YL
Non-Aligned
Either Labour leaning or Lib Dem leaning but not sure which
Posts: 4,917
|
Post by YL on Sept 29, 2022 7:09:31 GMT
As ever with MRPs, staggeringly high Labour shares in LD/Con marginals including: 23.6% in Winchester (2019 share 4.6%) 24.1% in Cheltenham (4.9%) 24.4% in Richmond Park (5.2%) And these are the *lowest* Labour shares in LD/Con marginals I’ve found. Why do pollsters look at this, think it’s credible and publish the results? Well it probably is credible, there will be a significant number of people in LD/Con marginals who would instinctively say they’re going to vote Labour (especially when Labour are far ahead nationally), but come a general election and being bombarded with literature telling them the Local Lib Dem candidate is great and Labour Can’t Win Here, will end up voting LD. I think it may be more that the model is inferring the voting intentions of people in Winchester (say) from those of similar demographics around the country, as it does not have enough voters who are actually in Winchester to pick up any sort of local effect there. It's quite likely that many of those who are intending to vote Lib Dem in Winchester would in fact vote Labour were they not in a Con/LD marginal, and so the model thinks they will even there. There was a YouGov MRP a few months ago which looked specifically at Con/LD marginals, and so would have avoided that problem, and it showed a lot more Lib Dem gains than either this ComRes one or the FindOutNowUK/Electoral Calculus one released yesterday. |
|
|
|
Post by batman on Sept 29, 2022 7:37:13 GMT
What does this MRP poll suggest re- Wimbledon? Ah I see Wimbledon is a Labour gain.
Not going to happen dogs and bones come to mind. I think the boundary changes kill Labour's chances in Wimbledon altogether but that on existing boundaries they aren't quite dead yet. However, no current opinion polls suggest anything other than a Lib Dem gain on UNS, and on the basis of recent local elections it's hard to see far beyond that, for most people at least. And I don't think you'll find that Wimbledon CLP is a very effective fighting force at the moment either whereas the Lib Dems certainly are. |
|
|
|
Post by bigfatron on Sept 29, 2022 8:03:43 GMT
As ever with MRPs, staggeringly high Labour shares in LD/Con marginals including: 23.6% in Winchester (2019 share 4.6%) 24.1% in Cheltenham (4.9%) 24.4% in Richmond Park (5.2%) And these are the *lowest* Labour shares in LD/Con marginals I’ve found. Why do pollsters look at this, think it’s credible and publish the results? we would (expletive deleted) kill for 24.4% of the vote in my constituency. This does show the glaring weaknesses of an MRP poll, as do some of the other silly results shown in the thread. At teh risk of stating the obvious, I think it's more about the way MRP information is mis-applied; my understanding is that n MRP poll says 'based on a demographic model of each seat and the general sentiment of the voters within each demographic category, to each party, this is how we would expect each seat to vote'. The MRP does this job well, but definitionally ignores local factors like popularity of the sitting MP, tactical voting, etc. Then the MRP output gets misused because it's output is quoted as a prediction for seats where that is inappropriate, precisely because of those local factors, leading to a view that MRP polls are pointless. |
|
clyde1998
SNP
Green (E&W) member; SNP supporter
Posts: 1,765 
|
Post by clyde1998 on Sept 29, 2022 8:11:34 GMT
Well it probably is credible, there will be a significant number of people in LD/Con marginals who would instinctively say they’re going to vote Labour (especially when Labour are far ahead nationally), but come a general election and being bombarded with literature telling them the Local Lib Dem candidate is great and Labour Can’t Win Here, will end up voting LD. I think it may be more that the model is inferring the voting intentions of people in Winchester (say) from those of similar demographics around the country, as it does not have enough voters who are actually in Winchester to pick up any sort of local effect there. It's quite likely that many of those who are intending to vote Lib Dem in Winchester would in fact vote Labour were they not in a Con/LD marginal, and so the model thinks they will even there. There was a YouGov MRP a few months ago which looked specifically at Con/LD marginals, and so would have avoided that problem, and it showed a lot more Lib Dem gains than either this ComRes one or the FindOutNowUK/Electoral Calculus one released yesterday. Honestly, I believe looking at specific groups of constituencies for MRP modelling is probably going to be its most effective use. At the very least, different types of constituency need to be separated as people will vote differently based on who the relevant parties are. |
|
graham
Non-Aligned
Posts: 1,350
|
Post by graham on Sept 29, 2022 8:11:45 GMT
dogs and bones come to mind. I think the boundary changes kill Labour's chances in Wimbledon altogether but that on existing boundaries they aren't quite dead yet. However, no current opinion polls suggest anything other than a Lib Dem gain on UNS, and on the basis of recent local elections it's hard to see far beyond that, for most people at least. And I don't think you'll find that Wimbledon CLP is a very effective fighting force at the moment either whereas the Lib Dems certainly are. Re-UNS I am not sure that is correct. A 45% vote share across GB would imply 36%/37% in Wimbledon. That might well be enough in a 3-way contest. 'A rising tide lifts all boats.'
