Dec 2019 electorates

[YL]

Finally, (g): Leeds, North Yorkshire and the unfortunate Askern. Two maps here, one covering the whole area and one a zoom in on Leeds.





Two wards are split in Leeds. I should acknowledge that Wetherby & Boroughbridge, as well as being cross county, contains parts of four districts, but I think it's reasonably coherent, along the A1 corridor, as such seats go. The other North Yorkshire seats are reasonably closely based on the current ones, with the York seats and Scarborough & Whitby essentially unchanged.

1. Selby & Askern 2. Wetherby & Boroughbridge 3. York Central 4. York Outer 5. Leeds North West 6. Pudsey 7. Leeds South & Morley 8. Leeds Central 9. Leeds East & Rothwell 10. Leeds North East 11. Leeds West 12. Skipton & Ripon 13. Harrogate & Knaresborough 14. Richmond (Yorks) 15. Thirsk & Malton 16. Scarborough & Whitby.

(That's eight split wards in the region: three in Kirklees, two in Leeds, and one each in Sheffield, Wakefield and Calderdale.)

[YL]

Aug 7, 2020 16:00:50 GMT Adrian said:

YL ilerda Though your Sheffield plans look okay, I'm not sure why you both feel the need to smash Rotherham. It's possible and preferable to create a Rotherham seat very similar to the current one.

I think the town together with its fringe is large enough that being covered by two seats isn't too bad; indeed edgbaston explicitly suggested a north/south split some way back in this thread. In the Ecclesfield option, the problem was that I didn't want to leave Ecclesfield too much out on a limb, which meant Keppel ward got included. But if you're not worried about that, you could do (in terms of the new wards):

Wentworth & Ecclesfield: Ecclesfield wards, Hoober, Wath, Swinton wards, Rawmarsh wards (no need to include the part of the old Silverwood east of the Don, so this isn't too bad apart from the weak link to Ecclesfield)
Rotherham: Keppel, Greasbrough, Rotherham W, Boston Castle, Rotherham E, Dalton & Thrybergh, Sitwell, Brinsworth, Rother Vale
Rother Valley: the rest of the borough

In the Mosborough option, I can't find a plan I like that maintains something close to the current Rotherham seat: it seems to me like it's going to end up either virtually cutting the Wentworth seat into two or splitting Aston-cum-Aughton. But perhaps I've missed something.

[J.G.Harston]

I'm sure I posted a Sheffield+Pensistone map, but I've scrolled back through the last 20 pages and can't find it. I'll see if I can remember it and compare it to the ones recently posted.

[islington]

Am I alone in finding some irony in the fact that both Adrian's and YL's ward-splitting plans exclude Wakefield S from their Wakefield seats (as indeed does the current map); whereas in my non-split plan all four wards of Wakefield proper were kept together?

If we're going to split wards, can't we at least get a town like Wakefield wholly in its own seat?

[islington]

Aug 7, 2020 16:16:06 GMT Adrian said:

Aug 7, 2020 13:01:41 GMT islington said:

Those involved in drawing up ward boundaries must be aware that it is long-standing practice to use wards as building blocks for Parliamentary seats. I'm not saying this the primary purpose of wards - of course not - but it's a significant secondary consideration surely. Sheffield's an example: I believe I'm right in saying (and if not, there are posters in this forum who will rapidly correct me) that the reason Sheffield has 28 wards is to make it easier to draw up the city's anticipated allowance of five-and-a-half constituencies. And if this is right, I imagine it is not the only case where wards were drawn with at least half an eye on their future use for Parliamentary boundaries.

You're right that the LGBCE does sometimes allocate a number of wards that it believes will be helpful to the BCE. But this usually proves to be a fool's errand. Taking the Sheffield example, is it actually straightforward to create a non-split plan based on these supposedly helpful wards? Such a policy by the LGBCE would only be effective if theirs and the BCE's rules were in sync, which they're not.

I'd say it is helpful. It allows us to draw four non-split seats wholly within Sheffield. It's true it breaks down in the east of the city, because the wards there tend to be somewhat smaller, but the shortfall of electors can be resolved fairly easily by borrowing a single Rotherham ward.

This leaves one further seat with the odd three Sheffield wards attached to parts of an adjoining authority, but this was always priced in to the 28-ward plan.

What it does suggest, though, is that the odd three wards should not be taken from the north of the city because in this area the wards are a bit bigger and it's possible to draw 5-ward seats that are within range (as the 28-ward plan envisaged). It makes pragmatic sense, therefore, to take the three seats from the eastern side despite the reservations some have expressed about a 'Rother Valley' seat. (Would it be less odious, I wonder, if instead it were called 'Sheffield Mosborough and Kiveton Park', or some such name?)

Anyway, none of this alters my central point, which is that wards are drawn in the knowledge that they are likely to form the basis for Parliamentary boundaries. 

