tydfil in 1970.

[iain]

Jan 10, 2016 0:38:39 GMT Davıd Boothroyd said:

No, you're wrong. The biggest swing in 2015 was 17.0% in Glasgow North East, and it did not break the record which was 20.8% in Merthyr Tydfil in 1970.

This uses the modern, commonly used definition of swing, between the top two parties. I do not care about your semantics.

[Davıd Boothroyd]

Jan 10, 2016 0:50:43 GMT iain said:

Jan 10, 2016 0:38:39 GMT Davıd Boothroyd said:

No, you're wrong. The biggest swing in 2015 was 17.0% in Glasgow North East, and it did not break the record which was 20.8% in Merthyr Tydfil in 1970.

This uses the modern, commonly used definition of swing, between the top two parties. I do not care about your semantics.

It is not fucking semantics. It is a matter of fact that voters' shifts between major parties are far more reluctant and therefore more meaningful. A shift between Labour and the SNP, or between the Conservatives and UKIP, is simply not on the same level of comparison as one between Conservative and Labour. Any analysis which treats it as the same is utter bunk.

[Deleted]

Jan 10, 2016 0:56:42 GMT Davıd Boothroyd said:

Jan 10, 2016 0:50:43 GMT iain said:

This uses the modern, commonly used definition of swing, between the top two parties. I do not care about your semantics.

It is not fracking semantics. It is a matter of fact that voters' shifts between major parties are far more reluctant and therefore more meaningful. A shift between Labour and the SNP, or between the Conservatives and UKIP, is simply not on the same level of comparison as one between Conservative and Labour. Any analysis which treats it as the same is utter bunk.

Oh let it go. You've banged this drum so often it doesn't make any sound.

[iain]

Jan 10, 2016 0:56:42 GMT Davıd Boothroyd said:

Jan 10, 2016 0:50:43 GMT iain said:

This uses the modern, commonly used definition of swing, between the top two parties. I do not care about your semantics.

It is not fracking semantics. It is a matter of fact that voters' shifts between major parties are far more reluctant and therefore more meaningful. A shift between Labour and the SNP, or between the Conservatives and UKIP, is simply not on the same level of comparison as one between Conservative and Labour. Any analysis which treats it as the same is utter bunk.

Who said they were the same? I, along with most people, use swing as between the top 2 candidates, rather than the Butler swing.

I am sure that you will agree that the swing from Labour to Conservative in Glasgow North East is a not at all meaningful statistic. There may have been a 17% swing from Labour to Conservative, there has been no 'shift between major parties'. Personally I am much more interested in the swing between the SNP and Labour here than I am in the swing between Labour and the Conservatives.

People can make up their own mind. Poll added.

[maxque]

Jan 10, 2016 0:56:42 GMT Davıd Boothroyd said:

Jan 10, 2016 0:50:43 GMT iain said:

This uses the modern, commonly used definition of swing, between the top two parties. I do not care about your semantics.

It is not fracking semantics. It is a matter of fact that voters' shifts between major parties are far more reluctant and therefore more meaningful. A shift between Labour and the SNP, or between the Conservatives and UKIP, is simply not on the same level of comparison as one between Conservative and Labour. Any analysis which treats it as the same is utter bunk.

We are talking of Scotland here. Conservatives and Labour are not the two major parties there, far from it.

And trying to apply England/Wales political logic to Scotland makes as much sense then trying to apply a left/right logic to Ulster.

[carlton43]

There must now be logic for regarding swing in more than one respect? In Scotland swing is surely SNP and another party. In England it is Labour-Conservative. But with the uniformity of swing having largely disappeared as a constant feature may we not consider the swing to be quoted two ways for all constituencies? That is the figure Labour-Conservative and also the swing between winning party and previous winner/runner-up as well. In some cases a three-fold compilation of swing could be instructive, where a party placed 4th last time wins, but there is also large movement between two other parties other than Conservative and Labour. The DB position by itself is an absurdist position in the current situation because it is unhelpful, misleading and sometimes utterly pointless.

