|
|
Post by James E on Jan 19, 2022 17:04:50 GMT
leftieliberal "I thought that it would be a good idea to look at average polling over the whole of 2021 from all the regular pollsters whose polling I have been plotting graphs of in my Polling Archive spreadsheet." You need to get out more mate. That was rather uncalled for, Steve. This is meant to be a polling site, and LL's post provided some useful insight into the 'house effects' of the various regular pollsters. It does rather confirm much that we've observed before (for example that YG have lower Lab figures because of the prompt for Greens) but it's good to have some actual summary figures. And many thanks to Leftieliberal. |
|
|
|
Post by James E on Jan 19, 2022 16:41:34 GMT
More on the savanta poll Among Con 2019 voters Stick with Con 64% etc... It's worth comparing those 'churn' figures with the most recent YouGov, as they are markedly different, even though both companies have similar headline results Of Con 2019 voters YouGov have: Conservative 51% (13 points less) Labour 8% (2 points less) Vote for others 14% Don't Know 21% (6 points more) Would not vote 6% I suspect that this is another case of 'Panel' v 'recalled past vote' difference. And as such, I'd trust the YG figures rather more. |
|
|
|
Post by James E on Jan 19, 2022 12:28:11 GMT
As I read it, that Red Wall poll shows an overall movement from Con 48/Lab39 in GE2019 to Con37/Lab 48 now.
So this is much in line with the current polls giving Labour a 10 point lead, and an 11% Swing from the last election.
To add: the details of the 'Red Wall' poll provided by Hireton above are well worth reading. I think these are are more worrying for the Tories than the headline voting figures, especially how the two main parties are perceived on a range of issues.
|
|
|
|
Post by James E on Jan 18, 2022 21:50:00 GMT
@ TED
Yes, there must be a fair chance that the vote is 'triggered' but then Johnson survives, just as May did in the same situation in Dec 2018.
This would probably be the ideal scanario for Labour.
|
|
|
|
Post by James E on Jan 18, 2022 19:56:36 GMT
@jimjam
The YouGov released today is actually less recent than the R&W and Deltapoll ones we had yesterday. Agree that the central range is a 10-11 point lead.
One trend that I've noted in all the recent polls for which I've seen tables: the gender gap is widening.
At the last GE, YouGov estimated that Lab did a net 6 points better with women than with men (i.e. Cons led by 15 points with men and 9 with women voters).
In the recent polls the differences between female and male VI are:
10 points with Focaldata (44/24 v 39/29)
10 points with YouGov (43/30 v 36/33)
17 points with Deltapoll (44/27 v 38/36)
24 points with R&W (50/24 v 38/36)
That's an average of 15 points; and it's still 12 if you exclude R&W as an outlier.
|
|
|
|
Post by James E on Jan 18, 2022 8:57:40 GMT
Other things I'm not too keen on are big blocks of twitter and pictures I think it's good practice (in replying) to shorten people's responses to the part you want to react to, and to remove long attachments, especially twitter blocks. As I've done here... So when, for example, replying to a post with a tweeted poll, just delete it in the boxed text, and add something like [R&W poll etc ] |
|
|
|
Post by James E on Jan 17, 2022 17:58:05 GMT
Trevor, Not sure if the lead has stretched to 13% ish. The last 7 polls have 1 x 9% lead, 3 x 10% and 1 x 11% which average nicely at 10%. Plus a 14%, which I did think at one end of MOE, and this 13%. So average exactly 11% but kind of meaningless until Johnson goes. Does give a feel of the potential for Labour as even with Tory DKs returning, which James reckons from the YG is about 3%, Labour are hitting 40%. This R&W has the highest lvel of Con>Lab movement we've seen yet, with 12% of those expressing a voting Intention (or 9% including the DKs) going from 2019Con to Lab. That's about 5.2% of direct swing. Con2019 Don't Knows are 