[YL]

Aug 8, 2020 11:55:01 GMT islington said:

Am I alone in finding some irony in the fact that both Adrian's and YL's ward-splitting plans exclude Wakefield S from their Wakefield seats (as indeed does the current map); whereas in my non-split plan all four wards of Wakefield proper were kept together?

If we're going to split wards, can't we at least get a town like Wakefield wholly in its own seat?

I'm sure we could, but as you know full well the rules say nothing about putting currently split communities back together, so it's hard to put much weight on doing so.

Unlike, say, Outwood, right next door, currently not split, not split in my plan or Adrian's, split in yours.

[YL]

Aug 8, 2020 10:56:02 GMT J.G.Harston said:

I'm sure I posted a Sheffield+Pensistone map, but I've scrolled back through the last 20 pages and can't find it. I'll see if I can remember it and compare it to the ones recently posted.

Possibly you were thinking of this.

[ilerda]

Aug 8, 2020 11:55:01 GMT islington said:

Am I alone in finding some irony in the fact that both Adrian's and YL's ward-splitting plans exclude Wakefield S from their Wakefield seats (as indeed does the current map); whereas in my non-split plan all four wards of Wakefield proper were kept together?

If we're going to split wards, can't we at least get a town like Wakefield wholly in its own seat?

There’s a problem here with the BCE’s lack of institutional memory.

I would imagine that Wakefield South was included in the current Hemsworth seat with some reluctance, but it was seen as necessary to avoid problems elsewhere and to make them fairly balanced.

Unfortunately what started as an imperfect solution to a problem can very easily become hard to undo because of the desire to maintain existing boundaries if that is plausible. What would be preferable would be to treat each review as a completely fresh start, although I find it hard to see that idea being met with any enthusiasm.

[J.G.Harston]

Aug 8, 2020 13:03:04 GMT YL said:

Aug 8, 2020 10:56:02 GMT J.G.Harston said:

I'm sure I posted a Sheffield+Pensistone map, but I've scrolled back through the last 20 pages and can't find it. I'll see if I can remember it and compare it to the ones recently posted.

Possibly you were thinking of this.

That is indeed the lupine lagomorph of interest. I'd forgotten I'd drawn an example of excluding the main part of Sheffield to show it could be dealt with on its own, and I hadn't actually tackled it.

On the subject, I've been updating my spreadsheet of polling district figures for 2020. I've got Broomhall & Sharrow Vale at 22,000, 52% over quota, which - as we're only one cycle into using the new wards - seems very suggestive of my figures being wrong. I suspect I transcribed something wrong back in March. Annoyingly, the library is still closed. I can see I'm going to have to email Election Services direct.

[Adrian]

islington Ironic? According to Alanis Morrisette's definition maybe.

I have always wanted to have all Wakefield wards in a Wakefield seat, and that's what I usually propose. But I have always said that redistricting is about balancing priorities. If the priority is keeping seats within LA boundaries, it's always going to be difficult to do this without dividing some of the communities in that LA. (See the 5th Review for plenty of examples, including Wakefield.)

I'd be happy to include Sandal in the Wakefield seat, but the numbers don't quite work. And rather than split a ward (or split Outwood or Ossett!) I decided it's okay to maintain the local ties established at the 5th Review.

[Adrian]

ilerda This is a tricky question. From the point of view of fairness, you're right that the map should be completely redrawn each time, since "minimum change" means that any bias in the map festers.

But MPs (and to some, lesser, extent, constituents) don't like change, and the continuity provided by "minimum change" is probably more beneficial than a scorched earth policy.

Anyway, once we *finally* have a 5% map, the potential for long-term bias in it is considerably reduced.

[YL]

Aug 8, 2020 13:24:42 GMT J.G.Harston said:

On the subject, I've been updating my spreadsheet of polling district figures for 2020. I've got Broomhall & Sharrow Vale at 22,000, 52% over quota, which - as we're only one cycle into using the new wards - seems very suggestive of my figures being wrong. I suspect I transcribed something wrong back in March. Annoyingly, the library is still closed. I can see I'm going to have to email Election Services direct.

I think this is probably more or less correct; is it a local government or Parliamentary figure?

The "December 2019" figure in Boundary Assistant is nearly 19000, out of just under 80000 in the existing Sheffield Central. However, the actual General Election electorate for Sheffield Central was nearly 90000. I'd expect that difference of 10000 to be concentrated in growing areas (e.g. parts of City) and areas with relatively transient populations, so 3000 extra within Broomhill & Sharrow Vale is not unexpected.

The thing is that the last ward review used numbers from just after IER was introduced, but before the University and the Council worked out how to get students registered under the new system. Hence there was a dip and then a recovery in the electorates of student areas with the review at pretty much the wrong time. You then get the extra registration surge associated with major elections and referendums.

