|
|
Post by 🏴☠️ Neath West 🏴☠️ on Oct 28, 2020 13:21:14 GMT
Given what it's like generally, the fact that it's solution for Lancaster is actually more sensible than the current boundaries is slightly disturbing The seat immediately to the south of that one though... |
|
|
|
Post by 🏴☠️ Neath West 🏴☠️ on Oct 28, 2020 13:23:42 GMT
I'm not sure what's worse - that or "Tees Banks" including Redcar and Billingham. Has anyone worked out what's going on with the Oswestry pocket? Or did he just hit an OUT OF CHEESE error? Is it perhaps a detached part of that Chester South thing? |
|
|
|
Post by greenchristian on Oct 28, 2020 14:12:49 GMT
It appears that the algorithm has a predisposition to leave any entire district that fits within quota alone, and as many of us will know that often leads to much worse knock-on effects elsewhere. It hasn't left Coventry alone, even though the figures we were working with on boundary assistant allowed it to stand alone. |
|
bsjmcr
Non-Aligned
Posts: 1,591 
|
Post by bsjmcr on Oct 28, 2020 15:40:57 GMT
Greater Manchester:
Some positives
(Manchester) Wythenshawe and Cheadle could work, I've not realised how contiguous they were till looking at the map - if the BC is getting bored with joining with Sale East. Sale is still split though, with West going with Urmston and Irlam, which is also not too dissimilar from 'Stretford and Irlam' is something sometimes recommended on here. Here though we have Stretford and Sale East.
Didsbury and Stockport North - I had great difficulty keeping the Didsburies together as M'Withington is too big. This could well happen, as pairing with Wythenshawe makes it slightly too big, and splitting Didsbury is both wrong and results in tenuous links.
Salford North and Swinton - not bad
Manchester East and Ashton West. Almost there, just a bit of tinkering to put Failsworth, Droylsden together with some of the Eastern Manchester wards to make Manchester East would work. Even as it is Manc East and Ashton West doesn't sound too ridiculous given there used to be Knowlsley North and Sefton East, and Gateshead East and Washington West, etc...
Somehow Prestwich and Middleton doesn't look as ugly as the actually proposed version. Then again, it only has half of Middleton. So that should be Prestwich and Middleton West, and Heywood and Middleton East. It is still a chalk and cheese pairing IMHO, but the fact that a computer drew it up again makes it seem even more likely to return, despite BuryS being within the numbers.
and a couple from Merseyside:
Formby and Ormskirk? I can't say much about the community links, but anything to get rid of the Sefton Central name must be a good thing?
Halewood and Huyton - keeping up the alliterative names now that Blackley and Broughton would be no more.
And the absurd:
Manchester City Centre and Blackley is quite something, the way the CC is hanging off.
Radcliffe West, Kearsley and Bury North. Then again, the old "Heywood and Royton" also snaked around Rochdale did it not?
Salford Quays, Eccles and Worsley. Reminiscent of the long Eccles seat that went from Irlam to Swinton, except that didn't split towns. Walkden is split, joining L'Hulton and half of Leigh and Warrington!
Ramsbottom, Darwin and Ribble Valley??? Mid and South Lancashire?
What did Wigan do wrong to deserve to be split 5 ways? At least all of them would be Labour, should Burnham want to return in Leigh, he just needs to pick East or West...
|
|
YL
Non-Aligned
Either Labour leaning or Lib Dem leaning but not sure which
Posts: 4,903
|
Post by YL on Oct 28, 2020 16:02:54 GMT
It is easy to mock of course, and there is a great deal to mock here, but I think we should also applaud Electoral Calculus for at least coming up with an attempt at an automated redistricting algorithm and putting the results of their alpha version out for all to see. It does raise a number of questions about how such an algorithm could be improved and whether it could assist the boundary drawing process. If I had more energy at the moment I'd actually be quite interested in playing with the algorithm to see whether other criteria could reasonably be incorporated and how much that would improve the results. Perhaps the first improvement, though, would be to just modify the input map so that wards on opposite sides of estuaries aren't considered as bordering each other. |
|
|
|
Post by greenchristian on Oct 28, 2020 17:16:49 GMT
Is there a way to manually adjust the zoom on their map? I can't see one, and it's a right pain trying to navigate around the map without it.
|
|
|
|
Post by Defenestrated Fipplebox on Oct 28, 2020 17:23:31 GMT
Is there a way to manually adjust the zoom on their map? I can't see one, and it's a right pain trying to navigate around the map without it.
Putting in a postcode zooms it in. Clicking on the map after that zooms it in further, no way to zoom it out again as far as I can see.
|
|
|
|
Post by greenchristian on Oct 28, 2020 17:43:22 GMT
Is there a way to manually adjust the zoom on their map? I can't see one, and it's a right pain trying to navigate around the map without it.
Putting in a postcode zooms it in. Clicking on the map after that zooms it in further, no way to zoom it out again as far as I can see.
