United Kingdom General Elections: December 12th, 2019

[Oryxslayer]

Remember that all this is is a demographic modelling tool that projects a large national survey down to constituency level. It relies heavily on a series of assumptions, and if any are even a little bit out then the projections will be wildly off. It's clever, but no more inherently accurate than doing a quick swing calculation and checking out the constituency pendulum.

Or, rather... everyone remembers that the model called Canterbury correctly last time. Everyone forgets its projected figures for Finchley & Golders Green...

This brings to mind the aphorism "All models are wrong, but some are useful." From what I've seen I find this model is useful at painting a picture at what a Conservative victory with a 10 point lead may look like, and for identifying marginals.

Good post. In two weeks time, when the model will likely have been fed more recent data, this model will be very useful at telling us what is going on in about 550 of the seats. Until then, it's just an accurate snapshot in time. 538 does it all the time and nobody is beseeching the heavens like the model is the word of god.

However, once you have entered a variable into the environment, it effects the resident subjects. People who see this model will act upon it. Why, already both Labour and the Lib-Dems are adjusting strategy, and those are just the  most important actors.  

[Oryxslayer]

Why is everyone freaking out over some polls two weeks before the election? Things can swing as much as ten points in two weeks without major news, and no one knows if major news will drop during that time in any case.

It's a model with over 100K respondents (ginormous compared to normal polls), targeted weights, was accurate in the past, and there is a good sum of money behind the model to ensure  it comes out right. All told, it's very similar to how 538 does their stuff, only it's coming from an inhouse pollster rather than a aggregator. Quite a lot of people trust 538, so you can see how a situation develops.

[Oryxslayer]

Anyway there is currently a 'debate' going on about the climate on Channel 4. Boris and farage ducked, and Channel 4 put up Ice sculptures to empty chair them. The problem? Placing the Ice sculptures implies a message rather than an empty chair, and implying a message is a partisan act. It hands BoJo a silver platter case that the debate was a partisan sham to begin with, especially since it looks like they tried to send Gove. I'm sure to the Tory voter it will soon look at this event as a case where they were right to duck. Whatever.

Here's their paper:

[Oryxslayer]

Ipsos-Mori for Scotland.

SNP 44% (+7)
CON 26% (-3)
LAB 16% (-11)
LD 11% (+4)
GRN 2% (+2)

Are the changes from 2017 or from the previous poll? And if the latter, when was it taken?

From the last election.

And I am calling out that Labour figure as too low.

I mean Scottish Labour has had a few bad months, and they somewhat backed a Yes vote when their base was previously unionist. In fact, I doubt the unionist parties really lost voters and more the SNP activated their base when compared to 2017, which was an election with low nationalist turnout. Remember, the youth is Yellow up there. On the other hand, their incumbents will benefit from unionist tactical voting, so, who knows what happens.

[Oryxslayer]

BBC supposedly has a clip of Gove and the rest of the Tory team turning up for the debate, but then it was the channel 4 dudes who told them they would accept nothing but Boris. This may turn... Trumpy, as Boris now can reorient back to his talking points about People vs Remain Elite/Parliament. *Shudders*

[Oryxslayer]

BoJo's warning Trump not to make brash statements about the election when he comes to the Nato summit. BoJo, after all, doesn't want to be seen as Trump's mini-me: the minority of trump approvers in the UK are likely already voting Con/Brexit/DUP.

There's also a 7 way debate in Cardiff tonight, but BoJo, Corbyn, and Farage are sending replacements. I get the feeling that this debate may have a lot of security related questions after what happened today.

[Oryxslayer]

This is probably more out of ignorance than anything else, but I was under the impression the Tories were pretty climate friendly when compared to other center right and right wing parties in the west?

Depends on who you are comparing the  Conservatives to. If you are comparing them to other deep-blue Right-wingers around the globe, then the Tories are more ecologically friendly than the rest. This is because they have historically faced Green party and Lib-Dem assaults in their southern shires based on ecological issues. It's mostly a bottom up effect from various council-level challenges which influences the MPs positions, even though their seats are safe. Sometimes this is more local in focus (don't cut down the trees on 'x' street, preserve the Greenbelt culture of our town), and sometimes this is more Global (recycling bins and programs to reduce  litter, insulating homes to lower electricity usage). However, if you are comparing to other UK parties, the Tories are the most ecologically right-wing party that actually has a chance and winning seats in the main 632.

[Oryxslayer]

Mods, can we please kick out the troll?

[Oryxslayer]

The biggest difference between ComRes and BMG is probably the polling dates. BMG polled Thursday/Friday, ComRes Wednesday/Thursday. Last week we had on bombshells news story each day, the kind of story that could have reoriented the campaign if it was given time to stew. But, we didn't, and instead the voting opinions could very well differ drastically depending on which stories are captured. I suspect we may see more variation tomorrow between those polled early in the week, those polled later, and those polled all 5 days.

[Oryxslayer]

https://twitter.com/britainelects/status/1200869070410846209

Another weekend poll. Their one of the  outliers  with the Tories that high though.

https://twitter.com/britainelects/status/1200884503704678402

I don't think this would change the seat-by-seat model that much if it was the results of their next 100K polled respondents.