LD support is little changed nationally - perhaps marginally lower than 2019. Certainly no reason for Labour to see the seat as a 'write off'.
|
|
|
|
Post by matureleft on Sept 29, 2022 8:22:49 GMT
1997 demonstrated that a rising tide did indeed lift almost all boats, and that the absence of a very effective campaigning party didn’t necessarily prevent Labour victories. There were a few results where effort must have been pretty modest and narrow.
But in the real world this isn’t (yet) anything like 1997 in feel and I’d suspect there’ll be a rational and focused Lib Dem strategy (almost certainly assisted by Labour connivance at national level).
The bit of 1997 I do recognise is the sharper Labour focus on removing negatives and risk, and, largely, the acceptance by the members of the limits to their goals. 12 years is a long time away from government, and that wears on all but the most ideological of members.
|
|
The Bishop
Labour
Down With Factionalism!
Posts: 39,075
|
Post by The Bishop on Sept 29, 2022 10:18:34 GMT
Penrith and the Border??? Well yes, roughly my reaction. Though it is worth pointing out to the uninitiated that it does have a decent Labour vote in some places. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 29, 2022 11:01:36 GMT
MRP looks at "the kind of people who vote for party X", not necessarily voting intention for party X. In the tables of the poll it’s literally described as Voting Intention. Ok, but imagine: you're a low-interest voter whose engagement in politics is voting in a GE once every five years. You're mostly aware of the Tories and Labour because they're constantly in the news, and although you know of other parties they kind of slip your mind right now. Two years at least from a GE you're not thinking about politics at all, or about how you'd vote or what the issues are when someone comes alonmg and asks your voting intention so you go for whoever's not in government right now. It's quite possible that that does add up to Labour shares in excess of 20% in some Con-LD marginals even though we all know Labour won't get anywhere near that in an actual election |
|
|
|
Post by liverpoolliberal on Sept 29, 2022 15:52:58 GMT
In the tables of the poll it’s literally described as Voting Intention. Ok, but imagine: you're a low-interest voter whose engagement in politics is voting in a GE once every five years. You're mostly aware of the Tories and Labour because they're constantly in the news, and although you know of other parties they kind of slip your mind right now. Two years at least from a GE you're not thinking about politics at all, or about how you'd vote or what the issues are when someone comes alonmg and asks your voting intention so you go for whoever's not in government right now. It's quite possible that that does add up to Labour shares in excess of 20% in some Con-LD marginals even though we all know Labour won't get anywhere near that in an actual election I am perfectly aware that low interest voters will not think tactically when saying how they will vote in an opinion poll, years out from an election. I also agree that there are very likely high numbers of Labour supporters in these Con/LD marginals, making it credible that at the moment 20-30% of voters in those seat would say they will vote Labour. I'm suggesting that, as we know this is not what they will actually do come election time, these MRPs that give seat by seat predictions are somewhat flawed and perhaps not a good use of polling outside of election time. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Sept 29, 2022 16:18:57 GMT
Ok, but imagine: you're a low-interest voter whose engagement in politics is voting in a GE once every five years. You're mostly aware of the Tories and Labour because they're constantly in the news, and although you know of other parties they kind of slip your mind right now. Two years at least from a GE you're not thinking about politics at all, or about how you'd vote or what the issues are when someone comes alonmg and asks your voting intention so you go for whoever's not in government right now. It's quite possible that that does add up to Labour shares in excess of 20% in some Con-LD marginals even though we all know Labour won't get anywhere near that in an actual election I am perfectly aware that low interest voters will not think tactically when saying how they will vote in an opinion poll, years out from an election. I also agree that there are very likely high numbers of Labour supporters in these Con/LD marginals, making it credible that at the moment 20-30% of voters in those seat would say they will vote Labour. I'm suggesting that, as we know this is not what they will actually do come election time, these MRPs that give seat by seat predictions are somewhat flawed and perhaps not a good use of polling outside of election time. But they aren't predictions, any more than an opinion poll is a prediction of how people will vote at the next election. It is a poll of current voting intention mapped onto the current constituency map to give an idea of what would (or might) happen if an election were held tomorrow. The problem is that people treat and paint them as predictions even though they aren't designed to be |
