|
|
Post by Pete Whitehead on Jan 4, 2020 22:36:52 GMT
|
|
🏴☠️ Neath West 🏴☠️
Conservative & Unionist
Party hats roasting on an open fire...
Posts: 3,971
Member is Online
|
Post by 🏴☠️ Neath West 🏴☠️ on Jan 4, 2020 22:52:00 GMT
Interesting (from an historical perspective) that the Lib Dems are now considerably weaker in the Isle of Ely than in Cambridgeshire proper.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jan 4, 2020 22:57:01 GMT
Would the Lib Dems have won the old South West Cambridgeshire seat? |
|
|
|
Post by Pete Whitehead on Jan 4, 2020 22:59:49 GMT
No - they are clearly weaker in the St Neots area than in those areas that replaced it
|
|
|
|
Post by bjornhattan on Jan 4, 2020 23:53:46 GMT
Huntingdon North remaining as resolutely Labour as ever - it must account for a decent chunk of Huntingdon constituency's Labour vote.
|
|
|
|
Post by Pete Whitehead on Jan 5, 2020 0:01:06 GMT
Huntingdon North remaining as resolutely Labour as ever - it must account for a decent chunk of Huntingdon constituency's Labour vote. I was actually slightly surprised it came out that way (and of course my figures could be totally wrong). It is a shit-hole of course, but also quite Brexity I imagine. It isn't large enough to account for that high a proportion of the Labour vote in the constituency even if everyone there had voted Labour and they would have had significant votes in the other towns - St Neots espcially but also St Ives (increasing Cambridge influence there). It's not impossible given the patterns of this election that Labour did a bit better than I have them in more middle-class liberal (big and small 'L') areas like Buckden and a bit less well in the sink estates of Huntingdon itself |
|
|
|
Post by evergreenadam on Jan 5, 2020 13:41:28 GMT
Berkshire would be quite good to see, particularly for Wokingham ward breakdown.
|
|
|
|
Post by akmd on Jan 5, 2020 16:01:42 GMT
Thanks for doing these maps, Pete. Would you mind doing Surrey and Hampshire at some point, please?
|
|
|
|
Post by Pete Whitehead on Jan 11, 2020 11:49:10 GMT
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jan 11, 2020 11:54:12 GMT
Berkshire would be quite good to see, particularly for Wokingham ward breakdown.     The Lib Dem performance in Wokingham was pretty disappointing. They were closer to winning Esher and Hitchin which is surprising because this was more of a Lib Dem target in the 2000s. But maybe Philip Lee running as an ex-Tory hurt the Lib Dems here. |
|
|
|
Post by evergreenadam on Jan 11, 2020 14:05:42 GMT
The Lib Dem performance in Wokingham was pretty disappointing. They were closer to winning Esher and Hitchin which is surprising because this was more of a Lib Dem target in the 2000s. But maybe Philip Lee running as an ex-Tory hurt the Lib Dems here. I suspect there was more tactical voting for the Lib Dems in the wards bordering SE Reading than the model reveals. The Lib Dems made very strong advances in 2019 local elections, whether that progress will be maintained at this year's local elections will be interesting to see, there are certainly a lot of marginal Tory wards being defended. |
|
|
|
Post by November_Rain on Jan 11, 2020 19:29:18 GMT
Thank you for these maps Pete Whitehead  They're very interesting! Don't suppose you have one for Bristol and North Somerset by any choice? I know according to one of our Labour councillors, the Conservatives won every council ward in the constituency of Weston-super-Mare even South ward (has always been safe Labour) which compiles Bournville and Oldmixon housing estates which means we would have been wiped out if it was a local election. |
|
|
|
Post by Pete Whitehead on Jan 11, 2020 20:28:55 GMT
Thank you for these maps Pete Whitehead They're very interesting! Don't suppose you have one for Bristol and North Somerset by any choice? I know according to one of our Labour councillors, the Conservatives won every council ward in the constituency of Weston-super-Mare even South ward (has always been safe Labour) which compiles Bournville and Oldmixon housing estates which means we would have been wiped out if it was a local election. I have done Avon for 2015 and 2017 which means it isn't too big a job to do it for the recent election. Its always an interesting area to do as well, so will be on the cards soon |
|
|
|
Post by November_Rain on Jan 11, 2020 20:29:54 GMT
Thank you for these maps Pete Whitehead They're very interesting! Don't suppose you have one for Bristol and North Somerset by any choice? I know according to one of our Labour councillors, the Conservatives won every council ward in the constituency of Weston-super-Mare even South ward (has always been safe Labour) which compiles Bournville and Oldmixon housing estates which means we would have been wiped out if it was a local election. I have done Avon for 2015 and 2017 which means it isn't too big a job to do it for the recent election. Its always an interesting area to do as well, so will be on the cards soon Thank you Pete Whitehead. I appreiciate what you are doing 100%. |
|
|
|
Post by andrewp on Jan 11, 2020 21:33:43 GMT
The Conservatives carrying every ward in Solihull and Dudley. Pete Whitehead which was the closest ward in Solihull borough? |
|
|
|
Post by Pete Whitehead on Jan 11, 2020 22:04:27 GMT
You can have the West Midlands though for now.. The Conservatives carrying every ward in Solihull and Dudley. Pete Whitehead which was the closest ward in Solihull borough?
