Two questions for British posters

[Linus Van Pelt]

Enjoying your various detailed election threads, but you've got some crazy lingo there.

1. What are "notionals"?

2. What is the "swing", such that, as seems to be the case, the "swing" is from just one party to just one other party, even though there are three significant parties (or more in some areas)?

[Filuwaúrdjan]

1. Notionals = notional results = estimates of the result from the previous election in a constituency affected by boundary changes. Lower level results are not released in British elections* and so it isn't possible to just add up ward level results. Instead, estimates based largely on databases of local election results are used. There's usually at least one howler every post-boundary review election.

2. This is a little bit controversial. Basically, 'swing' is the difference between the vote percentage changes of party x and party y. Example:

John Jones, Party X, -2.3
Thomas Thomas, Party y, +1.7
Swing = 2.0

Originally, swings were only calculated between the Labour and Conservative parties (even in constituencies where another party finished ahead of one of them) and there are still people who argue that this is the correct way of doing this (given the original point of working out swings - forecasting the election result from early results - they are essentially correct. Swings done in this way are called Butler Swings, after their inventor, David Butler), but the media has (for a long time now) worked out the swing between whoever comes first and second in a given constituency. The media being the media, the tend to do this even if one of the top two parties didn't actually run in the previous election.

*With the exception of GLA and Scottish Parliament elections.

[Linus Van Pelt]

Thanks!

[KuntaKinte]

2. This is a little bit controversial. Basically, 'swing' is the difference between the vote percentage changes of party x and party y. Example:

John Jones, Party X, -2.3
Thomas Thomas, Party y, +1.7
Swing = 2.0

I may be completely off, but wouldn't that be a swing of 4.0 pionts?

[Хahar 🤔]

2. This is a little bit controversial. Basically, 'swing' is the difference between the vote percentage changes of party x and party y. Example:

John Jones, Party X, -2.3
Thomas Thomas, Party y, +1.7
Swing = 2.0

I may be completely off, but wouldn't that be a swing of 4.0 pionts?

Take the average.

[Harry Hayfield]

Enjoying your various detailed election threads, but you've got some crazy lingo there.

1. What are "notionals"?

A "notional" result is an estimation of the result in that constituency if tyhe boundaries for the election being contested where in place at the last election. In the UK, this is quite a new science having been done for the Feb 1974, 1983, 1997, 2005 and 2010 general elections/b]

2. What is the "swing", such that, as seems to be the case, the "swing" is from just one party to just one other party, even though there are three significant parties (or more in some areas)?

Swing is the average of change from one party to another. For instance: Con +5% Lab -3% Lib Dem -2% means a Lab to Con swing of 4%, Lab to Lib Dem of 0.5% and Lib Dem to Con of 3.5%. When people say "The Conservatives need a 6.9% swing to win a majority, it means that if 7 in every 100 Lab voters switch to Con, then 116 seats which are currently Lab would be turned into Con seats

Or is that too complex for you?

[Hash]

Notionals are basically what is sometimes referred to as redistributed results in Canada (see the 2000 redistributed results available on Pundits Guide.ca or Elections Canada for the 2004 ridings).

[minionofmidas]

2. This is a little bit controversial. Basically, 'swing' is the difference between the vote percentage changes of party x and party y. Example:

John Jones, Party X, -2.3
Thomas Thomas, Party y, +1.7
Swing = 2.0

I may be completely off, but wouldn't that be a swing of 4.0 pionts?

Only in the United States of America.

[afleitch]

There's usually at least one howler every post-boundary review election.

Gordon 1997 springs to mind

[Harry Hayfield]

There's usually at least one howler every post-boundary review election.

Gordon 1997 springs to mind

Dunbartonshire East (2005)
Birmingham, Erdington (1983)

[Chancellor of the Duchy of Little Lever and Darcy Lever]

Dumfriesshire, Clydesdale and Tweeddale (2005)

[afleitch]

Dumfriesshire, Clydesdale and Tweeddale (2005)

Quite the howler. Though they can be forgiven for that one due to the shape of the damn thing.

[Filuwaúrdjan]

Dumfriesshire, Clydesdale and Tweeddale (2005)

Quite the howler. Though they can be forgiven for that one due to the shape of the damn thing.

It's probably the most insane constituency currently existing, which is saying something.

Forest of Dean in 1997 is another one. Presumably nostalgia got the better of analysis in that case. The 1983 notionals for Brecon & Radnor were clearly 'wrong' as well... Huddersfield area was bad too, IIRC.

[afleitch]

Dumfriesshire, Clydesdale and Tweeddale (2005)

Quite the howler. Though they can be forgiven for that one due to the shape of the damn thing.

It's probably the most insane constituency currently existing, which is saying something.

When you look at it, you genuinely have to stop and think; "where the hell are the Tory voters located?" Its only when you look at what's not in the seat then it starts to make sense.

I fully expect Wales to suffer its own abomination if it's representation is reduced at the next review

[Filuwaúrdjan]

When you look at it, you genuinely have to stop and think; "where the hell are the Tory voters located?" Its only when you look at what's not in the seat then it starts to make sense.

Pretty much; it's also pretty clear that they didn't realise quite how strong Labour is in Dumfries proper. Though when you look at it town-by-town it becomes a little clearer... Lochmaben, Annan and Gretna are all Labour, but Moffat, Langholm and Lockerbie are all Tory. Sanquhar and Kirkconnel are very Labour but they're also pretty small, and then you have the far south of Lanarkshire and the area in the Borders.

You must be logged in to read this quote.

Urgh, yes. Depending on how they do it, we might get more than one. If they're logical and understand that English assumptions about rural-urban divisions ought not be applied to Wales, then we might get away with one or two. If not... urgh. Mid Wales will be an absolute mess.