UK AV/STV constituencies w/Ireland

[andrew111]

This is a bit of an odd scenario where we are united with the RoI, which already has a perfectly good set of STV constituencies. Starting by reducing the number of MPs even further is also strange, and will reduce the benefits of STV.
The answer to your question is "a bit more proportional". If you have more members per seat, as in RoI or Stormont, you get better proportionality.You could try and make a statistical prediction based on the proportionality of Scottish local elections, which are mainly 3 member seats. Trying to predict any more than that would be a Herculean task, but I don't think we would have a Tory plus DUP government right now..

[tiberius]

Sept 16, 2018 6:57:08 GMT andrew111 said:

This is a bit of an odd scenario where we are united with the RoI, which already has a perfectly good set of STV constituencies. Starting by reducing the number of MPs even further is also strange, and will reduce the benefits of STV.
The answer to your question is "a bit more proportional". If you have more members per seat, as in RoI or Stormont, you get better proportionality.You could try and make a statistical prediction based on the proportionality of Scottish local elections, which are mainly 3 member seats. Trying to predict any more than that would be a Herculean task, but I don't think we would have a Tory plus DUP government right now..

Come to think about it, there are a ton of unknown unknowns as well about how the Irish political scene would develop in this scenario, as well. It certainly seems theoretically possible that a Tory+regional Irish political party (a la some present day version of the Irish Parliamentary Party) deal might be done, and that would give a very narrow majority or strong minority along the lines the Tories have now.
It would really all come down to Ireland, wouldn't it?

[minionofmidas]

Come to think of it, has anybody ever tried drawing 3-5 member seats of the UK, on the Irish quota and trying to emulate the Irish commission's practices? What would that turn out like? (Probably better to ignore the obvious next fantasy question - who would win those seats if the Irish party system extended to the UK? - because it's just absurd to assume FF could exist in the UK.)

[tiberius]

Sept 16, 2018 8:16:17 GMT minionofmidas said:

Come to think of it, has anybody ever tried drawing 3-5 member seats of the UK, on the Irish quota and trying to emulate the Irish commission's practices? What would that turn out like? (Probably better to ignore the obvious next fantasy question - who would win those seats if the Irish party system extended to the UK? - because it's just absurd to assume FF could exist in the UK.)

3-5 member seats emulating the Irish commission's practices is feasible, but not the Irish quota (which is roughly 30k per member). (Unless you want a roughly 1,500 person parliament) However, fantasy maps for a regional parliament would be a different story.
The exact results would hinge on the quota.
A peak at what you might be looking for might be in the map I posted - just double the number of members elected in every seat. Voila. 930 member parliament.

[tiberius]

A possible setup for Wales:

Snowdonia: 3 members. As Plaid, Labour, and the Conservatives all have MPs in the area covered here, it likely is 1 LAB, 1 PC, and 1 CON.
Denbighsure: 1 member. LAB-CON marginal under FPTP. Marking this down as LAB because favorable preference flows from Plaid voters likely ensures Labour victory here more often than not.
Wrexham and Deeside: 3 members. Likely elects 2 LAB and 1 CON, on account of the fact only Labour and the Conservatives have enough support to elect an MP. *insert Lib Dem bar chart here* *insert bar graphs of results from the 1906 GE*
Powys: 2 members. Covers all of Ceredigion and Powys. 1 CON and 1 LD.
Carmarthen and Pembrokeshire: The election of a Conservative is guaranteed here; of 4 MPs whose RL constituencies lay in this three-member seat, 2 are Tories, 1 is Plaid, and 1 is Labour. However, both of the Tory seats are at least somewhat marginal, the one Labour seat has a very safe majority, and Labour placed second in the Plaid-won seat. Ultimately, the Plaid vote is too weak to elect an MP, but the preference flows ensure 2 of 3 members are Labour. 2 LAB, 1 CON.
Swansea: This seat eats into Southern Wales, hard Labour territory, but it also includes the area of the Gower seat, won by the Tories in 2015. A very solid 2 LAB, 1 CON.
Rhondda Cynon Taf Neath: heartland Labour territory. 3 LAB.
Brigend: The inclusion of the Vale of Glamorgan prevents Tories from being shut out here. 2 LAB, 1 CON.
Caerphilly: This is a purely South Welsh Valleys seat. 'Nuff said. 3 LAB.
Cardiff: This seat is home to the Welsh capital. If the Liberal Democrats ever recover their lost glory they could have a chance at shutting the Tories out here. 2 LAB, 1 CON.
Monmouth and Newport: Covers Newport, Monmouthshire, and the far eastern parts of the Valleys. 2 LAB, 1 CON.
Total: 20 LAB, 8 CON, 1 LD, 1 PC
11 seats electing 30 members

