|
|
Post by aargauer on Jul 30, 2024 8:01:48 GMT
JK turns out some excellent work. It is a surprise to see Reform underperformed UKIP15 in some Leave type seats but concentrate their vote in select areas. Potential to embed or flash-in-a-pan? The beginning of lib demification |
|
|
|
Post by hullenedge on Jul 30, 2024 15:45:09 GMT
|
|
|
|
Post by Pete Whitehead on Jul 30, 2024 16:14:57 GMT
Looks good.. mostly. I see a lot of similarity with my results in Hertfordshire. A bit troubling that he has 'others' carrying Werrington In Peterborough and the rural South Bedfordshire wards, these being the Gaza Independents and Workers Party. Similar issues in Birmingham. I don;t know why so many of these models have so much trouble discerning where these Independents get their votes from. I didn;t think the demographic basis of their support was that much of a mystery..
|
|
jamie
Top Poster
Posts: 7,053 
Member is Online
|
Post by jamie on Jul 30, 2024 16:24:05 GMT
Looks good.. mostly. I see a lot of similarity with my results in Hertfordshire. A bit troubling that he has 'others' carrying Werrington In Peterborough and the rural South Bedfordshire wards, these being the Gaza Independents and Workers Party. Similar issues in Birmingham. I don;t know why so many of these models have so much trouble discerning where these Independents get their votes from. I didn;t think the demographic basis of their support was that much of a mystery.. I presume there is modelling based on local elections? That looks to be the case in Newcastle where the Gaza Independents apparently did best in the very strongly independent ward (very white, older, owner-occupier). |
|
|
|
Post by Pete Whitehead on Jul 30, 2024 16:31:40 GMT
I presume so. Werrington for example has long been an Indepenednt stronghold in local elections and I think Caddington has been recently. I base my models on local elections* but there's no reason to be doing this when its obvious people who live in White suburbs and rural areas who vote for Independent councillors are not the people voting for quasi-Islamists in a general election
*obviously in these cases one uses the census
|
|
|
|
Post by John Chanin on Jul 30, 2024 16:51:27 GMT
Total crap for my constituency. It has the Muslim independents carrying Hall Green South and Moseley, and Labour carrying Sparkhill and Sparkbrook, when it is self evident to anyone with a brain that the opposite is true. The same idiocy applies to Yardley, Hodge Hill, Ladywood, and less obviously Parry Barr. |
|
john07
Labour & Co-operative
Posts: 15,774
|
Post by john07 on Jul 30, 2024 16:51:55 GMT
One thing that puzzles me is that the map is for the General Election results. If you click on the tab to see the spreadsheet, it gives the 2024 Local Election results? |
|
|
|
Post by edgbaston on Jul 30, 2024 17:05:27 GMT
He’s got the strongest Labour ward in Ladywood, Hall Green and Yardly all down as independent. Actually impressively ignorant of the local situation |
|
|
|
Post by swanarcadian on Jul 30, 2024 17:06:44 GMT
Ben, if you’re reading this, thank you for your fine efforts. |
|
Sibboleth
Labour
'Sit on my finger, sing in my ear, O littleblood.'
Posts: 16,025
Member is Online
|
Post by Sibboleth on Jul 30, 2024 17:46:36 GMT
He’s got the strongest Labour ward in Ladywood, Hall Green and Yardly all down as independent. Actually impressively ignorant of the local situation Yeah, the models... um... well there's an obvious problem in certain constituencies isn't there. |
|
|
|
Post by Andrew_S on Jul 30, 2024 17:51:32 GMT
The first thing I look at is Tamworth and he has Bourne Vale being won by Labour, probably the safest Tory ward in the constituency. |
|
YL
Non-Aligned
Either Labour leaning or Lib Dem leaning but not sure which
Posts: 4,905
|
Post by YL on Jul 30, 2024 18:07:59 GMT
The first thing I look at is Tamworth and he has Bourne Vale being won by Labour, probably the safest Tory ward in the constituency. This ward has a nasty habit of having uncontested local elections; maybe this is being treated as essentially no data rather than giving the information that it's such a safe Tory ward that no-one else can be bothered challenging them there? |
|
jamie
Top Poster
Posts: 7,053
Member is Online
|
Post by jamie on Jul 30, 2024 19:01:57 GMT
The first thing I look at is Tamworth and he has Bourne Vale being won by Labour, probably the safest Tory ward in the constituency. This ward has a nasty habit of having uncontested local elections; maybe this is being treated as essentially no data rather than giving the information that it's such a safe Tory ward that no-one else can be bothered challenging them there? That’s a problem in Wales where 2 usually strongly ex mining Labour wards are shown as Conservative won in Brecon etc. |
|
|
|
Post by greenchristian on Jul 30, 2024 19:33:20 GMT
I only have good sampling data for one ward in Coventry, but it's interesting to compare my figures for the on-the-day vote in Holbrook to what's implied by the model Pete is using for the maps. Here's how they compare: Party
| Sample | Model | | Labour | 52% | 50-55% | Reform
| 18%
| 12.5%-15%
| | Conservative | 15% | 10-15% | Green
| 11% | 15-17.5% | Lib Dem
| 3% | 0-5% |
Three parties clearly within Pete's range and two a few points outside isn't bad. It could be the result of different voting patterns between on the day votes and postal votes, or it could be that we're keeping fewer of our local election voters in the GE than Pete's model assumes. |
|
|
|
Post by bjornhattan on Jul 30, 2024 19:48:46 GMT
I only have good sampling data for one ward in Coventry, but it's interesting to compare my figures for the on-the-day vote in Holbrook to what's implied by the model Pete is using for the maps. Here's how they compare: Party
| Sample | Model | | Labour | 52% | 50-55% | Reform
