clyde1998
SNP
Green (E&W) member; SNP supporter
Posts: 1,765
|
Post by clyde1998 on Oct 17, 2022 13:49:50 GMT
Implied regional vote share, based on 2019 electorates. | Con | Lab | LDm | SNP | PC | Grn | Oth | North East | 25.0% (-13.2) | 55.4% (+12.8) | 7.5% (+0.6) | | | 5.1% (+2.8) | 7.0% (-3.0) | North West | 23.8% (-13.7)
| 54.0% (+7.5) | 10.3% (+2.4) | | | 6.7% (+4.2) | 5.2% (-0.4) | Yorkshire | 27.0% (-16.1) | 51.2% (+12.3) | 8.0% (-0.1) | | | 6.5% (+4.3)
| 7.3% (-0.5) | East Midlands | 33.7% (-21.1) | 46.5% (+14.8) | 8.0% (+0.2) | | | 6.1% (+3.5) | 5.8% (+2.6) | West Midlands | 31.6% (-21.8) | 47.0% (+13.2) | 7.6% (-0.3) | | | 7.9% (+5.0) | 5.8% (+3.9)
| East of England | 33.9% (-23.3) | 41.2% (+16.8) | 13.8% (+0.4) | | | 5.7% (+2.8) | 5.4% (+3.4) | South East | 33.5% (-20.5) | 33.2% (+11.0) | 19.5% (+1.2) | | | 8.7% (+4.7) | 5.2% (+3.5) | South West | 33.5% (-19.3) | 34.3% (+10.9) | 20.2% (+2.0) | | | 7.5% (+3.7) | 4.5% (+2.6) | London | 20.0% (-12.0) | 52.0% (+3.9) | 13.8% (-1.1) | | | 8.4% (+5.4) | 5.7% (+3.8) | NET: England | 29.1% (-18.1) | 45.1% (+11.2) | 12.9% (+0.5) | | | 7.2% (+4.2) | 5.6% (+2.2) | Scotland | 15.0% (-10.1) | 29.1% (+10.5) | 10.7% (+1.2) | 39.9% (-5.2) | | 2.8% (+1.8) | 2.5% (+1.8) | Wales | 23.5% (-12.6) | 42.7% (+1.8) | 7.2% (+1.3) | | 19.4% (+9.4) | 3.4% (+2.4) | 3.8% (-2.3) | NET: Britain | 27.6% (-17.1) | 43.6% (+10.7) | 12.4% (+0.6) | 3.5% (-0.5) | 1.0% (+0.5) | 6.7% (+3.9) | 5.3% (+2.0) |
|
|
|
Post by gwynthegriff on Oct 17, 2022 13:51:29 GMT
Weird map... Much as I would like to see them returning to the fold I doubt either Dunbartonshire East nor North Norfolk will be Lib Dem gains with "new" candidates. And Rhondda is just weird!
|
|
|
Post by bigfatron on Oct 17, 2022 14:08:27 GMT
On the other hand I would expect us to hang on to all three by-election gains...
|
|
|
Post by bjornhattan on Oct 17, 2022 14:43:31 GMT
I actually wouldn't be surprised that much if we held Banff. Scotland isn't that bad from a Unionist point of view. The Labour Party finally getting its act together is a necessary prerequisite for saving the unity of the country. Us maintaining second places across the south, Perthshire and north east is also good. It wouldn't surprise me either, but it would still be an extraordinary result. Twenty years ago, Banff and Buchan was by far the SNP's strongest seat in Scotland; if this MRP is right it is now one of the weakest.