[Oryxslayer]

If you don't like Constituency Polls, look away now.

https://twitter.com/britainelects/status/1200857453904220161

The Tories lead by 13% when told only Gauke candidate a real shot of defeating them, and 12% when told the Lib-Dems. YouGov rates this seat as Likely Conservative, and this poll conforms to that rating, with the Tories always around 50%, but not by much.

https://twitter.com/britainelects/status/1200856948893196289

Beaconsfield had a reasonable non-tory base back in 2017. It appears Grieve has bit into the Conservative vote, but not by much. When told the only party with a realistic shot at winning against the Tories was Labour, the margin matched 2017's Tory landslide. When told it was only Grieve with a realistic shot at winning in a 1v1, he still loses, 54%-45%. YouGov rates the seat as Likely Conservative, and Grieve's large voteshare coupled with independent uncertainty justify that rating.

https://twitter.com/britainelects/status/1200856360721760257

Frankly, I'm not sure why they polled this seat. It's more or less safe Tory, no matter how many Lib-Dems want to return to the good old days. It voted for leave and gave the Tories a majority in 2017.  The Tories lead by 32% and 19% when told only Labour and the Lib-Dems have a realistic shot at winning the seat. YouGov rates the Seat as Safe Tory, and nothing should change that.

https://twitter.com/britainelects/status/1200855028229115905

This was a Lib-Dem target not that long ago, and it's one of the few constituencies we actually have two constituency polls of. Last poll the Lib-Dems were within striking distance of Labour, that's all gone now. Shows just how much constituency polls can vary because of their small voter pool, and it also shows what happens when the Lib-Dems pull their targeted resources and their voters scatter. When told there are  only two realistic parties: Labour and the Tories, Labour leads by 52% to 44%. When told it is only the Lib-Dems and the Tories, The Lib-Dems  only lead by 44% to 43%, since labour keeps a hold on 10% of the vote. YouGov rates the seat as Likely Labour, and I see no reason to change that because the  Lib-Dem base seems happy to cast Red ballots now that the  party has no chance.

Now for the big one:

https://twitter.com/britainelects/status/1200855757362782209

When told that only Labour and the Conservative's have a chance of winning, Raab wins in a landslide. When told that it's a Lib-Dem verses Conservative race with no other threats, it's a 48%-48% tie. Even though this is an uncertain constituency poll, this confirms Raad has royally screwed up. The Lib-Dems were right to target this seat and it looks like they are got a  real race on their hand. Raab's personal problems make him an ideal target for the Lib-Dems. I think the Lab-Con numbers here also are  important, since they show that Corbyn is worse  of a fit for the seat than Raab. YouGov rates this as likely Tory, but I get the feeling from everything going on here that this is going to be one of the seats YouGov and other models miss because of how the Lib-Dems targete their campaign resources.

[Oryxslayer]

Maybe the YouGov poll is actually the most interesting one out of that lot, given their tendency going back almost two years now as Tory-friendly - Labour equalling their highest rating since the launch of Change UK (remember them?) and the second lowest YouGov lead since Johnson became PM.

As for Opinium - lol. Literally nobody - including in Tory HQ - actually believes they are ahead 46-31.

You are probably mostly correct.
But is it possible the other pollsters are over herding to 2017 results?  Is it possible they might over compensentate for tactical voting?

Because a pollster notifies the polled person with the tactical choice, does not mean all the voters really get the notice.  It also does not mean all of the possible shifters will actually shift.

And finally there may be no reliable way to gage shifting Leavers, who may be shifting in greater numbers than in 2017.

Let’s wait till 12/13 to judge the pollsters.

Yes, the big worry is that pollsters have overcorrected since 2017 (as pollsters tend to do when their previous house effect was off) and BoJo is on track for an even bigger majority than it seems thanks to previous nonvoters. Hell, their previous overcorrectiong from when the pollsters missed 2015 was part of the reason why Corbyn's surge was totally off the radar. The issue with this line of thought though is that the MRP poll seems to suggest that overcorrection is going on to a slight degree, but it's main effect is the solidification of already expected Tory flips, rather than padding an already large margin. But yes, the best play as I always say, is to watch the tracking average models and see what 10 days bring.

[Oryxslayer]

I made my ten days out, long-spelled-out prediction in the other thread.

[Oryxslayer]

2. Corbyn hating was, if anything, more intense than it has been this year. Seriously, I was shocked at how vitriolic the rhetoric was; I sort of assumed it had gained volume and intensity over the past 2 years.

Even though I'm here now? 😁

Number 2 and number 5 seem to go hand in hand. The forum back then was more unruly, banning Krazen was arguably the turning point when things became more civil, even though he likely didn't post in that thread.

[Oryxslayer]

https://twitter.com/EuropeElects/status/1201512655166554114

Don't think we have ICM here yet. They are one of the higher-rollers  on Labour, which is fine. Like I always say, look to the modelling trackers. It's dangerous if we don't have outliers for the models because that leads to herding.