[islington]

Aug 8, 2020 13:51:20 GMT Adrian said:

islington Ironic? According to Alanis Morrisette's definition maybe.

I have always wanted to have all Wakefield wards in a Wakefield seat, and that's what I usually propose. But I have always said that redistricting is about balancing priorities. If the priority is keeping seats within LA boundaries, it's always going to be difficult to do this without dividing some of the communities in that LA. (See the 5th Review for plenty of examples, including Wakefield.)

I'd be happy to include Sandal in the Wakefield seat, but the numbers don't quite work. And rather than split a ward (or split Outwood or Ossett!) I decided it's okay to maintain the local ties established at the 5th Review.

I'm not splitting Ossett; I have both Ossett and Horbury wards in  the same seat.

I am splitting Outwood, which I agree is unfortunate but it's not so bad, in my view, as splitting Wakefield. And I agree with ilerda's point that the current placing of Wakefield S in the Hemsworth seat is, presumably, an imperfect expedient adopted at the last review. And since seats in this area need to change anyway, because NP&C is too big, I thought it made sense to shift Normanton into Hemsworth. This (a) gets P&C (now devoid of its 'N') within range, and (b) creates an opportunity to unite Wakefield S with the rest of the town.

Why is dividing Outwood less of a problem than splitting Wakefield?

Wakefield is a long-established historic town, well known not only in its local area but throughout the country. Although not an ancient Parliamentary borough it was the centre of a large and important parish (that included Outwood) and it was of sufficient substance to have been one of the 41 English towns picked out to be new Parliamentary boroughs in 1832. It later functioned as the county town of the West Riding and it was a county borough in its own right until 1974.

Outwood prior to 1974 was a ward of Stanley UD.

Obviously I'd prefer to keep them both whole, but if the choice must be made then for me it must be Wakefield.

[ilerda]

Aug 8, 2020 13:58:39 GMT Adrian said:

ilerda This is a tricky question. From the point of view of fairness, you're right that the map should be completely redrawn each time, since "minimum change" means that any bias in the map festers.

But MPs (and to some, lesser, extent, constituents) don't like change, and the continuity provided by "minimum change" is probably more beneficial than a scorched earth policy.

Anyway, once we *finally* have a 5% map, the potential for long-term bias in it is considerably reduced.

Yes I agree it's tricky to find the balance between the two. Certainly MPs and parties have an interest in no chance, but I suspect the new rules mean we're going to get pretty substantial changes at every future review to the extent that we perhaps didn't before.

Personally I think the sweet spot is neither minimum change nor scorched earth. Where an existing seat can objectively be said to meet a lot of the discretionary factors such a community, LA boundaries etc. and is also within limits then I think there should be a higher burden of proof needed for any change proposed.

But where an existing seat is in someway 'flawed', like the situation with Wakefield South in Hemsworth, I think there should be presumption in favour of change at the following review.

[greenchristian]

Aug 8, 2020 16:41:48 GMT ilerda said:

Aug 8, 2020 13:58:39 GMT Adrian said:

ilerda This is a tricky question. From the point of view of fairness, you're right that the map should be completely redrawn each time, since "minimum change" means that any bias in the map festers.

But MPs (and to some, lesser, extent, constituents) don't like change, and the continuity provided by "minimum change" is probably more beneficial than a scorched earth policy.

Anyway, once we *finally* have a 5% map, the potential for long-term bias in it is considerably reduced.

Yes I agree it's tricky to find the balance between the two. Certainly MPs and parties have an interest in no chance, but I suspect the new rules mean we're going to get pretty substantial changes at every future review to the extent that we perhaps didn't before.

Personally I think the sweet spot is neither minimum change nor scorched earth. Where an existing seat can objectively be said to meet a lot of the discretionary factors such a community, LA boundaries etc. and is also within limits then I think there should be a higher burden of proof needed for any change proposed.

But where an existing seat is in someway 'flawed', like the situation with Wakefield South in Hemsworth, I think there should be presumption in favour of change at the following review.

It's worth noting that the rules as written only take account of community ties within existing seats, and not when they cross existing constituency boundaries (i.e. reviews should try to avoid splitting communities, but there is no obligation to try re-uniting split communities). This makes it more difficult - at least in theory - to fix obviously bad seats in subsequent reviews.

[J.G.Harston]

Aug 8, 2020 14:02:01 GMT YL said:

Aug 8, 2020 13:24:42 GMT J.G.Harston said:

On the subject, I've been updating my spreadsheet of polling district figures for 2020. I've got Broomhall & Sharrow Vale at 22,000, 52% over quota, which - as we're only one cycle into using the new wards - seems very suggestive of my figures being wrong.

I think this is probably more or less correct; is it a local government or Parliamentary figure?

