|
|
Post by James E on Jul 20, 2023 21:50:31 GMT
Just a little more historical perspective on this set of by elections. Take Selby and Ainsty. Should Labour overturn the Tory majority of 20,000, it will be the biggest majority they've overturned anywhere, and in any circumstances, since the Second World War. We're in new territory in terms of Tory by-election defeats if they lose Selby to Labour. I think it was 512th on Labour's target list? No sign of anything other than 3 Tory losses tonight from the twitter vibes. Sometimes you do get a bit of gossip around this time that things may be closer than anticipated. I would also say that on the Anthony article linked earlier- he might be right in one respect about "not telling us anything" about a General Election and I imagine Selby will return to Tory, probably Frome as well, but it does tell us the determination to turn out, albeit with lower turnouts than General Elections. Compared to earlier in this parliament where Labour were underperforming in by elections and locals, they are definitely more in line with opinion polls now I think.Selby and Ainsty is the 117th safest Conservative seat per the 2019 results. Alternatively, it is/was Labour target number 225, so might fall if Labour achieved 427+ seats - or at least 400 if the SNP were to stay dominant in Scotland. www.electionpolling.co.uk/battleground/defence/conservativeRegarding Labour 'underperforming' in by-elections: I think there may be an interesting comparison to be made between Selby, and the results in the 3 'Lab-defence' by elections we had around 6-8 months ago. The current polls with Lab lead averaging 19% are very much the same as they were for Stretford and Urmiston (Dec 2022), and Labour were enjoying slightly higher polling leads at the times of City of Chester (Dec 2022) and W Lancashire (Feb 2023). But in each of those, Labour had a lower swing in the by-elections than the opinion polls at that time, by around 2-5%. However, Labour did achieve a greater swing than the polls were suggesting when winning the Wakefield by-election (June 2022). In other words, Labour seem to be getting better swings where they are attacking than in defending seats they already hold. So if Selby (and Uxbridge?) provide swings of greater than the 15% that current polls are showing, then maybe this can tell us something useful for the next General Election. |
|
|
|
Post by James E on Jul 20, 2023 17:03:40 GMT
Based on fairly limited polling evidence, here are my predictions for today's trio of by-elections: Uxbridge & South Ruislip Lab 42% (+4) Con 37% (-16) swing 10% Selby and Ainsty Lab 42% (+18) Con 36% (-24) swing 21% Somerton and Froom LD 53% (+27) Con 28% (-25) swing Con to LD 26% Anyone else want to have a guess? The first two look quite feasible, James E , but your S&F prediction serms a bit racy to me. Drove through a small town in the constituency earlier and saw plenty of wrinklies near the polling station. I think LDEM will probably take it, but the margin might be less emphatic. We're going down to vote at 8 pm so will give some further thoughts later. For a bit of context to my S&F prediction - look at the Con to LibDem swings in the three LD by-election gains we've seen in the past few years: 17 June 2021 Chesham and Amersham : 25% swing Con to LDs (when Con VI was averaging 44% in polls) 16 Dec 2021 North Shropshire : 34% swing Con to LDs (when Con VI was averaging 33% in polls) 23 June 2022 Tiverton and Honiton: 30% swing Con to LDs (when Con VI was averaging 32% in polls) The Tories' current polling average is 27%, while the LDs are at much the same VI as they were in these previous BEs. For the Conservatives' vote share in Somerton&Froom: UNS would have them down by 18 points to 35%. Proportional loss would have them on about 32 % (27/44.7 x 53%) By-election history suggest that their vote share may be halved to around 26%. |
|
|
|
Post by James E on Jul 20, 2023 15:35:18 GMT
Based on fairly limited polling evidence, here are my predictions for today's trio of by-elections:
Uxbridge & South Ruislip
Lab 42% (+4)
Con 37% (-16)
swing 10%
Selby and Ainsty
Lab 42% (+18)
Con 36% (-24)
swing 21%
Somerton and Froom
LD 53% (+27)
Con 28% (-25)
swing Con to LD 26%
Anyone else want to have a guess?