[therealriga]

Sept 16, 2018 8:16:17 GMT minionofmidas said:

Come to think of it, has anybody ever tried drawing 3-5 member seats of the UK, on the Irish quota and trying to emulate the Irish commission's practices? What would that turn out like? (Probably better to ignore the obvious next fantasy question - who would win those seats if the Irish party system extended to the UK? - because it's just absurd to assume FF could exist in the UK.)

Yes. I posted it here a couple of years ago covering 1983-2015: vote-2012.proboards.com/thread/7174/stv-uk-1983-2015

2017 figures I got were

Con 292 (+11)
Lab 292 (+61)
SNP 23 (-13)
LD 22 (-3)
DUP 9 (+2)
SF 4 (=)
UUP 2 (-1)
SDLP 2 (=)
PC 1 (-4)
APNI 1 (=)
GP 1 (-1)
IND 1 (+1)
IU 0 (-1)
UKIP 0 (-52)

Edit: just noticed you said Irish quota. That's population based so wouldn't be so doable. Ireland currently has 1 TD per 30,356 people. The same ration applied to the UK would give you a 2176 member parliament (!)

[tiberius]

Redcar and Cleveland: 2 seater. A solid 1 LAB, 1 CON.
Hartlepool and Stockton: 3 seater. A solid 2 LAB, 1 CON.
Darlington: 1 seater. Preference flows from other candidates ensures a Labour victory. 1 LAB.
Durham East: 2 seater. A solid 1 LAB, 1 CON.
Durham West: 3 seater. A solid 2 LAB, 1 CON.
Sunderland: 3 seater. A solid 2 LAB, 1 CON.
South Shields and Gateshead West: The Conservative vote here is weak enough to make the second seat a marginal. In 2017 however, a Tory would be able to nab the second seat, tying Labour. 1 LAB, 1 CON.
Gateshead East: Labour always finishes first here, but the Tory vote is strong enough to usually take the second seat. 1 LAB, 1 CON.
Newcastle upon Tyne: Labour's preformance here is better than elsewhere in Tyne and Wear. 2 LAB.
North Shields: see Gateshead East. 1 LAB, 1 CON.
Northumberland: Until recently this would have been 1 seat each for the three biggest non-regional parties in the UK. But in the aftermath of the LD vote falling dramatically since 2010, the LD seat for now is taken by Labour. 2 LAB, 1 CON.
Total: 16 LAB, 9 CON.
11 seats electing 25 members.

[tiberius]

Herefordshire: Labour is clearly in the minority, but its vote is big enough to net it an MP. 1 CON, 1 LAB.
Worcester: the areas in the RL marginal seat of Worcester is the biggest source of Labour votes in this seat, and ensuring the election of a Labourite. But the Tories have the advantage. 2 CON, 1 LAB.
Redditch and East Worcestershire: replace Worcester with Redditch, and the same situation holds. 2 CON, 1 LAB.
South Warwickshire: not much to see here. 2 CON, 1 LAB.
North Warwickshire: A very solid 1 CON, 1 LAB.
Coventry: this is a mirror image of Herefordshire. 2 LAB, 1 CON.
Solihull: This once would have been an easy 1 CON, 1 LD. But the Liberal Democrats have been overtook by Labour, who likely pick up the second seat by a narrow margin. 1 CON, 1 LAB.
Birmingham South: Not much of note here. 2 LAB, 1 CON.
Birmingham Central: Labour pulls such massive margins from parts of this seat that it's likely a Tory shutout. 3 LAB.
Birmingham North: The inclusion of some rather marginal Labour territory ensures a Conservative gets elected here. 2 LAB, 1 CON.
West Bromwich: suburbia of Birmingham? Anyway, it appears that Labour is certainly in advantage here. 2 LAB, 1 CON.
Dudley: most of this area is represented by Tories and the one Labour MP has a majority of 22. Likely result is 2 CON, 1 LAB.
Walsall: Walsall and three wards in the Wolverhampton council area. what was said about Dudley can also be said about here. 2 CON, 1 LAB.
Wolverhampton: Labour leaning, but not enough to shut the Tories out. 1 LAB, 1 CON.
Shropshire: A land of natural beauty, of gorgeous scenery. 2 CON, 1 LAB.
Telford: This is Wolverhampton's opposite. 1 CON, 1 LAB.
South Staffordshire: Labour gets just enough voters here to be certain to effect a tie. 1 CON, 1 LAB.
Tamworth and Burton: Labour gets enough support here to tie the Tories in seat count. 1 CON, 1 LAB.
Newcastle-upon-Lyme: see Tamworth and Burton. 1 CON, 1 LAB.
Stoke on Trent: Labour has a narrow advantage in regards to getting the third seat, but in a good year for the Tories, they'd flip it. 2 LAB, 1 CON.
Staffordshire Moorlands: this is a one-seater, and one that remains solid for Karen Bradley. 1 CON.
Total: 27 CON, 27 LAB.
21 seats electing 54 members.