|
|
|
Post by michaelarden on Sept 29, 2022 18:53:10 GMT
Ok, but imagine: you're a low-interest voter whose engagement in politics is voting in a GE once every five years. You're mostly aware of the Tories and Labour because they're constantly in the news, and although you know of other parties they kind of slip your mind right now. Two years at least from a GE you're not thinking about politics at all, or about how you'd vote or what the issues are when someone comes alonmg and asks your voting intention so you go for whoever's not in government right now. It's quite possible that that does add up to Labour shares in excess of 20% in some Con-LD marginals even though we all know Labour won't get anywhere near that in an actual election I am perfectly aware that low interest voters will not think tactically when saying how they will vote in an opinion poll, years out from an election. I also agree that there are very likely high numbers of Labour supporters in these Con/LD marginals, making it credible that at the moment 20-30% of voters in those seat would say they will vote Labour. I'm suggesting that, as we know this is not what they will actually do come election time, these MRPs that give seat by seat predictions are somewhat flawed and perhaps not a good use of polling outside of election time. Labour polled over 23% in Kingston and Surbiton in 1997 for example. |
|
iain
Lib Dem
Posts: 11,455
|
Post by iain on Sept 29, 2022 19:23:37 GMT
I am perfectly aware that low interest voters will not think tactically when saying how they will vote in an opinion poll, years out from an election. I also agree that there are very likely high numbers of Labour supporters in these Con/LD marginals, making it credible that at the moment 20-30% of voters in those seat would say they will vote Labour. I'm suggesting that, as we know this is not what they will actually do come election time, these MRPs that give seat by seat predictions are somewhat flawed and perhaps not a good use of polling outside of election time. Labour polled over 23% in Kingston and Surbiton in 1997 for example. Despite a local campaign, Kingston & Surbiton was not a Lib Dem target seat in 1997 (I believe it is one of two seats we have won in GEs which weren't targeted, but I've forgotten what the other was). When it was a target seat, in 2001, the Labour vote fell to 9%. |
|
|
|
Post by michaelarden on Sept 29, 2022 21:13:47 GMT
Labour polled over 23% in Kingston and Surbiton in 1997 for example. Despite a local campaign, Kingston & Surbiton was not a Lib Dem target seat in 1997 (I believe it is one of two seats we have won in GEs which weren't targeted, but I've forgotten what the other was). When it was a target seat, in 2001, the Labour vote fell to 9%. I think you're making my point - in 1997 when Labour were polling well into the 40s (and even 50s) consistently for years their vote was high pretty much everywhere - even winning a number of Lib Dem target seats from third. The circumstances in 2001 were different with Labour not under threat of losing nationally and a rather controversial Tory candidate in a hyper marginal seat with active and high profile Lib Dem MP. |
|
timmullen1
Labour
Closing account as BossMan declines to respond to messages seeking support.
Posts: 11,823
|
Post by timmullen1 on Oct 3, 2022 15:44:36 GMT
Stand by, poll incoming!
And here it is:
|
|
timmullen1
Labour
Closing account as BossMan declines to respond to messages seeking support.
Posts: 11,823
|
Post by timmullen1 on Oct 14, 2022 13:44:41 GMT
|
|
Chris from Brum
Lib Dem
What I need is a strong drink and a peer group.
Posts: 9,771
|
Post by Chris from Brum on Oct 14, 2022 14:49:47 GMT
LD down from 10.5 to 10.4, Green up from 3.4 to 3.5 ...
|
|
|
|
Post by michaelarden on Oct 14, 2022 16:11:48 GMT
LD down from 10.5 to 10.4, Green up from 3.4 to 3.5 ... Or LD down from 11.4 to 9.5 and Green up from 2.5 to 4.4? |
|
Chris from Brum
Lib Dem
What I need is a strong drink and a peer group.
Posts: 9,771
|
Post by Chris from Brum on Oct 14, 2022 16:31:55 GMT
LD down from 10.5 to 10.4, Green up from 3.4 to 3.5 ... Or LD down from 11.4 to 9.5 and Green up from 2.5 to 4.4? I think mine is way more likely. |
|
maxque
Non-Aligned
Posts: 9,318
|
Post by maxque on Oct 15, 2022 2:11:11 GMT
Or LD down from 11.4 to 9.5 and Green up from 2.5 to 4.4? I think mine is way more likely. Crosstabs of the last polls are not out yet, but LD were on 10.9 last time and Greens on 3.3. |
|
timmullen1
Labour
Closing account as BossMan declines to respond to messages seeking support.
Posts: 11,823
|
Post by timmullen1 on Oct 19, 2022 14:59:55 GMT
|
|