|
|
|
|
Post by James E on Jul 18, 2023 10:32:00 GMT
New Delta poll, recent polls have shown a slight narrowing, so this may be an outlier 🚨New Voting Intention🚨🚨 Labour lead is twenty-four percentage points in the latest results from Deltapoll. Con 24% (-4) Lab 48% (+2) Lib Dem 11% (+2) Other 17% (-1) Fieldwork: 14th - 17th July 2023 Sample: 1,000 GB adults (Changes from 7th-10th July 2023) Deltapoll quite often seem to be on a different trajectory to the rest of the polling industry. While most current polls currently show the Labour lead slightly lower than it was at the start of 2023, Deltapoll's last 5 polls average a lead of 21% compared to 16% in their first 5 2023 polls. And i remember the opposite happening at one point last year, when their figures narrowed, again contrary to the general trend. I suspect that they may have made unannounced methodological changes. Not all pollsters are as open as Opinium of YouGov as to their methodologies. |
|
|
|
Post by James E on Jul 17, 2023 14:11:08 GMT
James. How much of the age remain/leave split is actually educational with more graduates in the under 35s? That's very hard to quantify, but the consensus seems to be that there is an age-effect superimposed on the education effect. In particular, it should be noted that only half of the 18-24 cohort (who voted 75/25 for Remain per Ipsos, or 71/29 per YouGov) was old enough to hold a degree, which makes any conclusions in this group very tricky. The best I can offer you is probably the 25-34 age group which Ipsos found voted 60/40 for Remain. If half of that group are graduates, and they followed the average graduate split of 68/32 for Remain, then that half of the cohort was 34/16 for Remain. It would follow that non-graduates aged 25-34 also voted Remain by 26/24, whereas YouGov have all non-graduates dividing 41/59 for Leave. Conversely, the over 55s (or over 65s) divided more strongly for Leave than that overall 'non-graduate' split of 41/59, despite containing maybe 5-10% graduates. So it must be clear that education alone cannot explain the age-patterns found by Ipsos or YouGov. d25d2506sfb94s.cloudfront.net/cumulus_uploads/document/oxmidrr5wh/EUFinalCall_Reweighted.pdf |
|
|
|
Post by James E on Jul 17, 2023 10:47:37 GMT
Colin (re likely consequences of brexit): "Don't the current polls indicate opinion on that question ?"Yes indeed, given the large swing to 'wrong to leave'. And re your: "I think Social Grade is interesting too-C1s having been the group to majorly vote Remain whilst all others voted majorly Leave. The Working Class by a large majority."
You appear to have got your psephological knickers twisted. Far from "all others voted majorly Leave" this is what the Ashcroft report actually said: "The AB social group (broadly speaking, professionals and managers) were the only social group among whom a majority voted to remain (57%). C1s divided fairly evenly; nearly two thirds of C2DEs (64%) voted to leave the EU." To adopt Sunspeak, it was the C2DEs wot dun it. I would argue that age was the more defining demographic than social class in the 2016 referendum. Per Ipsos Mori's analysis at the time, Under 35s of all social classes showed a majority for Remain, and over 55s of all classes had majorities for Leave. Their figures include: ABs, aged 55+ : 48% Remain, 52% Leave C1s, aged 55+ : 37% Remain, 63% Leave and C2s, aged 18-34: 54% Remain, 46% Leave DEs, aged 18-34: 56% Remain, 44% Leave. It's notable that in both the 18-34 and 35-54 age cohorts, the lowest social class (DEs) were slightly more Remain-inclined than the C2s. www.ipsos.com/en-uk/how-britain-voted-2016-eu-referendum |
|
|
|
Post by James E on Jul 15, 2023 19:12:50 GMT
"While the boundary changes generally favour the Conservatives..."
It's certainly true that the Tories will go into the next election with notionally more seats than they won in 2019. However, when I have experimented with Electoral Calculus, applying the same vote shares on the 2019 and 2023 boundaries, it seems to make very little difference. At worst, Labour appear to take about 3-5 fewer seats, and in some scenarios, they take 1 or 2 more.
I would be interested to see ordered lists of Lab and LD targets on the new boundaries - has anyone come across these?
