UK parliamentary boundary review

[Oryxslayer]

If there was a tool like DRA for the UK we could have some fun drawing maps or hilarious gerrymanders

Sadly, no such tool exists

Gerrymandering in the UK is a lot harder than the US - the building blocks are much bigger (in Scotland, London and Yorkshire especially). So you can probably make a gerrymander by turning safe seats into more marginal ones and pack the others, like the 2016-2020 NC Map, but you would never be able to make the current Maryland Map or seats like TX-21. Still, see how much fun you can have with it.

Don't be so naive. Countries with population flexibility, unlike the US's strict adherence to equal pop per district, don't need to draw tentacles to gerrymander. In fact a tentacle would do a disservice to the state mappers legitimacy, Instead, one simply has to ensure that the seats which you know from the get go will be won by the opposition have more population per seat than those seats you will be won by the governing party. Another trick is to cut or add seats to the count, depending on if the governments areas are growing compared to the opposition. Cutting seats, like the 600 map, would hurt incumbents, but produced maps more favorable to the Tories.

[Oryxslayer]

I decided to take a look at Wales for fun to see how things stand, since that is the region which will change the most, and it seems like there will be a lot of hard choices made.

- The Northeast has the pop for four seats. One will be entirely Denbighshire given it falls within the 5% threshold. You'll then have one in Flintshire, one in Wrexham, and one crossing mostly in Flint. The Wrexham seat will have the city, but the Flintshire seat could be Deeside+Coast, Buckley+Coast, or a weird Wrexham Suburbs+Coast seat.

- Carmarthenshire has the pop for two seats. Nothing too meaningful on its own, but In combination with the 20.15 seats for the Valleys, this leads to weirdness.

- Penbrokeshire and Ceredigion combined have another two seats. It means the Ceredigion seat will take a bit from her neighbor...

- Which forces the unholy pairing of Powys and the Northwest. Which is hard given the lack of roads and the mountains. Preferably Powys would go with the valleys given it has 1.2 seats and the southern areas are easiest to remove.

- So lets follow this orientation and take Breckon and the rest of south Powys to the valleys. This means Carmarthenshire must be cut along with Pembrokeshire, so that the second Carmarthen seat  can get complete. The Pembroke+Carmarthen+Valleys+Breckon has 24 seats on the dot.

- Ceredigion now goes northwards and everything almost works out on paper. But the Powys surplus is not solved (I guess it goes with Monmouthshire?), the unholyness now plays out in Carmarthenshire, and the Northwest pairing is just barely overpopulated for 3 seats even at maximum allowed deviation.

Theres also the puzzle of how you cut Conwy since the natural cuttings from the city to Llanwst, or the coast vs interior both leave the Conwy seat overpopulated.  


And this is before I even bothered with the valleys.

[Oryxslayer]

Initial Boundary report proposal to be released tomorrow.

[Oryxslayer]

Twitter rumors (AKA grain of salt)

Devonwall is not happening

Cumbria-Lancashire seat because of losses in the NW. Wyre and Preston region particularly affected.

NE losses partially forced on Sunderland.

[Oryxslayer]

I have the shapefile and will be uploading thoughts here.

MAP KEY:

Colors are 2019 (2021 Hartlepool) winners. New Lines are black, old red.

Cornwall:



Old Camborne and Redruth is gone. Its now a north coast seat with Truro and St. Ives absorbing the south

[Oryxslayer]

Devon:



Little bit of Churn in the West, SW Devon changes shape. Exmouth has more of Exeter and less of the outlying areas. Say hello to new SomerDevon seat Tiverton & Minehead around Exmoor. 

[Oryxslayer]

Somerset



Lots of churn. Taunton Deene is now just Taunton thanks to shrinkage.  New Birdgwater seat. Weston Super Mare has shrunk, might be held by Labour whenever they form the next government. bath and Bristol got a lot of work....



