2018 Review - West Midlands

[Adrian]

I'll post my WM plan(s) soon. I prefer to deal with Sandwell and Birmingham on their own, although pairing them, with one Birmingham ward in a Sandwell seat (which is excusable on account of the fact that it would equalise the size of seats in the two areas), produces quite a nice arrangement of seats, albeit with several split wards.

Since I prefer to deal with Birmingham and Sandwell on their own, it follows that I prefer a Halesowen and Stourbridge seat, even though that means putting a (very small) part of Stourbridge in the Dudley West seat. Having said that, plans like YL's with a cross-border seat have their merits, one of which of course is that a cross-border seat already exists.

(Just to reiterate my opinion: there are plenty of good reasons for crossing LA boundaries (and each case can be assessed on its merits), but avoiding ward splits is not one of them.)

As for the rest of YL's plan, there are some good features, and a couple of obvious problems, i.e. the splitting of Tipton and the butchering of Walsall.

[greenchristian]

May 18, 2016 21:12:14 GMT Adrian said:

I'll post my WM plan(s) soon. I prefer to deal with Sandwell and Birmingham on their own, although pairing them, with one Birmingham ward in a Sandwell seat (which is excusable on account of the fact that it would equalise the size of seats in the two areas), produces quite a nice arrangement of seats, albeit with several split wards.

Given that the border between Birmingham and Sandwell is somewhat arbitrary in places with a lot of people in Sandwell thinking of themselves as Brummies, crossing that border isn't necessarily a big deal.

[islington]

EAL -

There is a way of keeping your Dudley South and Stourbridge seats without crossing into Wolverhampton/Walsall (meaning that those boroughs, with Oscott, can have 5 whole seats). How about this?

HALESOWEN & RR - Compared with your plan, loses Tividale and gains Old Warley. 75937

SMETHWICK AND WEST BROMWICH - Loses Old Warley and gains Great Bridge. Ugly, I know, but I agree with your underlying point that you have to get a 9-ward seat out of Sandwell somehow. 77932

DUDLEY NORTH AND OLDBURY - Keeps the Dudley wards from your version, gains Oldbury and Tividale and loses the Wolverhampton element. 78197

WEDNESBURY AND TIPTON - The rest of Sandwell. Another ugly seat but contiguous and legal. 73449

Also, looking at your Birmingham, I think the reason it works with only one ward split is that the removal of Oscott effectively reduces the electorate of the city so that it's much nearer the quota for nine seats (with Oscott: 686804 = 9.19; without Oscott: 669321 = 8.95). This gives you much more scope to allow the wards to cluster in groups of four, which is what most of them seem to want to do, even though the resulting seats tend to be toward the lower end of the permitted range. Whereas, if you don't cross the city boundary, you need seats that average out on the large side; and this forces you to resist the tendency of the wards to form seats on the small side - hence the need for multiple ward splits.

But I still think the best answer is to take slightly greater liberties with the city boundary and avoid any ward splits at all.

[East Anglian Lefty]

You're right that it's possible to split Dudley and Sandwell on the one hand and Walsall and Wolverhampton on the other into separate groups. I hadn't noticed that and it probably is worth doing. Personally, rather than putting Princes End into Smethwick, I'd rather add Great Barr. Equally ugly, but Great Barr will be peripheral whatever you do.

But I can't agree that it's better to cross the city boundary on multiple occasions than to split a ward. Local authority boundaries are, according to the legislation, at least as important a criterion to consider as wards, and they're considerably more permanent. On a practical level, the ward split I'm proposing is right in the city centre, where there will be a high degree of churn and transience and hence ward boundaries will be less important to the locals anyway.

[Adrian]

May 18, 2016 21:20:15 GMT greenchristian said:

May 18, 2016 21:12:14 GMT Adrian said:

I'll post my WM plan(s) soon. I prefer to deal with Sandwell and Birmingham on their own, although pairing them, with one Birmingham ward in a Sandwell seat (which is excusable on account of the fact that it would equalise the size of seats in the two areas), produces quite a nice arrangement of seats, albeit with several split wards.