|
|
|
|
Post by James E on Jul 14, 2023 15:45:32 GMT
"I always felt that significant benefits wouldn't be apparent until 2030 or so...." (Mercian) Actually, the late TOH himself was quite revisionist over this. He did not claim that the benefits of Brexit would not appear until 2030 until some time after the referendum, and neither did any of those who campaigned for Leave. I did write down somewhere TOH's comments at the time when May triggered Article 50 in March 2017: the gist was that he expected difficulties for the UK when we were in the process of leaving, but that once we actually left, there would be a major boost to both GDP and investment in the UK. It struck me at the time as likely to be disproved, and it has been. The 'wait until 2030 before we can evaluate Brexit' argument wasn't seen until the talks started to founder in late 2017-2019. Exactly. The actual process was handled so badly that he became more cautious. As Keynes said " When the facts change, I change my mind. What do you do?" What has happened since the Referendum has been entirely consistent* with the facts as I understood them back in 2016. [Edit: for the benefit of Trevor (below) - I expected the Brexit negotiations would go rather badly, and that we would suffer economically ] |
|
|
|
Post by James E on Jul 14, 2023 12:06:26 GMT
... Also along with the late TOH I always felt that significant benefits wouldn't be apparent until 2030 or so. It could be later because of all the faffing about in Parliament. The point is not whether Brexit has had some marginal downside or not, but it is a reality for the foreseeable future. UKIP started around 1990 I think and it took 25 years to get to a referendum. I wish you and your comrades well in your forthcoming 25-year battle. Actually, the late TOH himself was quite revisionist over this. He did not claim that the benefits of Brexit would not appear until 2030 until some time after the referendum, and neither did any of those who campaigned for Leave. I did write down somewhere TOH's comments at the time when May triggered Article 50 in March 2017: the gist was that he expected difficulties for the UK when we were in the process of leaving, but that once we actually left, there would be a major boost to both GDP and investment in the UK. It struck me at the time as likely to be disproved, and it has been. The 'wait until 2030 before we can evaluate Brexit' argument wasn't seen until the talks started to founder in late 2017-2019. |
|
|
|
Post by James E on Jul 14, 2023 11:08:46 GMT
You have to remember that if (when?) the country starts to discuss rejoining, the debate will swiftly become about freedom of movement and how much migration from the continent people are willing to accept. That will certainly shift those polling numbers. Given how close it was last time it is reasonable to assume that there would still be a majority for rejoin, but the debate would be acrimonious and the country badly split once more. I don't think there is much appetite at Westminster to reopen that just yet (for one thing it would consume a huge amount of government time and effort and prevent much else getting done - as was the case 2016-19). This explains Labour's position and it is my belief is that the Lib Dems will also fall short of advocating rejoining at the next election (TBC on that one). Totally agree with you on this. Labour MP's who speak at meetings I attend talk about neither the Labour Party or the voters having the bandwidth to contemplate rejoining at present; I do think things can change quite quickly though, I don't think it's necessarily a once in a generation decision. Things have changed quickly over the past 12 months. Looking at YouGov's Brexit 'hindsight' tracker, neither side was able to get to (or even near to) a 60/40 lead until mid 2022. But a year ago in July 2022 we had the first 40/60 lead for 'Wrong to Leave' and now from Sept 2022 onwards every YouGov poll on the issue shows 60%+ thinking the decision was 'Wrong'. www.whatukthinks.org/eu/questions/in-highsight-do-you-think-britain-was-right-or-wrong-to-vote-to-leave-the-eu/?pollster%5B0%5D=yougov&removedThe overall picture is that the figures moved back and forth for 5 years up to Summer 2021 when the 'right' side was enjoying a boost due to the fast vaccine roll-out. From then on, it has been a one-way street. Taking the figures from around mid-summer for the past 3 years, the averages are: June/July 2021 48% Right, 52% Wrong. June/July 2022 42% Right, 58% Wrong. June/July 2023 37% Right, 63% Wrong. Another 4% movement would take us to a position where more than two-thirds consider the decision to be wrong. And I'd accept that the 'Rejoin/Stay Out' polls show a lower lead for the 'Rejoin' option (typically 58/42) but these are close to becoming a consistent 60/40 lead, and age-demographics alone would take us to 60/40 in around 3 years. And then there's the YouGov poll from May 2023 on whether voters consider Brexit a success. Just 9% of respondents said it had been a success, compared to 62% who say it's a failure. And even among Leave voters the figures show a plurality for 'Failure' (37%) over 'success' (20%). So there is room for further movement there. [EDIT - Re Leaver disilusionment. See the graph in the middle of this article from Nov 22. While the proportion believing Brexit to be the wrong decision has accelerated since mid-2021, it has really bean a very slow, steady decline. There is little impact of 'events' as seen in VI polling.] yougov.co.uk/topics/politics/articles-reports/2022/11/17/one-five-who-voted-brexit-now-think-it-was-wrong-d |
|
|
|
Post by James E on Jul 13, 2023 10:32:01 GMT
Some rather revisionist comment by Peter Kellner, who now sees a Labour majority in the next General Election as a 50/50 chance. I do wonder whether he us placing any bets on a hung parliament which is available at 14/5 with the bookmakers. kellnerpolitics.com/2023/07/12/labour-has-a-50-50-chance-of-an-overall-majority/As I have mentioned before, I disagree with his contention that UNS will provide a decent model for seats at the next GE, and it now seems that he' accepts this. His latest blog states that "Taking everything into account, Labour may need a lead of 8-10 per cent in the Britain-wide popular vote to secure an overall majority in the next House of Commons " whereas just 7 weeks ago he was still arguing that Labour needed a 12-13% lead (as per UNS) in his 'grumblers and defectors' article. But other views are available. John Curtice has argued that the May 2023 Local Election results put us back to around the position at GE2017, leaving Labour needing around a 7% lead for a majority. Professor Rose on Electoral Calculus has argued that a 5% lead is needed, although it should be noted that this is without tactical voting. EC itself shows Labour can take a majority on a lead of 3% if there is Lab/LD tactical voting, which I would have thought is a near-certainty. My own view is that Prof Rose is the probably right. It's also worth noting that on vote share he is says that "I currently expect Labour to end up with 40-44 per cent of the vote, while the Tories win 31-35 per cent" - very close to most people's current expectations. The mid-point of this (Lab 42, Con 33) produces a comfortable Labour majority on all projections other than UNS. |
|
|
|
Post by James E on Jul 12, 2023 9:32:26 GMT
While I'm on the subject of the R&W 'walls': their most recent 'Blue Wall' figures show a swing 3% ahead of their comparative GB polls. This is with a sample of seats which voted Con 50%, LD 27%, Lab 21% in 2019.