|
|
jamie
Top Poster
Posts: 6,842
|
Post by jamie on Oct 17, 2022 15:15:43 GMT
Implied regional vote share, based on 2019 electorates. | Con | Lab | LDm | SNP | PC | Grn | Oth | North East | 25.0% (-13.2) | 55.4% (+12.8) | 7.5% (+0.6) | | | 5.1% (+2.8) | 7.0% (-3.0) | North West | 23.8% (-13.7)
| 54.0% (+7.5) | 10.3% (+2.4) | | | 6.7% (+4.2) | 5.2% (-0.4) | Yorkshire | 27.0% (-16.1) | 51.2% (+12.3) | 8.0% (-0.1) | | | 6.5% (+4.3)
| 7.3% (-0.5) | East Midlands | 33.7% (-21.1) | 46.5% (+14.8) | 8.0% (+0.2) | | | 6.1% (+3.5) | 5.8% (+2.6) | West Midlands | 31.6% (-21.8) | 47.0% (+13.2) | 7.6% (-0.3) | | | 7.9% (+5.0) | 5.8% (+3.9)
| East of England | 33.9% (-23.3) | 41.2% (+16.8) | 13.8% (+0.4) | | | 5.7% (+2.8) | 5.4% (+3.4) | South East | 33.5% (-20.5) | 33.2% (+11.0) | 19.5% (+1.2) | | | 8.7% (+4.7) | 5.2% (+3.5) | South West | 33.5% (-19.3) | 34.3% (+10.9) | 20.2% (+2.0) | | | 7.5% (+3.7) | 4.5% (+2.6) | London | 20.0% (-12.0) | 52.0% (+3.9) | 13.8% (-1.1) | | | 8.4% (+5.4) | 5.7% (+3.8) | NET: England | 29.1% (-18.1) | 45.1% (+11.2) | 12.9% (+0.5) | | | 7.2% (+4.2) | 5.6% (+2.2) | Scotland | 15.0% (-10.1) | 29.1% (+10.5) | 10.7% (+1.2) | 39.9% (-5.2) | | 2.8% (+1.8) | 2.5% (+1.8) | Wales | 23.5% (-12.6) | 42.7% (+1.8) | 7.2% (+1.3) | | 19.4% (+9.4) | 3.4% (+2.4) | 3.8% (-2.3) | NET: Britain | 27.6% (-17.1) | 43.6% (+10.7) | 12.4% (+0.6) | 3.5% (-0.5) | 1.0% (+0.5) | 6.7% (+3.9) | 5.3% (+2.0) |
The two regional Con-Lab swings that stick out are London and Wales. London makes some sense as that’s where the Conservatives have the least to lose in England, there’s fewer pensioners/Leave voters (who are in most polls swinging the most), and the Labour vote is most vulnerable to the Greens. Wales is a bit more peculiar. PC tend to poll best outside elections (I suspect a decent chunk of their natural support thinks they’re a bit irrelevant come GEs), but even then it’s higher than their support in Welsh only polling (YouGov find around 15%, and Opinium’s sole post-2019 poll was lower than nearby YouGov ones). The Conservative vote perhaps looks a little too high, as does the Lib Dem. I strongly suspect Labour would do quite a bit better than that should the top line figure be borne out in a general election.
|
|
Sibboleth
Labour
'Sit on my finger, sing in my ear, O littleblood.'
Posts: 15,282
|
Post by Sibboleth on Oct 17, 2022 15:20:18 GMT
Very poor Welsh samples are pretty common in national polling o/c and the record of proper polling in Wales isn't great.
|
|
|
Opinium
Oct 17, 2022 15:30:31 GMT
via mobile
Post by mattbewilson on Oct 17, 2022 15:30:31 GMT
Can I just clarify. When we talk about poor subsampling, in the traditional polling sense this is just the raw data from a small group of respondents who are part of a wider data set. But MRP isn't a sample of people polled with a headline opinion plastered on the front page. Using key demographic indicators they predict behaviour based on historic and contemporary trends. Projecting that onto a wider audience. So what we talking about here when we say subsamples?