Jacob Rees-Mogg's constituency got cracked HARD, but I suspect he can win Frome. Bath Expands. Fifth new Bristol seat in the east of the city and includes New Cheltenham. +1 Lab. Filton & Bradley Stoke loses the coastal territory. May also be a Labour target in a competitive election.

[Oryxslayer]

Can anyone post a picture of London?

I'm not going to bother with having old and new lines on one map here, too much confusion. First thing that stands out to me is the end of the Parallel Hammersmith and Kensington seats, and CoL is in a Labour seat.

[Oryxslayer]

Dorset:



Nothing major I can see. Maybe Mid Dorset & Poole North getting an arm to reach around the harbor  is notable?

Hampshire Coast:



Southampton, Portsmouth, and New Forest seats remain untouched. New Tory seat on Wight as ordered. Eastleigh shrinks, Fareham moves east, and Hedge End is a new seat.

Wiltshire:



Seats cross the Gloucestshire border...more than once I think. Swindon is a point of oddity. All the urban areas remain in the two seats, but the double encirclement of Swindon South by East Wiltshire looks off for some reason.

[Oryxslayer]

Gloucestershire



Stroud shrinks, one has to assume this hurts the Tories. The lines around Gloucester city are awkward given that the urban area is larger that one seat. Tewkesbury shrinks, and The Cotswolds grows.

[Oryxslayer]

Sussex(s)



A lot of churn but from afar the lines look better than previously - no Mid Sussex squiggle.

From west to east, Bogner Regis expands. New Littlehampton seat that reaches inland. Old Shoreham seat is dead, it shifts west and is now Worthing with Goring by Sea. New Shoreham seat is the closest thing to the old Arundel, but Arundel lacks any real heir. the inland was carved up to benefit the coast. Brighton shrinks in the east but stays the same.  Lewes loses the north. The seats further east continue the trend of a shrinking coast and a growing interior.

Hailsham and Crowborough succeeds Wealdon. East Grinstead and Uckfield is a new seat created via the carve-up of others in the region. Crawley is the same. Mid Sussx is truly in the middle and built around Haywards Heath.


Kent



The Thanet seats are now east and west. Canterbury loses Sturry, not any rural areas. Ashford is now a east-facing seat rather than west. This reorganizes the two rural seats to the northwest. New Maidstone and Malling seat, fixes the previous carve-up.

[Oryxslayer]

North Hampshire



Basingstoke is one of the weirder seats to change. There's an urban ward in the NW that is outside the seat and in the rural one, and there are rural areas to the SW of the town that could be traded for it. North East Hampshire no longer surrounds the town on both sides.

Surrey



Reigate shrinks. New Godalming & Ash seat takes up the areas that were once in south Guildford and SW Surrey. Guildford now faces west, and I'm tempted to say the Lib-Dems would win this given local results. Godalming & Ash has a weird Aldershot arm that takes it from Surrey Heath. Epson & Ewell now goes across the M25 for Leatherhead. Esher and Walton drops Cobeham - maybe making it a Lib Dem gain off 2019?


I'm stopping for now with most of the south done.

[Oryxslayer]

New Statesman did a bit of math off the 2019 ward results. These are obviously grains of salt though since partisanship is not set in stone like in the US, and new lines would change things. Some seats were uncompetitive, others were marginals. Also 2019 will likely be remembered as a Tory high mark. Some changes:

- The desire to preserve 3 Hull seats with them all reaching outside the city flipped two for the Tories.
- Lots of political churn in the Black country.
- Esher & Walton would be won by the Lib Dems under the new lines.

[Oryxslayer]

What is the most Conservative seat in the country under these new boundaries?

I mean the place with the best Tory fundamentals - cause voting patterns are a bit more fluid - is still the rural parts of the east. Particularly Lincolnshire, Essex, and almost everything (not Cambridge) in between. The boundaries weren't ever going to change so drastically that seats were not comparable to their previous iterations.

[Oryxslayer]

I didn't realize the new Barrow in Furness seat utilized water connectivity wtf.