On my numbers the Conservative lead in Smiths Wood is 3 votes. Chelmsley Wood is also close (but a couple of hundred). Obviously these wards are particularly difficult to gauge because of the Green strength in local elections in recent years
|
|
|
|
Post by afleitch on Jan 11, 2020 22:52:42 GMT
What is your method for calculating?
|
|
|
|
Post by nobodyimportant on Jan 12, 2020 19:16:18 GMT
The Lib Dem performance in Wokingham was pretty disappointing. They were closer to winning Esher and Hitchin which is surprising because this was more of a Lib Dem target in the 2000s. But maybe Philip Lee running as an ex-Tory hurt the Lib Dems here. I suspect there was more tactical voting for the Lib Dems in the wards bordering SE Reading than the model reveals. The Lib Dems made very strong advances in 2019 local elections, whether that progress will be maintained at this year's local elections will be interesting to see, there are certainly a lot of marginal Tory wards being defended. I'd definitely expect fewer Lib Dem votes in Barkham, Arborfield and the Western part of Wokingham constituency and more in Maiden Erlegh, Hillside and Shinfield North than shown on that map. Those three wards between them probably make up a large proportion of the Lab 2017 Lib Dem 2019 vote. Pete Whitehead How close did you have Twyford (in Maidenhead constituency) to being Lib Dem, incidentally? My gut feeling would have been to put both Con and Lib Dem in the same categories as you did, but the Tories near the bottom of theirs and the Lib Dems near the top, thus coming out Lib Dem overall. |
|
|
|
Post by Pete Whitehead on Jan 12, 2020 20:36:24 GMT
I suspect there was more tactical voting for the Lib Dems in the wards bordering SE Reading than the model reveals. The Lib Dems made very strong advances in 2019 local elections, whether that progress will be maintained at this year's local elections will be interesting to see, there are certainly a lot of marginal Tory wards being defended. I'd definitely expect fewer Lib Dem votes in Barkham, Arborfield and the Western part of Wokingham constituency and more in Maiden Erlegh, Hillside and Shinfield North than shown on that map. Those three wards between them probably make up a large proportion of the Lab 2017 Lib Dem 2019 vote. Pete Whitehead How close did you have Twyford (in Maidenhead constituency) to being Lib Dem, incidentally? My gut feeling would have been to put both Con and Lib Dem in the same categories as you did, but the Tories near the bottom of theirs and the Lib Dems near the top, thus coming out Lib Dem overall. I suspect you're right about the Early area having a higher Lib Dem vote and the West Berkshire wards a bit lower. There's only a certain extent to which I can defy the data though before it becomes a case of basically plucking numbers out of my arse. I certainly can and do finesse the results a little though if they don't 'feel' right or I'm aware of certain factors which would suggest a different outcome. As far as Twyford goes it was more or less neck and neck - less than 100 votes in it |
|
|
|
Post by Pete Whitehead on Jan 12, 2020 21:16:45 GMT
What is your method for calculating? It's not easy to describe without demonstrating the process but it is essentially similar I expect to other systems used by EC etc, based on local election results, since that is the only available data (census data can have it's uses in informing results, especially in the absence of good local election data, but mostly how areas vote locally is going to be the best guide). In London I draw heavily on GLA election data as well. I don't use a single set of local elections as EC tends to do - especially in areas which elect by thirds I'll use an average of results, in order to iron out any anomalies caused by personal votes and so on. So I'd use a basket of recent local election results (but with the most recent carrying more weight) to bring together a dataset showing percentage support for each party in each ward (in most cases just the three main parties, in some the Greens and/or UKIP/Brexit party as well) I will then determine the number of votes cast in each ward based on the electorate and the turnout on a general election day (so the 2015 local elections are useful for this as most areas had elections on the same day as the general and those turnout figures are going to be a pretty good guide. I will then adjust these figures so the total number of votes matches the total number of votes cast in the constituency in the general election (column Q) I'm then able to create a base vote for each party in each ward where the total number of votes matches the total number of votes case in the actual election. It is then just a matter of distributing those votes. If a party gains fewer votes in the election than in the model then I just multiply there vote by the appropriate number (eg if Labour had 10,000 votes on the model and won 8000 in the election, I multiply their vote in each ward by 0.8). Where parties are gaining votes against the model I distribute the extra votes from the other parties as appropriate (and this will vary from case to case) In theexample below, the Brexit party vote is determined as being a fraction (c 0.34) of the UKIP base vote. The Lib Dem shortfall is made up from redistributed Green votes and the Conservative shortfall from the UKIP base vote. In this case the Labour vote is the total vote less the number of votes already assigned to other parties. I always do it this way so as not to alter the total number of votes in each ward from that already determined (though of course it is not always the Labour vote which is determined in this way - often it is the Conservatives)  |
|
They're very interesting!