7th May 4% Lab lead. 16.5% swing v 12% GB
22nd May 1% Con lead. 14% swing v 12% GB
4th June 4% Lab lead. 16.5% swing v 13% GB
18th June 7% Lab lead. 18% swing v 16%GB
2nd July 7% Lab lead. 18% swing v 15% GB
So these average as a 16.5% 'Blue Wall' swing compared to 13.5% for GB.
|
|
|
|
Post by James E on Jul 12, 2023 9:10:20 GMT
Have we looked at the most recent R&W 'Red Wall' which was released yesterday? Figures, with comparisons to 2 weeks ago are: Labour 52% (-1) Conservative 27% (+1) Reform UK 9% (–) Liberal Democrat 6% (–) Green 4% (–) Plaid Cymru 1% (–) Other 2% (+1) redfieldandwiltonstrategies.com/latest-red-wall-voting-intention-9-july-2023/This is a 17 point swing compared to GE2019 compared to 16 points in the most recent R&W (Lab 48, Con27) This seems to be consistent with the trend of the past couple of months with this sample. Previously, their 'Red Wall' polls tended to show a swing a point or two lower then average than their full GB polls, which is what might have been expected with the Conservatives getting an incumbancy boost in seats they gained at the last GE. 23rd May 23% Lab lead. 16% swing v 13%GB 28th May 17% Lab lead. 13% swing v 13%GB 11th June 22% Lab lead. 15.5% swing v 13% GB 25th June 27% Lab lead. 18% swing v 15%GB 9th July 25% Lab lead. 17% swing v 16%. So the average 'Red Wall' swing is now 16% compared to 14% in R&W's GB polls on the same dates. Incidentally, this does look consistent with 'proportionate swing' for this sample, as these seats voted Con 46.7%, Lab 38%, BXP 6.5%, LD 4% in 2019. The differences to UNS become greater in safer Tory seats. |
|
|
|
Post by James E on Jul 11, 2023 10:17:20 GMT
I think if you have to use a word that few people have heard of to describe a policy it's a waste of time. Depending on what it entails, I actually don't think this is a bad idea. I'm remembering a friend, who went to a private school for a few years and had something like wwhat is being proposed. It wasn't about accents or sounding posh, but, things like learning to project your voice. A skill tyhat I, like most people, sadly lack. Had I not known my friend, I guess I would have thought this was a silly idea/waste of time/gimmick, but, having seen how this (what she calls a life skill) has helped her, I'm for it. Oracy is not a new thing. I have been a Primary School Governor since 2005, and recall our first Headteacher, who retired in 2009, explaining the concept and how it can be promoted in our school. So that must have been around 15 years ago. And it really isn't about accents nor projecting your voice. It's more a matter of learning the skills to express yourself well verbally, through things such as story-telling, debating and presenting. |
|
|
|
Post by James E on Jul 10, 2023 18:04:19 GMT
I don't think we've seen this one as it's only just appeared on Election Maps twitter, but there have been 7 other polls with more recent fieldwork. Westminster Voting Intention: LAB: 45% (=) CON: 30% (+2) LDM: 11% (=) SNP: 3% (=) GRN: 3% (=) RFM: 3% (-2) Via @survation , 30 Jun - 2 Jul. Changes w/ 23-26 Jun. With 10 polls released so far with fieldwork in July, the average Labour lead is exactly 20%. en.wikipedia.org/wiki/Opinion_polling_for_the_next_United_Kingdom_general_election |
|
|
|
Post by James E on Jul 10, 2023 17:20:01 GMT
Full results "Labour leads by 21%, the largest lead for Labour since 19 March. Westminster VI (9 July): Labour 48% (+2) Conservative 27% (-1) Liberal Democrat 11% (–) Reform UK 5% (–) Green 4% (-1) Scottish National Party 3% (–) Other 1% (-1) Changes +/- 2 July" As so often with R&W, beneath the apparently 'normal' headline figures there are some really odd details. They have the Conservatives ahead by 43/40 in London, the 4th time in their past 7 polls in which this has happened. And this is not a small sub-sample, as they had well over 500 respondents in London, so more than double the number they needed within an overall sample of 2000 ( the London sample is down-weighted from 569 to 280). They also have the Tories ahead in the West Midlands, which is quite a turnaround, as their polls in May and June averaged a Con to Lab swing of well over 20% there. On the other hand, Labour will be more than pleased to have a lead of 36 to 34 with the over 65s (N=415), and delighted with their 61/27 lead in the Eastern region (N=126). Their figures by past vote look credible enough (e.g. 17% of Con2019 to Lab, 1% of Lab2019 to Con) although as always, they find higher levels of voter retention than most other pollsters. But there is something very strange in their sampling. |