The Tories apparently got the most 2019 votes in the new Wirral West according to Britain Elects.

[Oryxslayer]

I posted these images to give some background for YL's posts.

[Oryxslayer]

Yeah electoral calculus has a twelve-seat gain for the tories while New Statesman has six and even has lib dems gaining seats. Who's more reliable.

https://www.electoralcalculus.co.uk/boundaries2023.html

I mean the correct answer is neither - constituency campaigns are not national and therefore swings are not entirely universal. A uncompetitive seat will behave differently than one carpeted by campaigners and their literature. So taking parts of one seat and adding it to another are always going to have a MOE. The 2019 election also has a good chance to go down as close to a high water mark for the Westminster Conservatives. In the event of a competitive election, the seat count will recede and likely partially in new areas when compared to 2019.

The other thing to keep in mind is very little of that seat change came from reshuffling the lines. Some areas certainly gained safe seats and others lost them, but the big factor is that England will gain 10 seats from Wales and Scotland. This map is only a partial picture.

[Oryxslayer]

I'm not familiar with how Wales looks on the ground, but even I can see how ugly this is. Ceredigion Preseli is probably the worst. Merthyr Tydfil and Aberdare, Islwyn and Blaenau Gwent and Rhymney are also very strange - why cross into neighbouring valleys instead of drawing them north-south like now? Ugh.

I guess it comes down to population distribution. Many of us played around with Wales a few months ago. The mutual conclusion was that one would have to make some unenviable tradeoffs somewhere on the map to get everything to work. This is better than some of the alternatives, at least from my persepctive.

[Oryxslayer]

https://twitter.com/ElectionMapsUK/status/1448452319977299973?s=20

Not much changes I think, outside of the Dundee and Elgin regions.

[Oryxslayer]

Weaver Vale is another name which disappears.  Northwich joins Winsford from Eddisbury and Middlewich from Congleton in Mid Cheshire, while Frodsham, Helsby and the parts of Runcorn in the constituency join with the rest of Runcorn from the existing Halton constituency and smaller parts of Ellesmere Port & Neston and Eddisbury to form Runcorn & Helsby.

The new Mid Cheshire looks like a marginal which would have voted Tory by a modest margin in 2019.

For our international posters, ‘Mid’ is another term for ‘Leftovers’ i.e. the places that were left once you drew coherent constituencies around it. In this case, it’s actually and very unusually ended up producing a logical constituency, certainly more so than either of its 2 main predecessors (especially Weaver Vale). ‘Northwich and Winsford’ would have been a perfectly decent name.

I definitely think Cheshire is one of the more interesting areas at the review for this reason among many. A seat is destroyed but it re-emerges a few miles away. The split of Chester seems fairly illogical - I know residents saw it as a gerrymander to split to city along the river and lobbied against it - but then other changes it facilitates are seemingly sensible. Mid Cheshire deals better with her towns than the previous map did.

[Oryxslayer]

I mean, at the end of the day, all constituency models like this are just best guess estimates.  Partially because of how swing behaves when the change is so large, Partially because the subsamples that would be required to determine relative concentrated change are small with high MOE,  and Partially cause it's now and the election is a year away.

However,  there is also one new reason we also have to take into consideration.  The limited evidence we have right now suggests a high level of tactical voting among the Lib-Dems, Labour, and Greens when relevant, in order to defeat the Conservatives. Who knows how long this will persist after Labour take power, but it is a thing right now. And obviously models can't really measure this phenomenon, except maybe the YouGov MRP. It's unlikely therefore that the Lib-Dems lose any seat where the Tories are the challenger. There also will probably be a lot more Lib-Dem gains than we expect.  The flip side of this though is that Labour’s seat count will probably be lower: not cause they are losing winnable seats, but because they aren't going to get the random out-of-nowhere marginal flips we see in the model. Because in those areas, the tactical voting may not go in the direction of Labour.