Given that the border between Birmingham and Sandwell is somewhat arbitrary in places with a lot of people in Sandwell thinking of themselves as Brummies, crossing that border isn't necessarily a big deal.

Hmm. I think quite a few people in Newton regard themselves as Brummies, but I'm not convinced that's true of anywhere else in Sandwell, even in Warley.

As for the "arbitrary" nature of boundaries, we're not going to see eye-to-eye on this. Boundaries are where they are for a variety of reasons, and in some cases the original positioning was indeed somewhat arbitrary. But that doesn't matter. The boundaries are a plain fact, and your and my opinion of their importance is of no consequence.

[YL]

May 18, 2016 20:11:08 GMT islington said:

First question: If the Birmingham boundary is so sacred, why isn't the same respect accorded to the Sandwell boundary? This is an especially pertinent question in view of the fact that Sandwell, unlike Birmingham, can actually be divided into a whole number of seats without ward-splitting. (There are several ways of doing it; for instance, your plan above could be adapted by combining the Sandwell bits of your two boundary-crossing seats above into 'Warley and Rowley Regis' (71631).) It's true that you then run straight into the Stourbridge problem as previously discussed; but if it's acceptable to solve this problem by crossing the Sandwell boundary, as in your plan, why is it  less acceptable to solve it by crossing the Birmingham boundary?

Your plan also solves the "Stourbridge problem" by crossing the Sandwell boundary. The question of how to solve it is basically one of finding an alternative partner for Halesowen, and you also had a Halesowen/Warley link. You could indeed put Halesowen with parts of Birmingham instead, but your crossing of the Birmingham boundary is to do with your desire not to split Birmingham's wards, not Stourbridge.

Also, I think boundary crossing seats should be considered on their merits. A seat which contains coherent components on either side of a boundary is generally preferable to one which is largely on one side and grabs a ward from the other side to make up the numbers, especially if the grabbed ward clearly belongs on its side of the boundary.

May 18, 2016 21:12:14 GMT Adrian said:

As for the rest of YL's plan, there are some good features, and a couple of obvious problems, i.e. the splitting of Tipton and the butchering of Walsall.

Tipton is the sort of mistake I knew I was going to make. I think it can be fixed relatively easily at the cost of moving the boundary between the two West Brom seats more into the town centre (e.g. swapping West Brom Central and Princes End) but there may be better options.

I found Walsall hard. At least with that combination of Wolverhampton wards in the cross-border seat, I couldn't find an obviously better configuration, at least not using whole wards; there was one with a neater looking Walsall seat, but it split Bloxwich. There are presumably alternatives with different ways of crossing the Wolverhampton/Walsall boundary. I'll wait to see your map.

[East Anglian Lefty]

Alternatively, how about this:

STOURBRIDGE and DUDLEY SOUTH as in the initial plan

HALESOWEN & ROWLEY REGIS (76753) - loses Tividale, gains St. Thomas

SMETHWICK (78468) - loses Newton and West Bromwich Central, gains Oldbury and Tividale

SANDWELL NORTH (72775) - compared to Wednesdbury, loses Great Bridge, Oldbury, Princes End and Tipton Green, gains Charlemont & Grove Vale, Great Barr with Yew Tree, Newton and West Bromwich Central

DUDLEY NORTH & TIPTON (78019) - loses St. Thomas and the Wolverhampton wards, gains Great Bridge, Princes End and Tipton Green

St. Thomas in with Halesowen is ugly, but everything else is reasonably clean. Names can be argued about.

For Walsall and Wolverhampton, I personally prefer adding Wednesfield to Walsall rather than Bilston. So WOLVERHAMPTON WEST and ALDRIDGE-BROWNHILLS as I had them; WOLVERHAMPTON EAST (78223) loses Wednesfield and gains Spring Vale and Ettingshall; WALSALL SOUTH (73172) gains Blakenall and Birchills-Leamore; WALSALL NORTH AND WEDNESFIELD (72512) is a little stringy but not excessively so.