|
|
clyde1998
SNP
Green (E&W) member; SNP supporter
Posts: 1,765
|
Post by clyde1998 on Oct 17, 2022 15:52:15 GMT
A lot more detail here: www.opinium.com/wp-content/uploads/2022/10/MRP_Tables_2022.pdfThe factors they've got VIs for are: 2019 vote (split by nation), 2016 referendum vote, region, age and gender, marital status, ethnicity, tenure and political attention. Current vote by 2019 vote
| Con | Lab | LDm | SNP | PC | Grn | Oth | Brx | DNV | Ineligible | Con | 63% | 2% | 4% | 0% | 5% | 3% | 11% | 14% | 17% | 12% | Lab | 19% | 84% | 27% | 10% | 18% | 31% | 40% | 42% | 47% | 68% | LDm | 7% | 5% | 62% | 2% | 0% | 8% | 14% | 10% | 14% | 7% | SNP | 0% | 1% | 1% | 85% | 0% | 0% | 0% | 0% | 2% | 1% | PC | 0% | 1% | 0% | 0% | 76% | 0% | 0% | 1% | 1% | 1%
| Grn | 4% | 6% | 6% | 2% | 1% | 56% | 9% | 2% | 10% | 10% | Oth | 7% | 1% | 1% | 1% | 0% | 3% | 25% | 32% | 9% | 1% | Turnout | 92% | 89% | 91% | 88% | 89% | 86% | 86% | 90% | 25% | 48% | Fraction | 39% | 27% | 10% | 3% | 0% | 2% | 1% | 2% | 13% | 3% |
Scotland
| Con | Lab | LDm | SNP | Grn | Oth | Brx | DNV | Ineligible | Con | 59% | 1% | 3% | 0% | 3% | 8% | 9% | 10% | 6% | Lab | 20% | 74% | 22% | 10% | 40% | 37% | 32% | 37% | 73% | LDm | 11% | 5% | 61% | 2% | 11% | 20% | 10% | 14% | 8% | SNP | 3% | 16% | 11% | 85% | 4% | 7% | 16% | 30% | 9% | Grn | 2% | 3% | 3% | 2% | 38% | 4% | 2% | 4% | 4% | Oth | 5% | 1% | 0% | 1% | 3% | 24% | 31% | 5% | 1% | Turnout | 93% | 89% | 92% | 88% | 87% | 88% | 93% | 21% | 50% | Fraction | 23% | 16% | 8% | 38% | 1% | 0% | 0% | 11% | 3% |
Wales
| Con | Lab | LDm | PC | Grn | Oth | Brx | DNV | Ineligible | Con | 63% | 2% | 6% | 5% | 4% | 11% | 18% | 14% | 11% | Lab | 16% | 75% | 31% | 18% | 38% | 38% | 42% | 36% | 61% | LDm | 5% | 4% | 54% | 0% | 8% | 10% | 10% | 9% | 5% | PC | 10% | 15% | 4% | 76% | 7% | 17% | 4% | 31% | 16% | Grn | 2% | 4% | 4% | 1% | 38% | 5% | 2% | 5% | 6% | Oth | 5% | 1% | 1% | 0% | 4% | 19% | 24% | 5% | 1% | Turnout | 90% | 90% | 90% | 89% | 77% | 95% | 91% | 22% | 45% | Fraction | 31% | 35% | 5% | 8% | 1% | 1% | 5% | 12% | 3% |
|
|
|
Post by Peter Wilkinson on Oct 17, 2022 17:50:43 GMT
This MRP actually doesn't underestimate Lib Dem chances, it seems to have worked around the limitations quite well (which has led to a slightly odd-looking Labour underperformance in Finchley and Two Cities). It does seem to have given us some slightly odd seats while missing some more obvious ones, but that is the same for all parties in this poll. Overall, I think that you are generally correct here, but the figures both for Finchley and for Two Cities certainly look as if they have overweighted the 2019 GE results (though possibly because of some less obvious underweightings on other factors). In terms of Lib Dem support, the 2019 GE results (and the 2019 Euro ones) were way out of line with any other elections in recent decades, whether earlier or later ones - and the only way I can see the Lib Dems getting near the figures in this poll would be to put in the same level of targetting as in 2019, for what would still only be clear third places. Not sensible, when there will be plenty of better opportunities elsewhere. The Finchley result in particular looks distinctly odd, particularly when compared with neighbouring Hendon - before 2019, the two constituencies had moved, at least in general elections, pretty much in lockstep, at least once one allowed for Finchley being a few per cent better for the Tories and worse for Labour than Hendon. And while there are certainly differences between the two constituencies, they certainly don't feel large enough for the Tories to win Finchley, even if only by 0.5%, but to lose Hendon by 40%. In particular, if (as might still be the