And what I want to do is zoom out from the postcode view to a sensible zoom for scrolling around the map.  |
|
|
|
Post by greatkingrat on Oct 28, 2020 17:58:45 GMT
If you have a mouse wheel you can zoom in and out with that.
|
|
ilerda
Conservative
Posts: 1,096
|
Post by ilerda on Oct 28, 2020 18:04:25 GMT
It appears that the algorithm has a predisposition to leave any entire district that fits within quota alone, and as many of us will know that often leads to much worse knock-on effects elsewhere. It hasn't left Coventry alone, even though the figures we were working with on boundary assistant allowed it to stand alone. I meant where a district can be kept as one entire constituency without any division. Coventry council area isn’t within quota so it has to split into 3 parts, whereas Rugby, Hartlepool, Cannock Chase etc don’t. |
|
|
|
Post by therealriga on Oct 28, 2020 18:45:47 GMT
If you have a mouse wheel you can zoom in and out with that. Mouse pad works just fine for zooming out. |
|
|
|
Post by islington on Oct 28, 2020 19:00:28 GMT
Do we know what electorate numbers were used for the EC exercise? Were they the Dec 2019 figures as on Boundary Assistant? Or did EC have access to the Mar 2020 numbers on which the real exercise will be based?
I'm asking because it was reported somewhere upthread that the Mar 2020 figures would be available in late October, which is more or less where we are now.
(Asking for a friend, obviously.)
|
|
YL
Non-Aligned
Either Labour leaning or Lib Dem leaning but not sure which
Posts: 4,903
|
Post by YL on Oct 28, 2020 19:14:36 GMT
Do we know what electorate numbers were used for the EC exercise? Were they the Dec 2019 figures as on Boundary Assistant? Or did EC have access to the Mar 2020 numbers on which the real exercise will be based? I'm asking because it was reported somewhere upthread that the Mar 2020 figures would be available in late October, which is more or less where we are now. (Asking for a friend, obviously.) I suspect it's Electoral Calculus's own projections of the figures. They have used new wards in some areas where we don't have the figures for them, e.g. Rotherham: you can see the new Bramley & Ravenfield ward in that weird Ecclesfield to west Maltby thing. (I'd say "bizarre", but given what appears in some other parts of the country the milder "weird" seems more appropriate.) |
|
ilerda
Conservative
Posts: 1,096
|
Post by ilerda on Oct 28, 2020 22:06:55 GMT
I think the overall weirdness and illogical nature of some of the boundaries comes from the fact that the algorithm has been instructed to create constituencies as close to the quota as possible.
That means that sensible combinations of wards that we would likely have come up with, and which are within tolerance but not necessarily close to the quota, don't get picked because there is a more illogical combination that is closer to the quota. That is the top priority for the algorithm.
|
|
Khunanup
Lib Dem
Portsmouth Liberal Democrats
Posts: 11,997
|
Post by Khunanup on Oct 29, 2020 1:41:35 GMT
The computer is currently being destroyed by a pitchfork wielding army from the Wirral. And Hayling Island... This is batshit craziness. Andover, & Fareham town being split between three seats, Mersey Banks+++, the horror show of Humber Banks x2. Martin Baxter, I never thought I'd say this, but you've outdone yourself! |
|
|
|
Post by evergreenadam on Oct 29, 2020 13:27:59 GMT
Do we know what electorate numbers were used for the EC exercise? Were they the Dec 2019 figures as on Boundary Assistant? Or did EC have access to the Mar 2020 numbers on which the real exercise will be based? I'm asking because it was reported somewhere upthread that the Mar 2020 figures would be available in late October, which is more or less where we are now. (Asking for a friend, obviously.) The webpage has been set up for the Mar 2020 statistics. Keep checking for updates! www.ons.gov.uk/releases/electoralstatisticsukmarch2020 |
|
|
|
Post by therealriga on Oct 29, 2020 14:55:17 GMT
I think the overall weirdness and illogical nature of some of the boundaries comes from the fact that the algorithm has been instructed to create constituencies as close to the quota as possible. Which almost seems to be a misunderstanding on the part of the developer. It used to be the case that constituencies were supposed to be as close to the quota as possible but that's no longer the case. Under the old legislation a constituency 0.5% above quota was better than one 3% above quota, now a constituency 4.5% above is as good as one 0.1% above quota. As the BCE itself said: "Legislation does not require the commissions to achieve constituency electorates that are as close as possible to the electoral quota figure, and we do not consider it appropriate to adopt such a policy objective" |
|
Harry Hayfield
Green
Cavalier Gentleman (as in 17th century Cavalier)
Posts: 2,922
|
Post by Harry Hayfield on Oct 29, 2020 15:44:33 GMT
Do we know what electorate numbers were used for the EC exercise? Were they the Dec 2019 figures as on Boundary Assistant? Or did EC have access to the Mar 2020 numbers on which the real exercise will be based? I'm asking because it was reported somewhere upthread that the Mar 2020 figures would be available in late October, which is more or less where we are now. (Asking for a friend, obviously.) The webpage has been set up for the Mar 2020 statistics. Keep checking for updates! www.ons.gov.uk/releases/electoralstatisticsukmarch2020And note "Provisional release date: January to February 2021" which I don't think is in the bill currently passing through the Lords |
|
|
|
Post by evergreenadam on Oct 29, 2020 16:15:23 GMT
And note "Provisional release date: January to February 2021" which I don't think is in the bill currently passing through the Lords The provisional release date has certainly been altered since the webpage was initially published. It originally said Oct/Nov 2020. |
|
|
|
Post by John Chanin on Oct 29, 2020 16:57:34 GMT
Remember that councils with new wards brought into force by parliament prior to 1 December will have to calculate electorates for the new wards as at March 2020. This isn’t a very difficult job, but it will have to be done.
The operative dates for wards and electorates are irrelevant to the process of passing the bill. Whenever this is the operative dates will stay the same. However ONS may delay asking for the data if there is a delay in bringing the Act into force. This really doesn’t matter. The Boundary Commission has years to produce its recommendations. As someone says upstream the legislation gives a date for the end of the process not the start.
|
|