[tiberius]

West Cornwall: roughly 40% of Cornwall resides in this seat. Also includes the Isles of Scilly. 1 CON, 1 LD.
East Cornwall: this seat covers the rest of Cornwall. 1 CON, 1 LAB, 1 LD.
Portsmouth: within Portsmouth itself the Tories and Labour are roughly evenly divided, but the rest of the seat is clearly Tory-leaning. 2 CON, 1 LAB.
Dartmoor and Torquay: this seat elects three members, and includes some of the most heavily Tory portions of Devon. 2 CON, 1 LAB.
Exeter: consisting of Exeter and areas in eastern Devon, this is the Devon constituency likeliest to elect two Labourites. However, the hinterlands give the Tories a slight edge. 2 CON, 1 LAB.
North Devon: not much to see here. 1 CON, 1 LD.
Poole: the unitary authorities/council areas of Poole and Purbeck. 1 CON, 1 LAB.
Bournemouth: the Lib Dem vote here is weak, meaning they do not get a seat. The Tories profit from this. 2 CON, 1 LAB.
North Dorset and Weymouth: the rest of Dorset. 2 CON, 1 LD.
South Somerset: there's a lot of ancestral Liberalness here, and Yeovil was Paddy Ashdown's Commons seat for 18 years. 1 CON, 1 LD, 1 LAB.
Bridgwater: see South Somerset. 1 CON, 1 LD.
Bath: natural Liberal territory. 1 CON, 1 LD.
Weston-super-Mare: the LDs were once second here. But not anymore. 1 CON, 1 LAB.
South Wiltshire: Tory heartland. 2 CON, 1 LAB.
North Wiltshire: see South Wiltshire. 1 CON, 1 LAB.
Swindon: Tory-leaning 2-seater. 1 CON, 1 LAB.
Bristol South: Labour-leaning 2-seater. 1 LAB, 1 CON.
Bristol North: see Bristol South. 1 LAB, 1 CON.
South Gloucestershire: the three constituencies in the seat all gave Tories absolute majorities in 2017. Ultimately, it is 2 CON, 1 LAB.
Severn and Dean: this one-seater, consisting of the Forest of Dean council area with the addition of one ward, is safe for the Tories. 1 CON.
Cotswolds and Stroud: designed to unite the vast majority of the Cotswolds AOMB. 1 CON, 1 LAB.
Gloucester and Cheltenham: a cluster of major urban centers in Gloucestershire. 1 CON, 1 LAB, 1 LD.
Total: 29 CON, 16 LAB, 8 LD.
23 seats electing 53 members.

[tiberius]

Richmond Park and Wandsworth: This one is tricky. I'd go with 1 CON, 1 LAB, 1 LD.
Kingston and Merton: each of the three parties has a strong chance of getting an MP here. 1 CON, 1 LAB, 1 LD.
Sutton: this one is easy. 1 CON, 1 LD.
Brixton: hard to tell what would happen here. Labour's chances of a sweep are rather low, though. I guess 2 LAB, 1 LD.
Croydon: the southern part of this seat is solidly Tory, but its northern part is even more solidly Labour. 2 LAB, 1 CON.
Southwark: Labour has solid support here, but not enough for a sweep. 2 LAB, 1 LD.
Lewisham: it's interesting...while the Tories in theory would have a shot at winning a seat here, it's a tough climb, and preference flows from LDs and minor party voters in the seat would favor Labour. Due to this, Labour likely sweeps, even though a Lib Dem might have a chance if they get a big enough vote. 2 LAB.
Bromley: this is the most solidly Tory borough in South London, and one of the most heavily Tory urban areas in all of Britain. In a good year for the Tories, they will sweep. 2 CON, 1 LAB.
Greenwich: Labour-leaning two-seater. Very solid 1 LAB, 1 CON.
Bexley: Conservative-leaning two-seater. Very solid 1 LAB, 1 CON.
Total: 13 LAB, 8 CON, 5 LD
10 seats electing 26 members