The "December 2019" figure in Boundary Assistant is nearly 19000, out of just under 80000 in the existing Sheffield Central. However, the actual General Election electorate for Sheffield Central was nearly 90000. I'd expect that difference of 10000 to be concentrated in growing areas (e.g. parts of City) and areas with relatively transient populations, so 3000 extra within Broomhill & Sharrow Vale is not unexpected.

I've uploaded this year's figures, both local and parliamentary, to: mdfs.net/maps/Sheffield/Wards/ with a note of figures I didn't collect in March and have calculated. The figures I had were for wards in Sheffield Central and Sheffield Brightside.

The figures I've got for Broomhill are:
Broomhill local parl'y
Botanical EA 3320 3243
Bowood EB 1405 1361
Brocco Bank EC 899 875
Collegiate ED 2154 2068
Harcourt EE 3016 2914
Hunters Bar EF 2444 2361
Summerfield EG 1620 1503
Tapton Hill EH 1702 1643
Whitham EI 2502 2419
Broomhall EJ 3300 22362 54.7% 3083 21470 50.0%


With City Centre on 15775 local 14578 parliamentary electors.

The fact that there's a 'Broomhall' polling district in City Centre with an electorate of zero suggests that deviating from the ring road and putting Broomhall into Broomhill in the local review (the opposite of the initial proposal) was a mistake.

[J.G.Harston]

This is my first draft of a Sheffield+Penistone model. I really really want this to work with just one ward split along the Ecclesfield parish boundary, but the 2019 figures don't let it, and the 2020 figures which I've added on make it even worse.

SouthEast is too small, so needs to eat some of Heeley. Heeley is already too small, so needs to east something across ther Sheaf. The 2020 figures makes Central 13% too big.

I've been working around Wales and Catcliffe, and the Waverley development feels quite like an overspill from Handsworth, so I think I'll tackle a reincarnated Sheffield Hillsborough model and hive off bits in the southeast to Rotherham.

[East Anglian Lefty]

Regarding Wakefield, uniting the town in a single seat is obviously desirable. But the town isn't just the areas in the four Wakefield wards. There's a continuous line of development between Wakefield and Wrenthorpe, and bits of Wakefield Rural are clearly part of the contiguous urban areas.

The four Wakefield wards and the two Outwood wards are just slightly too small to stand alone (and adding rural makes it too large), but even if it wasn't then that would mean there's no sensible place to cross the Wakefield-Leeds boundary, so I think there's a very good case for just biting the bullet and accepting that Wakefield gets the short end of the stick.

[🏴‍☠️ Neath West 🏴‍☠️]

Aug 8, 2020 17:56:42 GMT J.G.Harston said:

I've uploaded this year's figures, both local and parliamentary, to: mdfs.net/maps/Sheffield/Wards/ with a note of figures I didn't collect in March and have calculated.

Thanks for this. On a complete tangent, what is the castle the Manor and Castle ward is partially named after? I can see the Manor Estate at one end, but the other end looks like it ought to be called Park Hill. (And this is the ward I most want to split!)

[J.G.Harston]

Aug 8, 2020 18:47:49 GMT 🏴‍☠️ Neath West 🏴‍☠️ said:

Aug 8, 2020 17:56:42 GMT J.G.Harston said:

I've uploaded this year's figures, both local and parliamentary, to: mdfs.net/maps/Sheffield/Wards/ with a note of figures I didn't collect in March and have calculated.

Thanks for this. On a complete tangent, what is the castle the Manor and Castle ward is partially named after? I can see the Manor Estate at one end, but the other end looks like it ought to be called Park Hill. (And this is the ward I most want to split!)

There uesed to be a ward at the western end that covered half the city centre and the site of Sheffield Castle under the markets, that ward was named Castle. There used to be a ward that covering the eastern end that covered the site of Sheffield Manor, that ward was called Manor. By 2002 both wards had fallen to about 60% of quota, so they were merged and tidied up. The new ward was Manor Castle, even though many people said it should be Manor *and* Castle to avoid perpetuating the common belief that the Manor is the "Manor Castle". It's not, it's the manor, the castle is something else.

The site of Sheffield Castle is now no longer in the ward, being about 200yds into City Centre, but the ward name 'Castle' persists in people's feeling for the name of the area, perpetuated in the habit of the council to name community things after the ward, so you'd have Castle Tenants' Centre for tenants in Castle Ward, not for any area itself called Castle, SOAR is the Southey Owlerton Area Regenation board, named after the wards, not named after the areas, Owlerton Ward no longer exists, and Owlerton village is long depopolated and tarmac'd under the A61.

I keep calling it Manor & Castle to try and pull away from people thinking there is a "Manor Castle". Manor & Castle can be quite easily split into the Manor half and the Castle half by drawing a line the City Road cemetary and along Manor Road or through Manor Wood, giving about 7500 in Castle and about 5500 in Manor.