|
|
|
Post by James E on Jul 9, 2023 14:15:04 GMT
FM - re ''What definitely has the potential to be hilarious  (as I suggested yesterday) is the choice Tory members (probably the least well equipped group in the UK to judge a person’s suitability to become party leader) make for next Party leader.'' I agree but after 2 terms in opposition the desire to get elected takes over. So Cameron beats Davis and for Labour, who have similar membership/voter disconnects, Starmer takes over from Corbyn. My expectation would be for the Tories, should they as seem most likely lose, elect the most right wing leader twice but for the person with a chance of victory (or at least losing with dignity) to take over after a third GE defeat. NB) I say twice but I guess could be more, I am thinking for 2 GEs more really. Interesting to note YouGov's most recent table for 'Most Popular Conservative Politician'. yougov.co.uk/ratings/politics/popularity/conservative-politicians/allThere's a distinctly "has-been" look to the Top 20 with 5 of them over 80, and one who is no longer even alive. Only 2 of the top 10 are still in Parliament, these being Sunak himself (No 1) and Theresa May at No 5. Ben Wallace at No 11 is the highest of any sitting MP other than May and Sunak, and he's closely followed by Patel, Braverman, Rees-Mogg and Morduant. |
|
|
|
Post by James E on Jul 8, 2023 19:36:42 GMT
I assume you're joking. Just in case you're not what on earth has 'tackling obesity' got to do with TfL, or any organisation for that matter? If people slim down then this helps TfL, because then they can fit more of them onto the trains.  |
|
|
|
Post by James E on Jul 8, 2023 19:20:12 GMT
Opinium
Latest poll for
@observeruk
Labour lead at 15 points.
Labour: 43% (-1)
Conservatives: 28% (+2)
Lib Dems: 9% (n/c)
SNP: 3% (n/c)
Green: 6% (-1)
Reform UK: 8% (+1)
(Changes are from a poll released in the Observer last week)
|
|
|
|
Post by James E on Jul 8, 2023 14:38:12 GMT
Just as Brexiters couldn't tell us what leaving the EU would mean in practical terms, until they completely botched it up and the disaster was clear for all to see, those same Brexiters can't tell us what rejoining would mean in practice because it needs a negotiation, and no-one knows what the outcome of that would be. The EU might quite like to have a UK on board, now that the anti-European boil has been lanced and the loons ad fruitcakes have been exposed for the blowhards they are, but we don't know for sure how many concessions they might be prepared to make to bring a weakened and chastened UK back into the fold. So anyone - on here, or in the press - who says rejoining would mean X,Y or Z with certainty is just talking through their hat. I think a willing UK is likely to get a better deal than many think, as we are still an economy with great potential and with a world view that chimes with many of the northern EU states, so we still have the potential for some support in there for rejoining. But I don't know for sure. It is very easy to lose the trust of others; far harder to regain it. With the EU already having trouble with recalcitrant members like Hungary and Poland it is very difficult to see them welcoming another. I expect them to demand both immediate acceptance of Schengen and replacement of the £ with the € as conditions of re-joining. Losing control of the currency that a Government borrows in is not to be taken lightly as countries like Greece and Italy have found. Re joining the Euro and EC Membership. I believe this is a fair explanation: ukandeu.ac.uk/explainers/how-new-member-states-join-the-eu-all-you-need-to-know/" Do new members have to join the euro?