[tiberius]

Twickenham: this seat is unchanged. Mark it down as 1 LD.
Hounslow: Labour is strong here but the Tories have enough voters here to effect a tie. 1 LAB, 1 CON.
Hayes and South Ruislip: both parties have strong areas here. 1 LAB, 1 CON.
Harrow: the Conservatives have a slim but real advantage here. 2 CON, 1 LAB.
Ealing: this is solidly Labour territory. 2 LAB, 1 CON.
Westminister North and Brent: This one is obvious. 2 LAB, 1 CON.
Barnet: this seat is on the knife-edge. Gun to head, I'd guess 2 CON, 1 LAB.
Cities of London and Westminster: this seat swung heavily to Labour in 2017, but the underlying Conservative lean was not entirely erased. 2 CON, 1 LAB.
Islington: solidly Labour territory just north of the City. 2 LAB.
Enfield: this is a more Labour-leaning version of Barnet. 2 LAB, 1 CON.
Haringey: this is yet another solidly Labour seat in North London. 2 LAB.
Hackney: see Haringey. 2 LAB.
Tower Hamlets: see Haringey. 2 LAB.
Waltham Forest: solid 1 LAB, 1 CON.
Newham: see Haringey. 2 LAB.
Ilford: This is a solid 2 LAB, 1 CON.
Dagenham and Romford: mirror image of Ilford. 2 CON, 1 LAB.
Total: 26 LAB, 15 CON, 1 LD
17 seats electing 42 members.

[tiberius]

Milton Keynes: all but two wards in the Milton Keynes council area. 1 LAB, 1 CON.
Buckingham: this is a hard one to prognosticate, thanks to conventions surrounding the Commons Speaker. But I'd assume that it would normally elect 1 CON, 1 LAB.
Chiltern Hills: this seat lies in southern Buckinghamshire. 2 CON, 1 LAB.
Oxford: this is Tory-leaning territory, but the opposition vote is strong enough to effect a three-way tie in seat count. 1 CON, 1 LD, 1 LAB.
South Oxfordshire: more Tory territory. 2 CON, 1 LAB.
Reading: this is Tory heartland affixed to minster town of Reading. 2 CON, 1 LAB.
Mid Berkshire: this is pure Tory heartland, containing about half of Maidenhead. 2 CON.
Bracknell and Slough: while Slough is heavily Labour-voting, the rest of the seat has an intense Conservative lean. PM May might live here. 2 CON, 1 LAB.
Epsom and Tandridge: Epsom, Reigate and Banstead, and Tandridge council areas. 2 CON, 1 LAB.
Surrey Hills: this seat contains the most rural portions of Surrey. 2 CON, 1 LD.
Woking: Woking and Surrey Heath council areas. Very safe Tory territory. 2 CON.
North Surrey: this 3-seater is, like all the other Surrey seats, safely Conservative. 2 CON, 1 LAB.
Aldershot: this seat is very Conservative, but Aldershot ensures a tie in seat count in most instances. 1 CON, 1 LAB.
Basingstoke: this constituency has an overwhelming Conservative lean, but a Tory sweep is nonetheless nearly impossible. 2 CON, 1 LAB.
New Forest: a Tory sweep is possible here, but by no means is ensured. 2 CON.
Southampton: this is a Labour-leaning two-seater. 1 LAB, 1 CON.
Eastleigh and Fareham: a collection of Tory-favoring coastal council areas. 2 CON, 1 LAB.
Portsmouth and South Havant: all of Portsmouth and four wards in Havant. A very solid 1 CON, 1 LAB.
Winchester and South Downs: this seat is very rural, very Conservative, and very forested. 2 CON, 1 LAB.
Horsham: northern and western areas in West Sussex. 1 CON, 1 LAB.
Crawley: this 3-seat contains the marginal seat of Crawley, but also has a large section of rural Tory country to its east, west, and south. A very solid 2 CON, 1 LAB.
Worthing: coastal areas in West Sussex. 2 CON, 1 LAB.
Brighton West: this seat contains all of Brighton Pavilion. 1 LAB, 1 GRN.
Brighton East and Eastbourne: the Liberals are stronger than usual here, resulting in a three-way tie in the seat count. 1 LAB, 1 CON, 1 LD.
Hastings and Rother: more solid Tory territory in Southern England. 2 CON, 1 LAB.
Dover: this seat covers the Thanet and Dover districts, as well as the town of Folkestone. 2 CON, 1 LAB.
Canterbury: coastal areas along the northern coast of Kent. This seat includes the Isle of Sheppey. 2 CON, 1 LAB.
Ashford and Maidstone: southeastern Kent. 2 CON, 1 LAB.
Rochester and Gravesand: coastal areas in Kent. 2 CON, 1 LAB.
Dartford: this one-seater is safe for the Tories. 1 CON.
South West Kent: another Tory-leaning Kent constituency. 2 CON, 1 LAB.
Total: 51 CON, 26 LAB, 3 LD, 1 GRN
31 seats electing 81 members.