case, despite removal of the main cause) the Labour vote in Finchley is still being somewhat suppressed by it having the largest Jewish vote in the country, then there should be a similar (if slightly smaller) effect in Hendon - with the second largest Jewish vote in the country. Though part of the oddity may be with Hendon - it has a projected Labour-Tory two-party vote of 86.6%, the highest in London and the eleventh-highest in Great Britain. While, from memory, it has always had a fairly high two-party vote, I didn't think it was that high. The only other constituency in London with a two-party vote projected higher than 85% is Romford - which at least is one of five in historic southern Essex (the others are Castle Point, Harlow, South Basildon and East Thurrock, and Thurrock). Elsewhere in England, there are 21 such constituencies (Luton, 15 in northern England and 5 in the Midlands) - scarcely surprisingly, considering the existence of the SNP and Plaid Cymru, there are none in Scotland or Wales. (By the way, this does relate just to Labour-Tory two-party votes - however, there is only one two-party vote in the figures which is not Labour-Tory, a Labour-"Other" two-party vote of over 94% in Chorley. But the calculation of figures for Chorley looks very specific to the job of the current incumbent.)
|
|
iain
Lib Dem
Posts: 10,707
|
Post by iain on Oct 17, 2022 17:58:01 GMT
Overall, I think that you are generally correct here, but the figures both for Finchley and for Two Cities certainly look as if they have overweighted the 2019 GE results (though possibly because of some less obvious underweightings on other factors). In terms of Lib Dem support, the 2019 GE results (and the 2019 Euro ones) were way out of line with any other elections in recent decades, whether earlier or later ones - and the only way I can see the Lib Dems getting near the figures in this poll would be to put in the same level of targetting as in 2019, for what would still only be clear third places. Not sensible, when there will be plenty of better opportunities elsewhere. To be clear, that's also what I was saying. I think the poll must have had some sort of weighting based on the second placed party in 2019, which has led to unrealistic results in these two constituencies.
|
|
|
Post by Peter Wilkinson on Oct 17, 2022 18:22:54 GMT
Overall, I think that you are generally correct here, but the figures both for Finchley and for Two Cities certainly look as if they have overweighted the 2019 GE results (though possibly because of some less obvious underweightings on other factors). In terms of Lib Dem support, the 2019 GE results (and the 2019 Euro ones) were way out of line with any other elections in recent decades, whether earlier or later ones - and the only way I can see the Lib Dems getting near the figures in this poll would be to put in the same level of targetting as in 2019, for what would still only be clear third places. Not sensible, when there will be plenty of better opportunities elsewhere. To be clear, that's also what I was saying. I think the poll must have had some sort of weighting based on the second placed party in 2019, which has led to unrealistic results in these two constituencies. I thought that that probably was what you were saying, though I'm pleased to have it clarified. It did, however, make a useful intro for my remarks on Finchley (and Hendon). By the way, the Chorley figures which I referred to at the end of my last post certainly suggest some kind of second placed party weighting, though they also suggest some further assumptions about which parties will (and will not) be standing there at the next election which they are applying nowhere else.