[islington]

May 19, 2016 19:53:15 GMT YL said:

May 18, 2016 20:11:08 GMT islington said:

First question: If the Birmingham boundary is so sacred, why isn't the same respect accorded to the Sandwell boundary? This is an especially pertinent question in view of the fact that Sandwell, unlike Birmingham, can actually be divided into a whole number of seats without ward-splitting. (There are several ways of doing it; for instance, your plan above could be adapted by combining the Sandwell bits of your two boundary-crossing seats above into 'Warley and Rowley Regis' (71631).) It's true that you then run straight into the Stourbridge problem as previously discussed; but if it's acceptable to solve this problem by crossing the Sandwell boundary, as in your plan, why is it  less acceptable to solve it by crossing the Birmingham boundary?

Your plan also solves the "Stourbridge problem" by crossing the Sandwell boundary. The question of how to solve it is basically one of finding an alternative partner for Halesowen, and you also had a Halesowen/Warley link. You could indeed put Halesowen with parts of Birmingham instead, but your crossing of the Birmingham boundary is to do with your desire not to split Birmingham's wards, not Stourbridge.

Also, I think boundary crossing seats should be considered on their merits. A seat which contains coherent components on either side of a boundary is generally preferable to one which is largely on one side and grabs a ward from the other side to make up the numbers, especially if the grabbed ward clearly belongs on its side of the boundary.

YL -

Yes, but your criticism of my plan was that I crossed the Birmingham boundary. So you've submitted a plan that respects the Birmingham boundary. Fair enough: but what I am asking is, if the Birmingham boundary is so important that it mustn't be crossed even at the price of multiple ward splits, why is the Sandwell boundary less important even though it can be respected without any ward splits at all?

As for my reasons for crossing the Birmingham boundary, you're partly right. Avoiding ward splits was a major factor and, given the rules, a perfectly legitimate thing for me to take into account (alongside other factors, of course). But another major reason, equally important if not more so, is that I felt it helped create a better map across the West Midlands area more generally. 

Crossing the Birmingham boundary is not ideal, I agree. But on the other hand, my plan treats the W Mids metropolitan area as a unit, avoiding any cross-border seats spilling into neighbouring counties. Apart from being an advantage in itself, this has important benefits for the rest of the region. Specifically, it means there is no need to reinforce Coventry with bits of Warwickshire, thus avoiding couplings (Coleshill / Coventry West or Kenilworth / Coventry South) that would be bound to be unpopular.

With an entitlement of 5.37, of course, Warwickshire has to be paired with somewhere. But Tamworth borough, by reason of its shape, location, and entitlement (0.76) couldn't be better suited if it had been designed on purpose. Treating it with Warwickshire allows a 6-seat plan which, I admit (since I'm owning up to drawbacks), involves a hair-raisingly close shave for the town of Rugby; but apart from that, results in a very satisfactory map and, in particular, a much better treatment of the Nuneaton/Bedworth area than is possible if you reinforce Coventry with part of Warwickshire (it doesn't matter whether you use Kenilworth or Coleshill - Nuneaton/Bedworth seems to be a problem either way). Moreover, treating Tamworth with Warwickshire comes with a bonus in Staffordshire of allowing the Lichfield seat to be coterminous with the borough, an example of respect for LA boundaries that I don't think can be achieved by any other approach.

And this is my point really. I'm not ashamed to acknowledge that I'm trying to avoid ward splits because I think this is an important factor (alongside other important factors): but in the end, you have to look at the overall map.

[islington]

May 18, 2016 22:10:49 GMT East Anglian Lefty said:

But I can't agree that it's better to cross the city boundary on multiple occasions than to split a ward. Local authority boundaries are, according to the legislation, at least as important a criterion to consider as wards, and they're considerably more permanent. On a practical level, the ward split I'm proposing is right in the city centre, where there will be a high degree of churn and transience and hence ward boundaries will be less important to the locals anyway.