[minionofmidas]

Sept 16, 2018 14:10:47 GMT therealriga said:

Sept 16, 2018 8:16:17 GMT minionofmidas said:

Come to think of it, has anybody ever tried drawing 3-5 member seats of the UK, on the Irish quota and trying to emulate the Irish commission's practices? What would that turn out like? (Probably better to ignore the obvious next fantasy question - who would win those seats if the Irish party system extended to the UK? - because it's just absurd to assume FF could exist in the UK.)

Yes. I posted it here a couple of years ago covering 1983-2015: vote-2012.proboards.com/thread/7174/stv-uk-1983-2015

2017 figures I got were

Con 292 (+11)
Lab 292 (+61)
SNP 23 (-13)
LD 22 (-3)
DUP 9 (+2)
SF 4 (=)
UUP 2 (-1)
SDLP 2 (=)
PC 1 (-4)
APNI 1 (=)
GP 1 (-1)
IND 1 (+1)
IU 0 (-1)
UKIP 0 (-52)

Edit: just noticed you said Irish quota. That's population based so wouldn't be so doable. Ireland currently has 1 TD per 30,356 people. The same ration applied to the UK would give you a 2176 member parliament (!)

Obviously a realistic proposal for a new Westminster system was not what I had in mind so much as seeing what it would look like, how much those rules would help avoid stupid seats (far from entirely, if their application to Ireland is any indication) etc. It helps that the number of constituencies wouldn't actually be far different.

[andrew111]

Sept 16, 2018 8:23:50 GMT tiberius said:

Sept 16, 2018 8:16:17 GMT minionofmidas said:

Come to think of it, has anybody ever tried drawing 3-5 member seats of the UK, on the Irish quota and trying to emulate the Irish commission's practices? What would that turn out like? (Probably better to ignore the obvious next fantasy question - who would win those seats if the Irish party system extended to the UK? - because it's just absurd to assume FF could exist in the UK.)

3-5 member seats emulating the Irish commission's practices is feasible, but not the Irish quota (which is roughly 30k per member). (Unless you want a roughly 1,500 person parliament) However, fantasy maps for a regional parliament would be a different story.
The exact results would hinge on the quota.
A peak at what you might be looking for might be in the map I posted - just double the number of members elected in every seat. Voila. 930 member parliament.

I see that the latest boundary changes in Ireland have reduced the number of 3 member seats from 13 to 9 - a step in the right direction in my view! They obviously are not allowed 6 member seats, which I would certainly allow in an STV system. For example Sheffield + Penistone would make a decent 6 member seat on the present quota, which would give all of Lib Dem, Con and Green a chance of representation compared to the current red desert...