[Deleted]

Jan 10, 2016 0:56:42 GMT Davıd Boothroyd said:

Jan 10, 2016 0:50:43 GMT iain said:

This uses the modern, commonly used definition of swing, between the top two parties. I do not care about your semantics.

It is not fracking semantics. It is a matter of fact that voters' shifts between major parties are far more reluctant and therefore more meaningful. A shift between Labour and the SNP, or between the Conservatives and UKIP, is simply not on the same level of comparison as one between Conservative and Labour. Any analysis which treats it as the same is utter bunk.

Is stating there was a 17% swing to the Conservatives in Glasgow NE any more meaningful?

[froome]

The concept of swing should be consigned to the political dustbin.

There are always more than two parties contesting constituencies, and people's votes can move in all directions from one election to the next. So the idea that there has been a movement of x percent between two parties is meaningless.

Its use has the effect of reinforcing the idea that there are just two main parties contesting elections.

[Davıd Boothroyd]

Jan 10, 2016 9:47:49 GMT froome said:

Its use has the effect of reinforcing the idea that there are just two main parties contesting elections.

There are just two parties who can take the lead in forming a UK government. You might not like that but it's still the product of a collective decision of British voters.

[Davıd Boothroyd]

Jan 10, 2016 1:44:58 GMT doktorb🏳️‍🌈🏳️‍⚧️ said:

Jan 10, 2016 0:56:42 GMT Davıd Boothroyd said:

It is not fracking semantics. It is a matter of fact that voters' shifts between major parties are far more reluctant and therefore more meaningful. A shift between Labour and the SNP, or between the Conservatives and UKIP, is simply not on the same level of comparison as one between Conservative and Labour. Any analysis which treats it as the same is utter bunk.

Oh let it go. You've banged this drum so often it doesn't make any sound.

Your knowledge of psephology matches your knowledge of percussive acoustics.

[Crimson King]

what is swing for? Perhaps we should decide this before getting into the detail of how it should be calculated.

If we are using it as a predictive tool, to (foolishly imho) judge the likely outcome of a future general or local election on the basis of a single result, or (slightly less foolishly) judge with increasing accuracy the likely outcome of a general election as more individual results come in, then it is necessary to use the same sort of swing in all elections, presumably Lab/Con in a UK scenario (but lab SNP in a Scottish one) How good a predictive tool it is, given the multi-party nature of our system is another matter of course.

If we are using it as a comparative tool, to see which results have shown the greatest shift in allegiance (from whomever to whomever) then the most appropriate measure of swim will differ from seat to seat.

[johnloony]

It is rather hahaha to read this thread. It is an established historical fact that Butler and Boothroyd both always use the Conservative/Labour swing as the swing they are talking about. It is also an established historical fact that Everbody Else and Normal People always use the Top Party/Second Party swing as the swing they are talking about. Everything else is irrelevant.

It may well be the case that "There are just two parties who can take the lead in forming a UK government" (as David says) but that is only a minor consideration in choosing which "swing" one wished to use.

[carlton43]

I have decided to compile the swing for a range of constituencies (perhaps all of them) to get a better feeling on the usefulness or otherwise of the whole concept of swing in a multi-candidate and multi-party era.

I decided to start with a compilation of all constituencies by size of majority as a percentage of total vote and to see what was needed to be useful to me and what could be meaningful at all.

I soon came to the conclusion that Classic Swing should be quoted first in all cases except for Ulster (until such time as both Labour and Conservatives contest all seats there). That is the swing from Conservative to Labour (or the reverse) even in such constituencies as those where neither of the contending majors is even runner-up. As David contends this is the major and necessary trend for the whole national contest and it does have meaning and usefulness in every case. I have termed this Column One Swing. The first case of this being 'sufficient' is Constituency Number 4 in the list of seats, East Ham which has a 65.5% Majority, but where the main contest in 2015 and 2010 were both between Labour and Conservative, albeit with the Conservatives very well behind. Here in East Ham the Swing was 5.2% Conservative to Labour and nothing else need be added.