EAL -

Well, yes and no. It depends how you assess the concept of respecting LA boundaries.

And let me be clear: I do agree with you that, all else being equal, it's better to have a single seat crossing the boundary and obviously, by this measure, my plan, with four seats crossing the Birmingham boundary, doesn't score very well.

But it's not the only possible measure. I have 38 Birmingham wards out of 40 in seats that are wholly or mainly in Birmingham: this means that 94.7% of all Brummies will find themselves in a Birmingham seat. This may not be much consolation to the other 5.3%, including Adrian; but it's still a pretty good proportion overall. And is this not an equally legitimate way of looking at it?
 

[greenchristian]

May 20, 2016 16:09:08 GMT islington said:

With an entitlement of 5.37, of course, Warwickshire has to be paired with somewhere. But Tamworth borough, by reason of its shape, location, and entitlement (0.76) couldn't be better suited if it had been designed on purpose. Treating it with Warwickshire allows a 6-seat plan which, I admit (since I'm owning up to drawbacks), involves a hair-raisingly close shave for the town of Rugby; but apart from that, results in a very satisfactory map and, in particular, a much better treatment of the Nuneaton/Bedworth area than is possible if you reinforce Coventry with part of Warwickshire (it doesn't matter whether you use Kenilworth or Coleshill - Nuneaton/Bedworth seems to be a problem either way). Moreover, treating Tamworth with Warwickshire comes with a bonus in Staffordshire of allowing the Lichfield seat to be coterminous with the borough, an example of respect for LA boundaries that I don't think can be achieved by any other approach.

In the Coventry/Kenilworth scenario, the Bedworth and North Warwickshire seat is probably slightly more coherent than it is currently (consisting of Bedworth and the entire North Warwickshire council, rather than Bedworth and most of North Warwickshire council), and the only problem with the Nuneaton seat is that it adds in some rural areas of Rugby council area.

The Coventry/Coleshill map does take actual parts of Nuneaton town out of Nuneaton and actual parts of Rugby town out of Rugby, but doesn't take a huge amount out of either.

Also, the Coventry/Meriden alternative could be just as controversial as Coventry/Kenilworth or Coventry/Coleshill. Go further into Solihull council than Meriden ward (which you will need to to make up the numbers), and you're well past the areas where you could reasonably claim any community ties to Coventry. Which is not the case with either of the Coventry/Warwickshire plans.

[Adrian]

May 20, 2016 16:09:08 GMT islington said:

With an entitlement of 5.37, of course, Warwickshire has to be paired with somewhere. But Tamworth borough, by reason of its shape, location, and entitlement (0.76) couldn't be better suited if it had been designed on purpose. Treating it with Warwickshire allows a 6-seat plan which, I admit (since I'm owning up to drawbacks), involves a hair-raisingly close shave for the town of Rugby; but apart from that, results in a very satisfactory map and, in particular, a much better treatment of the Nuneaton/Bedworth area than is possible if you reinforce Coventry with part of Warwickshire (it doesn't matter whether you use Kenilworth or Coleshill - Nuneaton/Bedworth seems to be a problem either way).

I'm struggling to see how this is possible without either splitting a ward or splitting Rugby.

[Adrian]

Here's my current version of Warks/Cov/Solihull.

Solihull needs a polling district from Meriden. I'm not prepared to swap Blythe and Elmdon wards - it'd mean taking a big bite out of the town.

I'd add the Whitley part of the Cheylesmore ward to Coventry East seat to bring down the size of the Cov S & Ken seat - hopefully there's enough electors there to do the trick.

(Or, without needing ward splits, move Stoneleigh to Warwick and Radford to Rugby.)

I don't see there being any great outcry at pairing south Coventry with Kenilworth - the links are very good, thanks partly to the existence of the university.