[pepperminttea]

Sept 16, 2018 21:26:09 GMT tiberius said:

Redcar and Cleveland: 2 seater. A solid 1 LAB, 1 CON.
Hartlepool and Stockton: 3 seater. A solid 2 LAB, 1 CON.
Darlington: 1 seater. Preference flows from other candidates ensures a Labour victory. 1 LAB.
Durham East: 2 seater. A solid 1 LAB, 1 CON.
Durham West: 3 seater. A solid 2 LAB, 1 CON.
Sunderland: 3 seater. A solid 2 LAB, 1 CON.
South Shields and Gateshead West: The Conservative vote here is weak enough to make the second seat a marginal. In 2017 however, a Tory would be able to nab the second seat, tying Labour. 1 LAB, 1 CON.
Gateshead East: Labour always finishes first here, but the Tory vote is strong enough to usually take the second seat. 1 LAB, 1 CON.
Newcastle upon Tyne: Labour's preformance here is better than elsewhere in Tyne and Wear. 2 LAB.
North Shields: see Gateshead East. 1 LAB, 1 CON.
Northumberland: Until recently this would have been 1 seat each for the three biggest non-regional parties in the UK. But in the aftermath of the LD vote falling dramatically since 2010, the LD seat for now is taken by Labour. 2 LAB, 1 CON.
Total: 16 LAB, 9 CON.
11 seats electing 25 members.

Not so sure about that. The Tories carried Northumberland by 2,667 votes in 2017 which would at the very least make the third seat a closely contested tossup. Obviously we don't know how people would have voted under a different system but I think it's more likely than not the Tories would have eked it in 2017 (and 2015 too).

But if you're assuming that Labour would win under the assumption that they're more 'transfer friendly' if you look at the police and crime commissioner elections in most cases transfers tend to favour the Tories in Lab-Con battles (though it does vary from area to area).

[Deleted]

Sept 17, 2018 14:55:12 GMT pepperminttea said:

Sept 16, 2018 21:26:09 GMT tiberius said:

Redcar and Cleveland: 2 seater. A solid 1 LAB, 1 CON.
Hartlepool and Stockton: 3 seater. A solid 2 LAB, 1 CON.
Darlington: 1 seater. Preference flows from other candidates ensures a Labour victory. 1 LAB.
Durham East: 2 seater. A solid 1 LAB, 1 CON.
Durham West: 3 seater. A solid 2 LAB, 1 CON.
Sunderland: 3 seater. A solid 2 LAB, 1 CON.
South Shields and Gateshead West: The Conservative vote here is weak enough to make the second seat a marginal. In 2017 however, a Tory would be able to nab the second seat, tying Labour. 1 LAB, 1 CON.
Gateshead East: Labour always finishes first here, but the Tory vote is strong enough to usually take the second seat. 1 LAB, 1 CON.
Newcastle upon Tyne: Labour's preformance here is better than elsewhere in Tyne and Wear. 2 LAB.
North Shields: see Gateshead East. 1 LAB, 1 CON.
Northumberland: Until recently this would have been 1 seat each for the three biggest non-regional parties in the UK. But in the aftermath of the LD vote falling dramatically since 2010, the LD seat for now is taken by Labour. 2 LAB, 1 CON.
Total: 16 LAB, 9 CON.
11 seats electing 25 members.

Not so sure about that. The Tories carried Northumberland by 2,667 votes in 2017 which would at the very least make the third seat a closely contested tossup. Obviously we don't know how people would have voted under a different system but I think it's more likely than not the Tories would have eked it in 2017 (and 2015 too).

But if you're assuming that Labour would win under the assumption that they're more 'transfer friendly' if you look at the police and crime commissioner elections in most cases transfers tend to favour the Tories in Lab-Con battles (though it does vary from area to area).

I think that's partly because Labour won on more outright first preferences than the Tories

[pepperminttea]

Sept 17, 2018 14:59:54 GMT @priceofdawn said:

Sept 17, 2018 14:55:12 GMT pepperminttea said:

Not so sure about that. The Tories carried Northumberland by 2,667 votes in 2017 which would at the very least make the third seat a closely contested tossup. Obviously we don't know how people would have voted under a different system but I think it's more likely than not the Tories would have eked it in 2017 (and 2015 too).

But if you're assuming that Labour would win under the assumption that they're more 'transfer friendly' if you look at the police and crime commissioner elections in most cases transfers tend to favour the Tories in Lab-Con battles (though it does vary from area to area).

I think that's partly because Labour won on more outright first preferences than the Tories

Sure because solidly Labour areas are more monolithic than solidly Tory areas are. I'm talking about the 'competitive' races where the two parties finished 1&2. In most of these the transfers went in the Tories's favour thus at least somewhat negating the often heard view that third party voters would prefer Labour.