|
|
|
Post by froome on Oct 17, 2022 20:04:53 GMT
Clearly some very strong regional (or even county) subsamples, e.g. Labour winning every seat in Suffolk and 5 in Wiltshire Lib Dems winning 6 seats in Surrey Conservatives doing decently in the NW, including holding Pendle The constituency results in Wiltshire are certainly odd. Here are the swings (from Tory to Labour) in the five Labour gains: South Swindon - 16% North Swindon - 17% South West Wilts - 20% Salisbury - 22% Devizes - 28% (!) Devizes ends up with a larger Labour majority than North Swindon - which doesn't seem remotely realistic. A Labour win there full stop would be a stretch, even in these extreme conditions. Having said that, the other two Labour gains are more plausible. Over a third of South West Wiltshire's electorate live in the town of Trowbridge, and you'd expect Labour to win a big majority there which could offset any rural Conservative strength. Similarly, some of the Salisbury city wards such as Bemerton would be virtually monolithic for Labour in these conditions which might be enough to make the seat as a whole Labour. The whole SW list seems odd to me. Devizes going Labour seems completely unthinkable, and what seems oddest is that Chippenham remains Conservative! Just to the west, it shows North Somerset going to Labour while Weston-super-Mare stays Conservative, which is again unbelievable. I assume this is mainly due to the way they each voted in the referendum, as I can't see any other reasoning for it.
|
|
iain
Lib Dem
Posts: 10,707
|
Post by iain on Oct 17, 2022 20:10:58 GMT
North Somerset is more marginal than Weston tbf
|
|
|
Post by batman on Oct 17, 2022 20:23:53 GMT
yes several recent polls on UNS have shown Labour gaining North Somerset but the Tories holding Weston-super-Mare. However, in recent days most have shown Labour gaining both of them. It is true however that most observers would tend to think that Labour has more potential in the latter than the former & in 2017 Labour came surprisingly close in relative terms there. AFAIK in North Somerset Labour's best area tends to be Clevedon
|
|
|
Post by Adam in Stroud on Oct 17, 2022 20:26:00 GMT
Much more of this, and come the next GE we're going to be walking round previously safe Tory seats telling Labour voters we're in poll position to get the Tories out and Tory voters that we're the only hope of stopping the place going Labour.
The bar charts are going to be tricky.
|
|
|
Post by John Chanin on Oct 18, 2022 6:45:06 GMT
Much more of this, and come the next GE we're going to be walking round previously safe Tory seats telling Labour voters we're in poll position to get the Tories out and Tory voters that we're the only hope of stopping the place going Labour. The bar charts are going to be trickyYou’ll find a way…..
|
|
|
Opinium
Oct 18, 2022 7:50:23 GMT
via mobile
Post by aargauer on Oct 18, 2022 7:50:23 GMT
Much more of this, and come the next GE we're going to be walking round previously safe Tory seats telling Labour voters we're in poll position to get the Tories out and Tory voters that we're the only hope of stopping the place going Labour. The bar charts are going to be trickyYou’ll find a way….. My LD friend (now on Durham CC) managed to create a bar chart showing him in second in a ward the Lib Dems had never run in before! Incidentally labour were in 1st on said chart despite not being the incumbent and ended up coming 3rd.
|
|
|
Post by Peter Wilkinson on Oct 18, 2022 10:29:06 GMT
Much more of this, and come the next GE we're going to be walking round previously safe Tory seats telling Labour voters we're in poll position to get the Tories out and Tory voters that we're the only hope of stopping the place going Labour. The bar charts are going to be trickyYou’ll find a way….. "Only the Lib Dems can beat ANYBODY here"
|
|
|
Post by batman on Oct 18, 2022 11:08:17 GMT
Maybe one day I'll wake up & hear of a leaflet where the Liberal Democrats say "vote for us because our policies on such-and-such are better"
|
|
The Bishop
Labour
Down With Factionalism!
Posts: 36,540
Member is Online
|
Post by The Bishop on Oct 18, 2022 11:41:42 GMT
And they will produce another leaflet for the next street promising totally different policies
|
|