[YL]

islington: my criticism of your plan is not that it crosses the Birmingham boundary, but
(a) that it does so in a bad way (that Oscott/Perry Barr arrangement) and
(b) that it is primarily doing so for what I think is a bad reason (putting too high a priority on not splitting wards, especially wards which are about to be redrawn anyway).

You have at least given some other reasons, and while I don't find them particularly convincing (I don't see anything wrong with Coventry South & Kenilworth given that Coventry can't have three seats of its own, and you're unnecessarily crossing the Staffordshire border as well as the Birmingham one) I'll accept there's a case to be made.

[East Anglian Lefty]

May 20, 2016 16:27:48 GMT islington said:

May 18, 2016 22:10:49 GMT East Anglian Lefty said:

But I can't agree that it's better to cross the city boundary on multiple occasions than to split a ward. Local authority boundaries are, according to the legislation, at least as important a criterion to consider as wards, and they're considerably more permanent. On a practical level, the ward split I'm proposing is right in the city centre, where there will be a high degree of churn and transience and hence ward boundaries will be less important to the locals anyway.

EAL -

Well, yes and no. It depends how you assess the concept of respecting LA boundaries.

And let me be clear: I do agree with you that, all else being equal, it's better to have a single seat crossing the boundary and obviously, by this measure, my plan, with four seats crossing the Birmingham boundary, doesn't score very well.

But it's not the only possible measure. I have 38 Birmingham wards out of 40 in seats that are wholly or mainly in Birmingham: this means that 94.7% of all Brummies will find themselves in a Birmingham seat. This may not be much consolation to the other 5.3%, including Adrian; but it's still a pretty good proportion overall. And is this not an equally legitimate way of looking at it?
 

It may be an equally legitimate way of looking at it. But with one ward split I've managed to get that up to 39 out of 40 and removed three of the boundary crossings. I think you've done a sterling job of showing what can be done without ward splits, but what can be done still isn't very nice, so I'm pushing to see what you can improve with a minimal number of ward splits.

[rocky]

There,is really no issue with crossing the Birmingham border. In the soho area no one would mind a cross border seat. In the Solihull/hall green area ( although not needed) or in the Halesowen Quinton Bartley green area as well areas could easily cross, likewise some kind of great Barr seat could work.

However the Erdington Perry Barr seat suggested by a few is just crazy and makes no community sense at all.

Reality is Birmingham is so big it often more sense in local communities to join with areas in different authorities that have more in common with them.

[islington]

This Great Birmingham Debate has been fascinating and I feel that we may be moving to some sort of resolution of it: not by arriving at a shared plan that we're all happy with, which is something that it is neither feasible nor (fortunately) necessary to achieve; but by having a shared understanding of the issues involved and the different ways of approaching them.

In that spirit, then, let me advance some thoughts arising from playing with Birmingham on BA and trying various approaches. And because this is a theoretical exercise, I'm allowing myself to look at Brum in isolation; if I were putting forward a more serious plan, I'd need to take account of the impact elsewhere because (as I keep saying) it's the whole map that matters.

Essentially, Brum is a trade-off between ward splits and crossing the boundary. If you want to avoid crossing the boundary at all, you need multiple ward splits - at least three. The reasoning goes like this. There are 40 wards of roughly similar size (14006 - 19255). They can be grouped into nine seats by creating one seat with five wards (which we know we can do with the smaller-than-average wards in the Edgbaston/Ladywood area), two with four wards (lots of options here - presumably Sutton Coldfield and one other), and bunching the remaining 27 wards into three sets of nine each. Then we divide one ward in each of the three sets, and this allows us to create six seats of four and a half wards each. QED.