Apart from Denmark, which has a permanent opt out, all current EU member states are expected to join the euro – although this depends on meeting certain economic criteria and there is no set timetable in which they have to do so. Countries can, therefore, join the EU before joining the euro. For instance, six of the 13 countries that have joined the EU since 2004 have still yet to join the single currency. There are four criteria that member states have to meet to do so: keeping inflation low and stable, ensuring public debt and the deficit are within EU rules, keeping interest rates low and tying the national currency to the euro for at least two years without serious issues, such as needing to devalue. Politically, there would be substantial pressure on new member states to join the euro, not least as they would have joined the EU knowing eurozone membership was an expectation. Having said that, once countries have become EU member states there is no mechanism to force them to join the euro nor to sanction them if they fail to do so." That final sentence is the part to bear in mind. There is no mechanism to force member States to join the Euro. So when someone claims that the UK would be 'forced' to join the Euro, the implication is that new and different rules would be created to treat the UK differently from the other non-Euro, EU Member States. |
|
|
|
Post by James E on Jul 8, 2023 13:27:19 GMT
Dont post that often but a little info on that Uxbridge poll showing labour only gaining 3.5% on the 2019 vote. Dont know how reliable the JL Partners poll is but this seat has little in common with most other London seats. Heavy favourites with bookies but would imagine Lab not taking this for granted. The northern part is heavily traditional Con, Boris type supporters, non transient white working class tradesmen types and some professionals. Stop for a pint sometimes on way back , these places could`nt be much more different. Places like Ickenham and Hillingdon still have a village feel to them, often overhear groups of workers having a pint referring to lefties and whilst Con support is 20% down in the poll many of these will not switch to Lab but opt for fringe parties, the ULEZ issue is having some effectThat Uxbridge constituency poll seems consistent with the cross-break analysis for London from YouGov and Opinium I provided last week. 10 YouGov London cross-breaks from May and June 2023 averaged: Lab 53% (+5) Con 20% (-12) So just a 8.5% swing compared to GE2019 (c/f 15% for GB as a whole). The 4 Opiniums for the same period average as Lab 54% (+6), Con 23 (-9) in London, so just an 7.5% swing. Deltapoll for the same period show both Con and Lab down 3 points in London, so no swing at all from GE2019, but I would have less faith in their figures, nor those of R&W, whose show Labour losing support in London. In fact R&W show Labour leading by more (8 to 28%) in all other regions of England than in London (4%)! The only recent London poll we have had was YouGov's 58/18 in late March, which I suspect was on the high side, with a 40% Lab lead, as the Opinium and YG figures above suggest more like 30%. But even that showed a lower Con to Lab swing of 12% compared to the GB-norm of around 16% then (and now). |
|
|
|
Post by James E on Jul 7, 2023 17:41:15 GMT
James, Re False recall. IMO, it is definitely a thing but I wonder if the one of the drivers the produces false recall also overstates it. I am thinking of the halo affect where people for some reason who voted for someone they would rather not have (or don't want to admit to voting for) or didn't vote at all, claim even in anonymous on-line surveys that they voted differently; the net result in this case is less Tory voters acknowledging their 2019 vote. By the same token, though, there will be some non voters (perhaps a few who did vote for someone else) saying just after the election that they voted for the winning team. Responses from panelist before GEs and just after form the basis for the false recall adjustment which if I am right may be overdone. Intuitively I think the real picture would be close to the YG/Opinium numbers of retention but not quite that low, maybe high 40s or around 50%. We will never know though. I thought I would test this by checking the earliest savanta polls from this Parliament. I have also looked at three from just before Partygate in Nov-Dec 2021. There were only two Savanta polls in the first 6 months after GE2019, one in Feb 2020 and one in May. Recalled past vote in these averaged: Con 43% Lab 35% For the 3 before Partygate (19 Nov - 5 Dec 2021) it was: Con 40% Lab 37% And now the most recent 4 show: Con 40% Lab 41% |
|
|
|
Post by James E on Jul 7, 2023 10:57:59 GMT
My post above about Savanta also reflects the difference between panel and 'recalled past vote' pollsters. Taking the recalled-past-vote ones first, here are Savanta's and R&W's most recent Con2019 cross-breaks:
Savanta
Con 64%
Lab 11%
LD 5%
Don't Know 13%
R&W
Con 59%
Lab 14%
LD 3%
DK 17%
But with the Panel Pollsters...