20% compared to 6% Lab, which would reduce the Lab lead by just over 3 points using the old ICM 50% reversion. So just about a 10 point lead even with this factored in. A couple of oddities in this poll: the Scottish Cross-break (n=135) is SNP 34%, Lab 31%, Con 19%. But before Graham gets excited, this is most likely a one-off. South of England cross-breaks put the Tories narrowly ahead, but this is still a huge (c14%) swing on GE2019. The gender cross breaks are remarkable. Labour are 2 points ahead with men (38/36) and 26 points ahead with women (50/24). I did say a few days ago that the political gender gap is widening - but surely not that much (yet....) |
|
|
|
Post by James E on Jan 17, 2022 9:57:04 GMT
I hope he doesn't re-invent himself as a national treasure. I might have to emigrate. To Scotland maybe.  Johnson's 15% majority does look vulnerable, especially as polls currently stand, but he would (on UNS) just hang on if the two main parties were level-pegging. However, the seat is trending to Labour, with 'better than UNS' results in each of the last three GEs. And it would surely be a likely Lab gain if there were a by election. alberto Former Prime Ministers tend to move off into obscurity to write memoirs or make speeches to the few who are still interested in their views. None of the current 5 living ex-PMs is a National Treasure, and neither was any of the previous 5, in my opinion. |
|
|
|
Post by James E on Jan 16, 2022 12:05:45 GMT
I too am very disappointed to see neila decide to leave, but as @crofty says, we have seen a really good influx of new posters, so the judgement has to be that there is more positive than negative in the change. I also wholeheartedly agree with statgeek 's post. I don't think anyone can really argue that posting has been more aggressive here than on the end days of UKPR1, and I have already noted that the rudest and most unpleasant transgressor from the old board - our Trevor - was never challenged at all by those now expressing angsty concern over the tone on here now. I think that says rather a lot. You can't claim any sort of concern for standards if you only ever apply those standards to one set of posters. I would suggest that people follow statgeek 's ABC, and try not to take it too personally if someone has a pop at you. We've all done it now and then. Indeed. Have a look back to his post on 13th Jan 8:17pm (bottom of page 42) accusing me of 'trolling', 'triggering', 'gotcha obsession' , creating a 'cesspit' and being 'at best exaggerated and at worst inacurate' . And at the top, 'Colin likes this'. ukpollingreport2.proboards.com/thread/19/jan-2022-lab-con-ldem?page=42[For the record, my post to JimJam on page 41 (re Turk's comments on Starmer's ratings) which provoked TW's comments was "On the standard 'Best Prime Minsiter' question, Starmer has led Johnson more often, and by wider margins than Labour have led the Tories in voting Intention polls. And looking at polling since Partygate started on 8th Dec, Starmer's leads of 13%, 9% and 7% over Johnson are somewhat ahead of the Labour polling leads which have ranged from 3% to 9%. en.wikipedia.org/wiki/Leadership_approval_opinion_polling_for_the_next_United_Kingdom_general_election#Johnson_vs_StarmerTwo of the recent polls are directly comparable to VI: with Survation (10 Dec) Starmer led Johnson 39-30 in the context of a poll giving Lab a 39-32 lead over Con. And with Ipsos Mori (3-10Dec) Starmer led Johnson 44-31 in the context of a VI lead of 39-34. ] |
|
|
|
Post by James E on Jan 15, 2022 23:49:51 GMT
tancred - a couple of recent quotes from you 13/1/22 (Page 39) A poll shows 9 point Lab lead (Lab 42,Con33) : "This poll is much better and encouraging for Labour. This is the level they need to be at." 15/1/22 (Page 66) Another poll shows a 10 point Lab lead (Lab41, Con 31) : "Pretty low lead considering the mess the Tories are in. I would have thought it would be at least 15 points by now." Hmmm...