Only it's not quite so easily done, once you try it. If you create a sensible five-ward seat by adding Selly Oak to the existing Edgbaston seat, you find that any possible nine-ward combination on the south side of the city is too big for two seats. I rescued this by shuffling the five-ward seat north, to take in Ladywood and Soho from the city centre and thus free up the below-average wards of Selly Oak and Bartley Green to reduce the average size of the nine wards in the southern combo, which then comes in at 154605. I went for my other four-ward seat with Hall Gn, Springfield, Acock's Gn, S Yardley (all above average size) and then combined the remaining two wards of Yardley, the four of Hodge Hill and three of Erdington (all except Kingstanding) for a nine-ward set (152588) and the rest of the city for the final set (154830). All that remains is to decide which wards to split.

I stress this is just a first stab; I was just trying to see whether a three-split version could be done, and I wasn't worrying too much about whether it made sense on the ground. I'm sure other combinations are possible, but what I'm not sure of is whether any arrangement will result in better seats, overall, that can be achieved by the alternative versions with one split (EAL) or no splits (me).

If you don't mind the occasional ward split but think three are too many, you can go for something like EAL's one-split option. But if you do this, you need to allow more wards to be gathered in groups of four. This isn't a problem in itself because it leads to quite natural groupings, but because they are all below average size they need to be balanced by a further five-ward seat in addition to the one in the Edgbaston area. And there's only one possible way of doing this, so far as I can see, which is why EAL's plan, like my no-split version, adds Perry Barr ward to the Erdington seat. Rocky says it's 'just crazy and makes no community sense at all' and she may well be right; but what else is there?

Or, of course, there's always my no-split option if you're willing to take more liberties with the city boundary.

There's no perfect solution to this. All these plans are workable, but each of them (including mine) has serious shortcomings. There are perfectly good arguments for and against adopting any of them.

(And just to reiterate the point that if you're making a real-world choice, you also need to look at the implications outside Birmingham of a decision whether (and if so, exactly how) to cross the city boundary.)

[East Anglian Lefty]

The Erdington and Perry Barr seat is undoubtedly bad. The problem is that I can't see any other non-split options in north Birmingham, assuming you leave Sutton Coldfield unchanged (almost a certainty) and don't bring in bits of north Solihull (which looks as bad as Erdington and Perry Barr.)

The follow-up question is what options you've got if you are willing to split a ward. The 5 wards between Sutton Coldfield and the M6 have a combined electorate of 80,035, so it wouldn't take a big split to make them legal. You can either add a few thousand voters to Sutton Coldfield (plenty of options there), or you can try to find somewhere to carve out in the south that doesn't cause too many problems.

That allows you to cross the Birmingham-Black Country boundary somewhere more sensible (Newton? Soho?), but the question is then how many splits you end up requiring. I reckon if it's possible to draw a map with no more than two splits in the West Midlands (one in north Birmingham, one elsewhere) and with no seat obviously sillier than Erdington and Perry Barr, you should do it. But is it possible? I will try to find out, but other attempts would be interesting to see.

[Adrian]

May 21, 2016 18:42:14 GMT East Anglian Lefty said:

The Erdington and Perry Barr seat is undoubtedly bad. The problem is that I can't see any other non-split options in north Birmingham, assuming you leave Sutton Coldfield unchanged (almost a certainty) and don't bring in bits of north Solihull (which looks as bad as Erdington and Perry Barr.)

The follow-up question is what options you've got if you are willing to split a ward. The 5 wards between Sutton Coldfield and the M6 have a combined electorate of 80,035, so it wouldn't take a big split to make them legal. You can either add a few thousand voters to Sutton Coldfield (plenty of options there), or you can try to find somewhere to carve out in the south that doesn't cause too many problems.

The options on the Erdington-Sutton border are actually quite limited. There'd be an outcry if any more areas south of the Chester Road are added to a Sutton seat. You could add the transroad side of PD CIF but it's probably not enough electors; you could also add the transroad side of DLF, i.e. move the border to the Chester Road as far as the Tyburn Road, but again this is a politically sensitive decision. And since Sutton is in quota, the Commission will probably quickly decide it's not worth the hassle.

My preferred option is to add PD CVE (3,000 voters) from Oscott to Perry Barr. This is the part of Perry Beeches that's split by the motorway, so this move wouldn't receive much local opposition.