| 18%
| 12.5%-15%
| | Conservative | 15% | 10-15% | Green
| 11% | 15-17.5% | Lib Dem
| 3% | 0-5% |
Three parties clearly within Pete's range and two a few points outside isn't bad. It could be the result of different voting patterns between on the day votes and postal votes, or it could be that we're keeping fewer of our local election voters in the GE than Pete's model assumes. For reference, my model figures were: Lab 51.9% Con 19.2% Ref 19.1% Grn 6.8% LD 1.7% Ind/Oth 1.3% I don't know if Pete's model goes off demographics at all (for parties other than Gaza independents etc) but it is interesting that the sampling results appear to be intermediate between mine (demographics based) and his (local elections based). He has the Green vote high while I have it low; I have the Conservatives and Reform votes high while he has them low. |
|
|
|
Post by Pete Whitehead on Jul 30, 2024 19:53:40 GMT
I only have good sampling data for one ward in Coventry, but it's interesting to compare my figures for the on-the-day vote in Holbrook to what's implied by the model Pete is using for the maps. Here's how they compare: Party
| Sample | Model | | Labour | 52% | 50-55% | Reform
| 18%
| 12.5%-15%
| | Conservative | 15% | 10-15% | Green
| 11% | 15-17.5% | Lib Dem
| 3% | 0-5% |
Three parties clearly within Pete's range and two a few points outside isn't bad. It could be the result of different voting patterns between on the day votes and postal votes, or it could be that we're keeping fewer of our local election voters in the GE than Pete's model assumes. It's always difficult with these kind of wards where a small party which doesn't have much strength in a constituency overall, has concentrated local election strength in a single ward. One tries to scale back the effects of this (which would have been more extreme without adjustmnents). It always used to be a nightmare with the Lib Dems who habitually had a single ward in which they won overwhelmingly while scoring low vote share everywhere else. You have to do something to level things out a bit but in the end its arbitrary - which is why this is always as much art as science. The Greens have annoyingly now adopted similar targeting strategies to the Lib Dems with some success and thus cause even more problems in this respect. |
|
|
|
Post by Pete Whitehead on Jul 30, 2024 19:59:22 GMT
I only have good sampling data for one ward in Coventry, but it's interesting to compare my figures for the on-the-day vote in Holbrook to what's implied by the model Pete is using for the maps. Here's how they compare: Party
| Sample | Model | | Labour | 52% | 50-55% | Reform
| 18%
| 12.5%-15%
| | Conservative | 15% | 10-15% | Green
| 11% | 15-17.5% | Lib Dem
| 3% | 0-5% |
Three parties clearly within Pete's range and two a few points outside isn't bad. It could be the result of different voting patterns between on the day votes and postal votes, or it could be that we're keeping fewer of our local election voters in the GE than Pete's model assumes. For reference, my model figures were: Lab 51.9% Con 19.2% Ref 19.1% Grn 6.8% LD 1.7% Ind/Oth 1.3% I don't know if Pete's model goes off demographics at all (for parties other than Gaza independents etc) but it is interesting that the sampling results appear to be intermediate between mine (demographics based) and his (local elections based). He has the Green vote high while I have it low; I have the Conservatives and Reform votes high while he has them low. The Grens got 8% in Coventry NW so it seems unlikely that they won a lower share than that in the one ward where they have local strength. The relationship can be very tenuous or even contradictory though. I remember the results published by Walsall Council showing no evidence whatsoever of the residual Lib Dem strength in the Willenhall end of the old Walsall North. I have used demographics sometimes to determine the UKIP/Reform vote where there is otherwise no useful local election data and occasionally so for the main parties where there is very little partisan competition in local elections (I've not uyet attempted Powys or Pembrokeshire, but if I did I imagine I would need to do that) |
|
|
|
Post by Pete Whitehead on Jul 30, 2024 20:04:13 GMT
I only have good sampling data for one ward in Coventry, but it's interesting to compare my figures for the on-the-day vote in Holbrook to what's implied by the model Pete is using for the maps. Here's how they compare: Party
| Sample | Model | | Labour | 52% | 50-55% | Reform
| 18%
| 12.5%-15%
| | Conservative | 15% | 10-15% | Green
| 11% | 15-17.5% | Lib Dem
| 3% | 0-5% |
Three parties clearly within Pete's range and two a few points outside isn't bad. It could be the result of different voting patterns between on the day votes and postal votes, or it could be that we're keeping fewer of our local election voters in the GE than Pete's model assumes. My actualy figures for Holbrook were Lab 53.9% Grn 15.1% Ref 14.6% Con 11.7% LD 3.5% So quite out on the Conservative vote which I guess is a consequence of them getting squeezed a but by the Greens locally. I'm not sure where their vote should be lower to compensate as it is hardly astronomical in Bablake and Woodlands. I suspect the Conservative share in Sherbourne has been a bit higher in recent local elections than it 'should' have been. |
|
Clark
Forum Regular
Posts: 744
|
Post by Clark on Jul 30, 2024 20:05:26 GMT
Highest Tory share I found was just under 59% in Chigwell Row - I see he has the Lib Dems just pipping the Conservatives in Gerrards Cross too!
|
|
|
|
Post by Pete Whitehead on Jul 30, 2024 20:08:10 GMT
Highest Tory share I found was just under 59% in Chigwell Row - I see he has the Lib Dems just pipping the Conservatives in Gerrards Cross too! He has 60.5% in Stanmore (which is also my highest share) |
|