[Oryxslayer]

It seems like the WWIII-present UK system is Labour wins a landslide once every generation, Conservative pluralities/narrow majorities otherwise.

I never realized radioactive mutants were so pro-Conservative.

[Oryxslayer]

https://twitter.com/britainelects/status/1201796560574537728

https://twitter.com/britainelects/status/1201822002476134401

Morning Poll roundup. I'm interested in the London poll, because it seems to confirm that the city will go one way (ousting tories) whereas the country may go another. The pollster even says in their article that we have reached the point where the numbers don't really matter (the safe seats are confirmed as safe), it's more local factors like candidates and targeted issues in the swing seats.

[Oryxslayer]

There are murmurs that 'internal polling' has Jo Swinson under water in her own seat.

Beautiful! I'm adverse to tactical voting as a matter of principle (even if I lived in a Tory/LD marginal there's no way that I'd vote for any party other than Labour as long as they have someone as flawless as Corbyn as their leader) but I'd certainly vote SNP in East Dunbartonshire to give the Yellow Tories yet another well-deserved humiliation.

I mean it makes sense that internal polling has them down. The SNP's revival is thanks to their present ability to convince 2014/15 but not 2017 voters to turn out. This demographic is concentrated most strongly in the urban strip, not the tory highlands and borders. This is unfortunately why Labour's Scottish prospects are rather dire right now. It's also why Swinson may be polling a tight race.

HOWEVER. Tactical voting among unionists is a thing and is the reason why SNP undershot every projection in 2017. It's something that is both hard to poll and occurs spontaneously. it may not be enough to save those with small majorities, but it will likely save Swinson with her 10% lead.  

[Oryxslayer]

https://twitter.com/ElectionMapsUK/status/1201911803632111617

New Poll. I'm not sure  if we have had a poll recently where Labour went down? If so, this is a first. However, this seems like a dead cat bounce poll more than anything.

[Oryxslayer]

Is this a last minute Labour surge or not?

I'm not seeing a surge. The weekend polls showed an uptick, but it was marginal at best. YouGov should update their model tomorrow which will give us another data point. This upcoming weekends polls will likely be make-or-break if Labour's getting a detectable surge since we got Trump and another 1v1 debate this week, both serious opportunities for Labour to benefit.

[Oryxslayer]

Anyway, there was a debate between Scottish leaders tonight, and the reaction from those that did watch it was that Leonard really stepped in it and did Scottish Labour no favors.

[Oryxslayer]

I'm sure everyone has heard about the leaked video of Trudeau, Macron, Princess Anne, and the rest gossiping about Trump. Well, the fallout from that video may result in one of the best possible outcomes for Boris, so much so that one has to wonder if his team was involved. Trump 's energy and fury is focused on Trudeau and not meddling in the GE, he leaves early so there is less worries about him making statements of friendship, and Boris is in the video as well, which helps counteract the narrative of him being a puppet of Donald.

[Oryxslayer]

What's the chance of a polling overcorrection from 2017?

¯\_(ツ)_/¯

If pollsters know what their are doing, they corrected for 2017. However, there will still likely be errors, it's just that those errors are  unlikely to be the same ones as previously. We won't know for 8 days whether that potential error was because of overcorrection.

[Oryxslayer]

YouGov's got an interesting chart out. While the numbers look bleak for Labour, remember that voters cast ballots on more issues than Brexit.

[Oryxslayer]

What's the chance of a polling overcorrection from 2017?

Quite possible, certainly I think part of the reason for the miss in 2017 was overcorrection from 2015.  Survation and Yougov were accurate so doubt you are seeing one there, but I noticed ComRes and ICM had most Pro-Tory numbers in 2017, while are now showing the narrowest gap so for those giddy about the closing gap, those ones may be overcorrecting.  That being said I think turnout amongst younger voters which is anyone's guess will be key.  If they show up in big numbers it will be closer than polls suggest, if they have usual turhout, Tories will probably win quite comfortably.

I would also focus on seat and regional polls as this election a uniform swing will do terrible.  A uniform swing would show Labour holding a lot of their traditional Northern seats they probably wouldn't while in South show them well back in ones they are likely to hold or at least be competitive in.  A uniform swing probably wouldn't show them in danger of losing Sedgefield which they are while in the same time suggest Canterbury is lost even though I think Labour has a decent chance at holding it.

Hey, we think alike! Since we do, I'll throw on another data point that I agree with. Today I was reading something from Peter Kellner and he made a good point about the YouGov model. His basic point was that the YouGov model broadly shows whats going on in seats 'like this one' not 'exactly this one.' He brings up local cases like Barnet, since YouGov's MRP poll doesn't weight for Jews (too small of a demo, explained why they got Barnet wrong in 2017 as well), IDS and Raab's prominence in the Tory party, and Caroline Flint's pro-Brexit views in Don Valley. Essentially, if a seat has X demos and filed candidates, it should have X percentages, which is good for almost every situation. In some situations though, individual issues matter that are beyond a MRP polls capability to cover. This is why we should believe that Canterbury is more likely to stay Red than Bassetlaw, despite Canturbury having a smaller labour Margin in 2017.