YouGov
Con 42%
Lab 11%
LD 2%
DK 23%
Opinium
Con 44%
Lab 13%
LD 3%
DK 24%
|
|
|
|
Post by James E on Jul 7, 2023 10:34:00 GMT
Savanna Poll, it appears their poll last week was a 'rogue' one 📈18pt Labour lead 🌹Lab 46 (+3) 🌳Con 28 (-3) 🔶LD 11 (+1) ➡️Reform 4 (-1) 🌍Green 4 (+1) 🎗️SNP 3 (-1) ⬜️Other 4 (=) 2,216 UK adults, 30 June - 2 July (chg from 23-25 June) That's Labour's (equal) highest lead with Savanta for 5 months - or 20 of their polls. Looking at their figures over that time, they generally show Labour leads around 3-4 points lower than the average. I don't normally bother checking Savanta's tables. The latest version (linked below) are at least a bit more readable than the 250-page version they used to provide. But I'm glad I have had a look as they are really interesting - not so much as for what they show for the next General Election as for how their respondents recall voting at the last one. We don't have figures for this latest poll, but these are the recalled 2019 votes of their previous 4 samples: 2-4 June: Con 40%, Lab 42%, LD 8% ( 556, 585, 115 per table 1) 9-11 June: Con 42%, Lab 41%, LD 6% (572, 556, 82 ) 16-18 June : Con 41%, Lab 40%, LD 8% (579, 560, 109 ) 23-26 June: Con 39%, Lab 43%, LD 7.5% (571, 633, 112) In each case, these are then re-weighted to more closely resemble GE2019 (table 12); it looks like they do make allowance for the demographic age shift, which alone reduces the Con vote and increases Lab by about 2 points, and/or new voters and non voters.. Needless to say, the re-weighting has a sigificant effect on the headline figures: that 43/31 poll from 25 June was 53/25 in the raw, unadjusted figures, while the other 3 polls show Labour 26 or 27% ahead in the raw figures. savanta.com/published-polls/ |
|
|
|
Post by James E on Jul 7, 2023 9:39:39 GMT
Re Selby, Based on national polls at the time of the resignation, applying proportionate swing adjusted for Tory By-Election slippage and tactical voting, I suggested that anything short of a victory by at least 5% would actually be disappointing for Labour. Since that time Labours' lead over the Tories has widened so a victory of just over 10% would not be surprising and in line with the adjusted national picture. Off there tomorrow for a days campaigning which will be interesting as my first time campaigning in a solid 'Tory' seat. Wakefield was about getting former Labour voters back or at least not voting Tory so I will be intrigued to see what the messaging is tomorrow when briefed. The odds for the Selby by-election with Betfred have now shifted to Lab 4/11, Con 2/1. My own prediction from the time Adams resigned the seat was a narrow Lab win. A very simple model of the Conservatives losing 40% of their vote from 2019 and Lab gaining 60% of that lost vote produces Con 36%, Lab 39%. This may well understimate the Tories losses. Although they are less common than LibDem by -election 'surges', there are precedents for huge Con to Lab swings in by elections when the Tories are really unpopular. Looking back to the 1992-97 Parliament, there were 4 where the swing exceeded even the highest swings of any constituency at the 1997 election (19%), let alone the overall GE1997 swing of 10-11%. These were Barking in 19994 (22%), Dagenham 1994 (23%), Dudley West 1994 (29%, and the current record) and SE Staffordshire 1996 (22%). The Tories vote retention in these 'early-Blair' by elections was worst in Barking (31%) and Dagenham (27%), although in each case this was partly due to losses to other parties of the right, much like to Corby 2012. Dudley West was the worst for the Tories in a seat they held, at 38% vote retention (or 48.9% to 18.7%). It looks unlikely that the 29% swing record will be broken, but local and national polling would suggest a similar swing to SE Stafffordshire in 1996 or Dagenham in 1994. The other comparison I'd make to those 1994-97 by-elections is what happens to the Conservative vote. I've noted that the the largest losses were in Barking and Dagenham, where far right parties took a larger share, but if you look at other by-elections from that time (e.g Eastleigh on the same day) it seems that the Tories vote was normally halved from GE1992. So the pattern I have noted of the Tories faring worse where they are defending is a phenomenon of the current parliament, rather than the historic norm. |
|
|
|
Post by James E on Jul 6, 2023 11:20:19 GMT
Savanna Poll, it appears their poll last week was a 'rogue' one 📈18pt Labour lead 🌹Lab 46 (+3) 🌳Con 28 (-3) 🔶LD 11 (+1) ➡️Reform 4 (-1) 🌍Green 4 (+1) 🎗️SNP 3 (-1) ⬜️Other 4 (=) 2,216 UK adults, 30 June - 2 July (chg from 23-25 June) That's Labour's (equal) highest lead with Savanta for 5 months - or 20 of their polls. Looking at their figures over that time, they generally show Labour leads around 3-4 points lower than the average. |