|
|
|
|
Post by James E on Jan 15, 2022 23:36:30 GMT
Opinium is the second of the recent polls to show Labour slightly ahead in the South of England (and also by 2 points, 46/34). Sample is 405 in this case. This would be a huge turnaround after the Tories lead of around 30-32 points across SE, SW and East England 2 years ago. On other interesting nugget from this poll is the huge gender gap (samples were a bit over 600 Male and Female). Labour lead by 17 points among women and just 3 among men. It's become quite normal for Lab to poll a net 5-10 points better with women, but the 14 point difference here is the largest I can racall. I appreciate you are normally forensic on these things but are you definitely comparing like with like there? For example does "South of England" definitely exclude London as your comparator seems to, and does "East England" in the comparator go beyond East Anglia and contain some pretty Tory areas further north into Lincs etc? Both YouGov and Opinium lump together the SE England, SW England and East England Government Regions in their 'South' cross-breaks. And both quote separate figures for London. See for example, the most recent YouGov tables here: docs.cdn.yougov.com/48dfh8v55q/TheTimes_VI_220113_W.pdfThat 37/35 Lab lead for 'rest of South' as YG call it contrasts to leads of 33,32 and 30 points (in E, SE and SW) as of 2019. So around a 17% swing. en.wikipedia.org/wiki/2019_United_Kingdom_general_election_in_England(The Wiki link shows the 'regional' cross breaks. And East of England region does not cover Lincs.) |
|
|
|
Post by James E on Jan 15, 2022 22:42:29 GMT
Didn't people used to remark/complain that it was women who kept the Tories in power? I suppose that must have been back in Maggie's day or possibly even earlier. I think it was more like Ted Heath's day, Mercian. More recently, YouGov's analysis showed a 6 point difference in 2019, with the Tories having a 15 point lead with men and a 9 point lead with women. It looks to me like current polling is now showing it as a few points larger. yougov.co.uk/topics/politics/articles-reports/2019/12/17/how-britain-voted-2019-general-election |
|
|
|
Post by James E on Jan 15, 2022 22:31:47 GMT
Opinium is the second of the recent polls to show Labour slightly ahead in the South of England (and also by 2 points, 36/34). Sample is 405 in this case.
This would be a huge turnaround after the Tories' lead of around 30-32 points across SE, SW and East England 2 years ago.
Another interesting nugget from this poll is the huge gender gap (samples were a bit over 600 Male and Female). Labour lead by 17 points among women and just 3 among men. It's become quite normal for Lab to poll a net 5-10 points better with women, but the 14 point difference here is the largest I can recall.
|
|
|
|
Post by James E on Jan 14, 2022 20:00:55 GMT
So do I [Good to agree with an ROC 'across the divide'] |
|
|
|
Post by James E on Jan 14, 2022 11:45:30 GMT
James - Thanks for extracting - your input on polls is appreciated. Forgive my pedantry but the swing is half the movement so 2.6% and as mentioned in the past my view is that Labour need this at around 3% (or 6% closing) to get near to vote share parity with differential turnout and demographic drift closing the rest of the gap. It's roughly 5% of those saying they would vote who are switching from Con>Lab [ (45%x 0.13) - (33%x0.02) ] This (alone) creates movement of around 10%, and constitutes a 5% swing. By way of illustration, the overall VI figures show 23% movement on GE2019 (Con lead of 12 to Lab lead of 11), which is an 11.5% swing. Per YG's figures, just under half of that (or 5.2%) is due to direct switching. |
|
|
|
Post by James E on Jan 14, 2022 11:09:38 GMT
Tables for YouGov here: docs.cdn.yougov.com/48dfh8v55q/TheTimes_VI_220113_W.pdfHighest yet Con>Lab switching of 13% of those who give a voting intention (or 9% of total Con2019 sample) Huge amounts of 'Don't Know' from Con2019 voters: 25% compared to 11% Lab. A 50% reversion of these would cut the Lab lead by 3-4 points. Lab are still a poor 3rd in Scotland, but the cross-breaks for England show huge swings. Lab lead Con by 37/35 in the South. This would be a swing of 16-17% on 2019, and sufficient (just on uniform swing) to turn seats such as Huntingdon, Devon Central and Isle of Wight red. |