A seat where I feel we require additional comment is Constituency Number 6 Birmingham Ladywood which has a Majority of 60.9%, but where the runner-up has changed from LD to Conservative and so the comparison from 2010 to 2015 is between different parties. Here the Column One (Conservative to Labour) Swing is 8.6%. But the Column Three (Runner-Up changes to Conservative or Labour from a Non-Major previously) is material to record the large LD vote collapse, and that Column Three Swing is LD to Labour 20.9%.

But the most safe seat of all, Seat Number One, Liverpool Walton with an amazing Majority of 72.3%, has a change in Runner-Up from LD to UKIP and that necessitates the compilation of three swing figures to reflect the full position. The Column One Swing (Conservative to Labour) is 5.6%. The Column Two Swing (the new Non-Major Runner-Up to Labour) is 1.5% UKIP to Labour. And the Column Three Swing (the former Non-Major Runner-Up to Labour) is 10.6% LD to Labour.

Having done over 25 of the most safe constituencies it is apparent that so far there is a prevailing Column One Swing to Labour even in some very safe Conservative seats and in all Labour seats. That where the Non-Major Runner-Up has changed there is usually a notional swing to Labour from that new runner-up despite the improved position of such runner-up. And that the enormity of the LD collapse shows through in very high figures in many Column Three entries and a few Column Two entries where the LDs have hung on to runner-up position.

So far I think this is turning out to be a most informative exercise.

[carlton43]

Still attending to swing, I noticed that the Leader of the Conservative Party and PM has the 55th most safe seat (Witney) with a 43% Majority over Labour. He had 60.2% of the poll with Labour on 17.2%. Right next to him as 56th most safe seat (Islington North)is the Leader of the Labour Party, also with a 43% margin and also with a 60.2% of the poll, but with the Conservatives on exactly 17.2% as well. In each seat there was also exactly the same swing of 1.4% from Conservative to Labour. That is quite a lot of coincidence.

[piperdave]

Question for Davıd Boothroyd - based on your view of swing, who is the swing between at Scottish Parliament elections? Labour and Conservative because swing can only be between them based on Butler? Or Labour and SNP because they have always been the largest two parties at said elections (since 1999 for the avoidance of doubt)?

[Davıd Boothroyd]

There probably isn't much point calculating swing at the Scottish Parliament elections, because it doesn't represent how voters behave.

[carlton43]

Jan 13, 2016 0:51:58 GMT piperdave said:

Question for Davıd Boothroyd - based on your view of swing, who is the swing between at Scottish Parliament elections? Labour and Conservative because swing can only be between them based on Butler? Or Labour and SNP because they have always been the largest two parties at said elections (since 1999 for the avoidance of doubt)?

Well, you have David's answer already, but I intend to apply my 3-fold approach to the Scottish situation for HOC results, but I don't see why it would not be applicable to the Scottish Parliament as well. All we are doing is providing over the years a series of snap-shots of the differential movement of gross polling. Those trends can be instructive and for some of us will always be interesting. Back to 1999 is a short history and the Chinese analysts amongst us might well observe that it is too early to say yet if the Old Order is quite dead and buried?

[piperdave]

Jan 13, 2016 1:20:10 GMT carlton43 said:

Jan 13, 2016 0:51:58 GMT piperdave said:

Question for Davıd Boothroyd - based on your view of swing, who is the swing between at Scottish Parliament elections? Labour and Conservative because swing can only be between them based on Butler? Or Labour and SNP because they have always been the largest two parties at said elections (since 1999 for the avoidance of doubt)?

Well, you have David's answer already, but I intend to apply my 3-fold approach to the Scottish situation for HOC results, but I don't see why it would not be applicable to the Scottish Parliament as well. All we are doing is providing over the years a series of snap-shots of the differential movement of gross polling. Those trends can be instructive and for some of us will always be interesting. Back to 1999 is a short history and the Chinese analysts amongst us might well observe that it is too early to say yet if the Old Order is quite dead and buried?

I only have David's answer as far as Westminster elections are concerned. I'm not sure we've had this debate before in respect of any of the devolved legislatures. I certainly can't recall one, so I'll wait and see what he says.