[East Anglian Lefty]

OK, a quick bit of experimentation produces this, which requires one split in north Birmingham and one split in West Bromwich Central. Quite a few of the seats are the same as in previous attempts, so I've only bothered describing them when they're new creations. There are two crossings of the Birmingham boundary, one to take in Newton and one in the Soho area. You can produce much neater seats in the south and east of Birmingham if you add a third crossing into Rubery (following which you can swap King's Norton for Bournville, Hall Green for Springfield and Springfield for Bordesley Green) but I thought that might be pushing it. And ever local authority boundary in the Black Country is crossed bar Walsall-Wolverhampton. Overall, a curate's egg. I suspect there's room for improvement if you can get a decent seat crossing from Quinton to Halesown, but right now I haven't got the energy to consider the knock-on consequences for Stourbridge.



Sutton Coldfield (73172+x) - gains part of Tyburn/Erdington/Kingstanding. I suspect the former would be easiest.
Birmingham Erdington (80035-x) - gains Oscott, loses part of a ward
Birmingham Perry Barr (76201) - compares to the current seat, loses Oscott, gains Aston and Newton. Probably the least controversial place to cross a boundary in the West Midlands
Birmingham Hodge Hill (71165)
Birmingham Yardley (71150)
Birmingham Small Heath (72783)
Birmingham Brandwood (71357)
Birmingham Northfield (71729)
Birmingham Edgbaston (76801)
Smethwick & Ladywood (73978) - successor to Warley, loses Langley and Old Warley, gains Ladywood and Soho
Halesowen & Rowley Regis (75937) - loses Cradley Green South, gains Langley and Old Warley
Stourbridge (78320)
Dudley South (71054)
Dudley North & Bilston (78250) - as islington didn't like the version going into the centre of Wolverhampton, here's an alternative
West Bromwich West (80762-y) - gains Greets Green & Lyng and most of West Bromwich Central
Walsall South & Great Barr (74186+y) - loses Pheasey Park Farm and Darlaston, gains four Sandwell wards, plus that part of West Bromwich Central north of the A41 (or thereabouts)
Aldridge-Brownhills (75164) - gains Pheasey Park Farm and Blakenall
Walsall West (74300) - successor to North, loses Blakenall and gains Darlaston
Wolverhampton East (77239)
Wolverhampton West (74882)

[East Anglian Lefty]

May 21, 2016 19:37:03 GMT Adrian said:

May 21, 2016 18:42:14 GMT East Anglian Lefty said:

The Erdington and Perry Barr seat is undoubtedly bad. The problem is that I can't see any other non-split options in north Birmingham, assuming you leave Sutton Coldfield unchanged (almost a certainty) and don't bring in bits of north Solihull (which looks as bad as Erdington and Perry Barr.)

The follow-up question is what options you've got if you are willing to split a ward. The 5 wards between Sutton Coldfield and the M6 have a combined electorate of 80,035, so it wouldn't take a big split to make them legal. You can either add a few thousand voters to Sutton Coldfield (plenty of options there), or you can try to find somewhere to carve out in the south that doesn't cause too many problems.

The options on the Erdington-Sutton border are actually quite limited. There'd be an outcry if any more areas south of the Chester Road are added to a Sutton seat. You could add the transroad side of PD CIF but it's probably not enough electors; you could also add the transroad side of DLF, i.e. move the border to the Chester Road as far as the Tyburn Road, but again this is a politically sensitive decision. And since Sutton is in quota, the Commission will probably quickly decide it's not worth the hassle.

My preferred option is to add PD CVE (3,000 voters) from Oscott to Perry Barr. This is the part of Perry Beeches that's split by the motorway, so this move wouldn't receive much local opposition.

OK, that's interesting, and does facilitate a decent Halesowen-Quinton seat. Unfortunately it seems likely to create a bit of a car crash in south-western Sandwell, so I'm going to think a bit more on that before I post any possible maps based on that.