|
|
|
Post by James E on Jan 13, 2022 16:48:05 GMT
Turk, Did you see the polling from James showing Starmer running ahead of Labour contrary to what you said. As I said to Colin won't challenge your opinions as they are well opinions but facts is facts as they say. Thank You, JimJam. this is what I posted yesterday: On the standard 'Best Prime Minsiter' question, Starmer has led Johnson more often, and by wider margins than Labour have led the Tories in voting Intention polls. And looking at polling since Partygate started on 8th Dec, Starmer's leads of 13%, 9% and 7% over Johnson are somewhat ahead of the Labour polling leads which have ranged from 3% to 9%. en.wikipedia.org/wiki/Leadership_approval_opinion_polling_for_the_next_United_Kingdom_general_election#Johnson_vs_StarmerTwo of the recent polls are directly comparable to VI: with Survation (10 Dec) Starmer led Johnson 39-30 in the context of a poll giving Lab a 39-32 lead over Con. And with Ipsos Mori (3-10Dec) Starmer led Johnson 44-31 in the context of a VI lead of 39-34. A reminder, too: neither of Starmer's two immediate predecssors achieved a lead even once as 'Best Prime Minister'. |
|
|
|
Post by James E on Jan 13, 2022 12:47:22 GMT
9 point Labour lead with Focaldata
Note that their comparative figures are from 20-21 dec, so at the height of 'Partygate 1'
LAB: 42% (+1)
CON: 33% (-1)
LDM: 11% (+2)
GRN: 4% (=)
SNP: 3% (-1)
RFM: 3% (-1)
Via
@focaldatahq
, 12-13 Jan.
Changes w/ 20-21 Dec.
|
|
|
|
Post by James E on Jan 13, 2022 11:11:08 GMT
tancred It's generally fairly obvious that tactical voting happens in close Con/Lab and Con/LD contests. If you look at the results of any General Election over the past 30 years, the lowest LD vote shares are in the Con/Lab contests and the lowest Lab shares in the Con/LD ones. Incidentally, I think there's good evidence of how LD voters split in the re-allocated votes from my local mayoral election in Cambridgeshire last year. This was contested only by Con, Lab and LD, which meant that when the second preference votes were applied, we got figures for LD voters only. Only 16% did not give a second preference, and of the 84% who did, nearly three-quarters opted for Labour and a quarter for the Tories. en.wikipedia.org/wiki/2021_Cambridgeshire_and_Peterborough_mayoral_election
|
|
|
|
Post by James E on Jan 13, 2022 10:45:43 GMT
What I find interesting about the latest YouGov poll is that Reform UK has seen no benefit from the drop in Conservative support Indeed, it is the lowest Refuk share in the last 10 YouGov polls. One other interesting stat. Looking at polling shares for 'Right of Centre' - so Conservative, plus UKIP, BXP and/or RefUK, the 32% ROC share is the lowest in ANY poll since at least 2005. And it may be earlier than that, because UKIP are not listed separately at that time among 'others' of 8% or so. en.wikipedia.org/wiki/Opinion_polling_for_the_2005_United_Kingdom_general_election |
|
|
|
Post by James E on Jan 13, 2022 10:02:37 GMT
[Model showing that Con 28 Lab 38 gives Labour 326 seats (maj 2) ] It looks to me like this model takes no account of tactical voting. So in their example of a narrow Con hold in Bournemouth West, it's: Con 37.3% Lab 37.3% LD 12.2% Grn 8.9% ref 4.2% Surely, as a matter of common sense, Con and Lab would both be able to squeeze the other parties, here. It would be Labour benefitting mainly, but there are one or two seats (Rother Valley?) where the opposite would apply. |
|
|
|
Post by James E on Jan 12, 2022 23:34:51 GMT
Steve ; "Impact on polling already evident from the normally Tory favouring yougov" YouGov prompt for Greens and Refuk, which tends to reduce Con and Lab somewhat (but Lab more). Also, YouGov tend to be more volatile than other pollsters. Hence if you look back by just over a year to Nov and Dec 2020 (the last time Labour were getting poll leads), YouGov were then producing some of the better polls for Labour. en.wikipedia.org/wiki/Opinion_polling_for_the_next_United_Kingdom_general_election |