[Deleted]

Sept 17, 2018 15:07:16 GMT pepperminttea said:

Sept 17, 2018 14:59:54 GMT @priceofdawn said:

I think that's partly because Labour won on more outright first preferences than the Tories

Sure because solidly Labour areas are more monolithic than solidly Tory areas are. I'm talking about the 'competitive' races where the two parties finished 1&2. In most of these the transfers went in the Tories's favour thus at least somewhat negating the often heard view that third party voters would prefer Labour.

but you do have to remember two things:
1) for political and financial reasons Labour & the Tories are the only parties to contest every region accross the UK and therefore transfers Labour might hope from the Lib Dems and Greens are limited. Whereas UKIP stood in more seats than the Lib Dems and their transfers certainly help vthe Tories
2)Supplementary vote doesn't reallocate the 2nd preferences of eliminated candidates. So lots of votes are lost amongst elimited candidates voting for each other

[tiberius]

Norfolk: this 3-seater has a slight Labour lean. 2 LAB, 1 CON.
Great Yarmouth: the Tories narrowly miss out on a second seat here. 1 CON, 1 LD, 1 LAB.
Fenlands: not much to say here. 2 CON, 1 LAB.
Lowestoft: a big chunk of rural Suffolk. 2 CON, 1 LAB.
Ipswich: a true marginal won by Labour in 2017, effectively paired with a safe Tory seat in the country. 1 CON, 1 LAB.
Bury Saint Edmunds: this seat centers on the eponymous minster town. 1 CON, 1 LAB.
Peterborough: the marginal area of Peterborough paired with tons of Tory hinterland. 2 CON, 1 LAB.
Huntingdon: areas around Cambridge itself, as well as areas to its north and west. Looks like a very fertile Lib Dem target in the next election. 2 CON, 1 LAB.
Cambridge: this is a Lib Dem-Labour marginal, that would have stayed Lib Dem in 2015. But in 2017, it would be different. 1 LAB.
Bedford and North Bedfordshire: this is the more Tory half of Bedfordshire. 2 CON, 1 LAB.
Luton and South Bedfordshire: this is the more Labour half of Bedfordshire. 2 LAB, 1 CON.
Watford: this seat is heavily Tory London exurbia. 2 CON, 1 LAB.
St. Albans: not much special about this seat as far as politics goes. 2 CON, 1 LAB.
Welwyn Hatfield: this one-seater has a clear Conservative lean, and is held by former Tory party chair Grant Shapps. 1 CON.
East Hertfordshire: see St. Albans. 1 CON, 1 LAB.
Stevenage: see Huntingdon. 1 CON, 1 LAB.
Basildon and Thurrock: this is the home of the fabled Essex Man, and holds some of the most well-known marginal territory in the country. 2 CON, 1 LAB.
Southend-on-Sea: Conservative-leaning coastal towns. 2 CON, 1 LAB.
Chelmsford: insert comment about how the Tories have the advantage here. 2 CON, 1 LAB.
Harlow: it's difficult to see Tories sweep here. 1 CON, 1 LAB.
Braintree: this is a collection of rural territory in the central and northwestern parts of Essex. 2 CON, 1 LAB.
Colchester: Tory-leaning, much like every other seat in Essex. 2 CON, 1 LAB.
Total: 33 CON, 23 LAB, 1 LD
22 seats electing 57 members

[tiberius]

Sept 17, 2018 15:07:16 GMT pepperminttea said:

Sept 17, 2018 14:59:54 GMT @priceofdawn said:

I think that's partly because Labour won on more outright first preferences than the Tories

Sure because solidly Labour areas are more monolithic than solidly Tory areas are. I'm talking about the 'competitive' races where the two parties finished 1&2. In most of these the transfers went in the Tories's favour thus at least somewhat negating the often heard view that third party voters would prefer Labour.

I gave Labour 2 of 3 seats in Northumberland because the Lib Dem and Green preferences would lean towards them, allow them to very slightly edge past the Tories (the remaining Lib Dem vote - a large portion of the more Tory-friendly ex-Lib Dem vote is now voting Conservative). In 2015, the opposite would hold in regards to UKIP preference flows - the Tories would win on the back of getting more first prefs and getting more (though not massively much more) from the UKIP.