|
|
|
Post by James E on Jul 6, 2023 11:15:14 GMT
Just because the polling on independence and republicanism is similar doesn't mean it is the same people in both polls. For example I know Liberal Democrats who are republicans, and I have no doubt that is also true in the Labour party. You could very well have majorities for both but only a minority in favour of both, which was the point I made. Alex Salmond said he would keep the monarchy; Nicola Sturgeon as far as I can remember never said anything either way; it is Hamza Yousef who has come out as a republican. Yousef could become the SNP's Sunak, leading them to defeat. Yes, it isn't simply the same 47% (or so) who support a Republic as those who want Independence. YouGov's most recent figures show 2014 "Yes" voters dividing 57/31 for an elected Head of State over the Monarchy. 2014 "No" voters divide 67/24 for the Monarchy, which is similar to as GB voters as a whole. But the point stands that Scotland is considerably less Monarchist than Britain as a whole. The 3 polls referenced in the 'whatscotlandthinks' link above show the pro-monarchy majority is around 25 points lower in Scotland. There is already a small-ish majority for a Republic among Scottish voters aged under 65. And in the same article, John Curtice notes the falls in support for the Monarchy among both Yes and No voters over recent months. docs.cdn.yougov.com/xfaufs6rcp/InternalResults_230420_Scotland_Royals.pdf |
|
|
|
Post by James E on Jul 5, 2023 21:46:44 GMT
oldnat "I agree that, since Flynn made those comments in May, the likelihood of England not installing a majority Labour government is much reduced." Labour's GB lead was averaging 16% when he made those comments, as opposed to 19% now, so really not much has changed. I think that the odds for a hung parliament have become very slightly longer. I picked up on his comments at the time (15 May, page 82 of this thread) and offered much the same coments then..... "May 15, 2023 at 11:35am neilj, moby, and 5 more like thisQuoteEditlikePost OptionsPost by James E on May 15, 2023 at 11:35am "Flynn said: "It's increasingly clear that the SNP can hold the balance of power after the next general election - putting Scotland in prime position to pull the strings of a minority UK government." It looks to me like this is another of those judgements based on UNS - and the false narrative that Labour need a lead of 13% in the overall vote to form a majority. My own calculations suggest that the point that Labour can get an overall majority is around 4% lead - such as 39/35. Electoral Calculus suggests a little lower with tactical voting. I am working on an assumption of just 5-10 Lab MPs in Scotland: the same dynamics which show Labour doing far better than UNS in England on current polling also suggest that they would underperform against the SNP in Scotland unless they can draw level in Scottish vote share with them."
|
|
|
|
Post by James E on Jul 5, 2023 21:12:12 GMT
....Mr Flynn's boastful assertions that Labour will depend on SNP support at Westminster if they are to form a government after the next election, will sound like hot air.
For clarity, this is what Flynn said in May 2023 -
“Everyone’s of the view that Keir’s going to be the next prime minister, and he’s likely to be the prime minister of a minority government...." [emphasis added]
So Flynn is predicting that Labour will be short of a majority. Anyone who believes this should be able to make some money on the betting markets, where a Labour majority ( and that means 326 seats or more) is priced at around 4/9 on. 'No overall majority' is priced at 5/2. www.oddschecker.com/politics/british-politics/next-uk-general-election/overall-majorityThis is because of the large Labour leads in GB polling, and the patterns in regional and other detailed data suggest that Labour would get a majority on around a 4-5% lead, even if the SNP remains the largest party in Scotland, which I believe it probably will. For what it's worth, the kind of result which would (in my opinion) enable the SNP to hold the balance of power would be something close to parity between Con and Lab: probably from a 3% Con lead to a 2% Lab lead, as with anything more than that a Lab+LD deal of some sort would probably be viable. |
|
|
|
Post by James E on Jul 5, 2023 17:27:00 GMT
|
|
(as I suggested yesterday) is the choice Tory members (probably the least well equipped group in the UK to judge a person’s suitability to become party leader) make for next Party leader.''