|
|
|
Post by James E on Jan 12, 2022 22:51:22 GMT
New YouGov poll has CON down to just 28%. This is 17% down on GE2019 Lab 38 (+1) Con 28 (-5) LDem 13 (+3) Green 7 (+1) RefUK 4 (-1) (via Smithson) The two main parties should be pretty even by March.  |
|
|
|
Post by James E on Jan 12, 2022 19:46:51 GMT
Turk, ''Starmers personal ratings are still way behind his party' Monday's R&W which has the leading 'at the moment' introduction Starmer was on 37% The headline Lab number was 39%. Actually, on the standard 'Best Prime Minsiter' question, Starmer has led Johnson more often, and by wider margins than Labour have led the Tories in voting Intention polls. And looking at polling since Partygate started on 8th Dec, Starmer's leads of 13%, 9% and 7% over Johnson are somewhat ahead of the Labour polling leads which have ranged from 3% to 9%. en.wikipedia.org/wiki/Leadership_approval_opinion_polling_for_the_next_United_Kingdom_general_election#Johnson_vs_StarmerTwo of the recent polls are directly comparable to VI: with Survation (10 Dec) Starmer led Johnson 39-30 in the context of a poll giving Lab a 39-32 lead over Con. And with Ipsos Mori (3-10Dec) Starmer led Johnson 44-31 in the context of a VI lead of 39-34. A reminder, too: neither of Starmer's two immediate predecssors achieved a lead even once as 'Best Prime Minister'. |
|
|
|
Post by James E on Jan 12, 2022 11:21:38 GMT
|
|
|
|
Post by James E on Jan 11, 2022 19:56:04 GMT
I'm old enough to remember Starmer being part of Corbyn's shadow cabinet. I imagine that we're all old enough to remember 2019, Trevor. |
|
|
|
Post by James E on Jan 11, 2022 16:10:53 GMT
"5% of members are female."
With a sample of 54 people, that means that we must have between 2.43 and 2.97 women.
|
|
|
|
Post by James E on Jan 11, 2022 14:49:43 GMT
This poll was released a couple of hours ago: Westminster Voting Intention: LAB: 37% (=) CON: 33% (+1) LDM: 11% (-2) GRN: 5% (=) SNP: 4% (=) Via @savantacomres , 7-9 Jan* Changes w/ 17-19 Dec. *Fieldwork pre-whatever we're calling this one. BYOB-Gate? So right in line with the 4 other pols we've had so far this year. Does anyone think that the main parties will be level by March? |
|
|
|
Post by James E on Jan 11, 2022 10:36:25 GMT
colinFrom the BBC (09:24 today) " Dr Hannah White, deputy director of the Institute for Government, a non-partisan think tank, says it is "extremely unlikely" that Boris Johnson's principal private secretary could have invited 100 people to a Downing Street garden party "without the assent of the prime minister". Asked if she thought Johnson would have had to have agreed to the event, she tells BBC Radio 4's Today programme: "I think he would have had to have done." Citing witnesses who say he attended, she says it's becoming "increasingly difficult" for the PM to claim he didn't know about this particular alleged breach of lockdown rules. White, who used to run the Committee on Standards in Public Life in the Cabinet Office, says it is "pretty surprising" the garden drinks happened at all and "even more surprising" that Martin Reynolds decided to put it on record via email." www.bbc.co.uk/news/live/uk-59948846
|
|
|
|
Post by James E on Jan 11, 2022 9:47:15 GMT
29 Dec 2021: 'Ditching Boris Johnson as Tory leader and replacing him with Michael Gove would hand Labour an 18-point poll lead, a new survey has found.' www.independent.co.uk/news/uk/politics/next-tory-leader-michael-gove-rishi-sunak-truss-b1983691.htmlcolin "I think there is another story emerging-a Civil Service whose senior members are so detached from the real world that they throw a Party to "celebrate the nice weather" when the country is locked away from a global pandemic. That we also have a PM who is ( allegedly) incapable of saying No to these people is the story in the headlines of course. And rightly so because he is accountable to us all-unlike the Mandarins in Whitehall." Probably cognitive bias on my part, but I imagined it was the other way round: I'd assumed that it was Johnson who 'threw a party' (or several) and the Civil Servants were not brave enough to tell him that he must not. |
|
