CatholicLeft
Labour
2032 posts until I was "accidentally" deleted.
Posts: 6,322
|
Post by CatholicLeft on Dec 10, 2019 22:27:39 GMT
Na h-Eileanan an Iar has 22% for Conservatives. I think not.
|
|
jamie
Top Poster
Posts: 6,892
|
Post by jamie on Dec 10, 2019 22:27:51 GMT
Bishop Auckland looks too good for Labour. Labour only 5% behind I.e. leans 4% towards Labour compared to the nation (same as 2017). Nearby seats showing much more swing and Bishop Auckland will have one of the most Brexity 2017 Labour votes in the country.
|
|
|
YouGov
Dec 10, 2019 22:37:14 GMT
Post by bjornhattan on Dec 10, 2019 22:37:14 GMT
The Brexit Party have some good news in this model, with their support estimated at 30% in Barnsley Central. According to the 95% confidence interval, it's now plausible that they could win a seat somewhere.
|
|
|
YouGov
Dec 10, 2019 22:38:30 GMT
Post by BossMan on Dec 10, 2019 22:38:30 GMT
Is there a spreadsheet somewhere with the seat estimates, as there was the other week?
|
|
|
YouGov
Dec 10, 2019 22:42:09 GMT
Post by BossMan on Dec 10, 2019 22:42:09 GMT
The Tories are actually doing much better in Wakefield than they were 2 weeks ago, now 8 points ahead - more than the 5 point lead in Shipley. Bradford South is now a tossup.
|
|
|
YouGov
Dec 10, 2019 22:47:40 GMT
via mobile
Post by Devil Wincarnate on Dec 10, 2019 22:47:40 GMT
Na h-Eileanan an Iar has 22% for Conservatives. I think not. Now you have grown, now you have shown, oh, Na h-Eileanan...
|
|
jamie
Top Poster
Posts: 6,892
|
Post by jamie on Dec 10, 2019 22:49:27 GMT
The Brexit Party have some good news in this model, with their support estimated at 30% in Barnsley Central. According to the 95% confidence interval, it's now plausible that they could win a seat somewhere. The confidence interval for Barnsley Central is fucking massive. Somewhere between a 6% Brexit lead and a 33% Labour lead. I’d take the outer edges of these intervals with a pinch of salt.
|
|
|
Post by bjornhattan on Dec 10, 2019 22:50:37 GMT
The Brexit Party have some good news in this model, with their support estimated at 30% in Barnsley Central. According to the 95% confidence interval, it's now plausible that they could win a seat somewhere. The confidence interval for Barnsley Central is fucking massive. Somewhere between a 6% Brexit lead and a 33% Labour lead. I’d take the outer edges of these intervals with a pinch of salt. It'll be a case of there being few seats with the Brexit Party in contention, so the model will be less reliable in areas they do well.
|
|
jamie
Top Poster
Posts: 6,892
|
Post by jamie on Dec 10, 2019 22:50:40 GMT
|
|
Tom
Unionist
Posts: 1,998
|
Post by Tom on Dec 10, 2019 22:52:27 GMT
Obviously the race has narrowed, but lots of these individual seat projections, particularly in Scotland, are bullshit.
|
|
|
Post by Devil Wincarnate on Dec 10, 2019 22:52:59 GMT
I'm going to repeat my suggestion that Leigh is not anywhere near like a toss-up and will be an easy Labour hold.
|
|
|
Post by evergreenadam on Dec 10, 2019 22:54:25 GMT
An interesting figure: assuming the YouGov figures are true (in particular - the 95% confidence interval), the probability of the Conservatives winning a majority is about 84%. Of course, that also assumes no change in the race between now and Thursday night. And some of the projected majorities are wafer thin. Margin of error would be enough to make a significant difference to whether there is a Tory majority.
|
|
|
Post by edgbaston on Dec 10, 2019 22:59:05 GMT
So do we believe this?
I think the overall seat prediction is very likely to be ballpark correct.
But some of these constituency vote share predictions... yeesh what garbage. And that undermines my faith in the whole exercise.
|
|
|
Post by Devil Wincarnate on Dec 10, 2019 23:05:17 GMT
So do we believe this? I think the overall seat prediction is very likely to be ballpark correct. But some of these constituency vote share predictions... yeesh what garbage. And that undermines my faith in the whole exercise. Let's face it, we effectively use this site to crowd-source a lot of local knowledge and build a picture. Between a few hundred of us, we probably have a better idea than many pollsters and all the bookies, because we've been tracking things for years between us, sharing views and arguing the toss. Local elections, demographics, local personalities... These models don't take of this stuff into account. As ever, things are treated as pure science when there is a heavy dose of art.
|
|
|
YouGov
Dec 10, 2019 23:08:33 GMT
via mobile
Post by thirdchill on Dec 10, 2019 23:08:33 GMT
So do we believe this? I think the overall seat prediction is very likely to be ballpark correct. But some of these constituency vote share predictions... yeesh what garbage. And that undermines my faith in the whole exercise. That is pretty much what I am thinking. Same majority, different seats. Norfolk North stood out to me as one of the predictions I do not see coming to fruition, and even if the conservatives did somehow gain it, it would be nowhere near that margin.
|
|
|
Post by thirdchill on Dec 10, 2019 23:11:45 GMT
Na h-Eileanan an Iar has 22% for Conservatives. I think not. There is quite a big problem in the model for the weakest seats for conservatives and labour respectively, in most cases the percentage is likely to be severely overstated, and the majorities in the safest seats are likely to be bigger than this model suggests, particularly in labour's case.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 10, 2019 23:19:05 GMT
Interesting that the last one had Caithness, Sutherland and Easter Ross as an SNP gain and this one has the Lib Dems miles ahead. Shows that they don’t really have a clue about these sort of seats.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 10, 2019 23:19:50 GMT
So do we believe this? I think the overall seat prediction is very likely to be ballpark correct. But some of these constituency vote share predictions... yeesh what garbage. And that undermines my faith in the whole exercise. According to Robert Cuffe, BBC head of statistics - The 2017 version of the [YouGov MRP] model…called 93% of seats correctly. [But] YouGov's correct-call-rate was lower (80%) for marginal seats.With that proviso, I suspect that YouGov is in the right area in terms of the situation at this moment. However, in very close contests (of which there will be more on this occasion) Lib Dems and tactical voters may cause opposition candidates to outperform where it matters, which would have an impact on the apparent accuracy of YouGov's projection. Also, of course, an election campaign is a dynamic process, and this one seems to be showing an anti-Conservative trend in its last days which may not yet have worked itself out. The latest Ashcroft poll provides further evidence for such a trend. Of course, the publication of the YouGov MRP and recent polling evidence may create a feedback effect which will have the effect of pushing reluctant or apathetic Conservative-inclined voters to the polls. Also, many people have already long ago voted by post. However, I wouldn't feel very comfortable tonight if I was Boris Johnson.
|
|
|
YouGov
Dec 10, 2019 23:25:47 GMT
Post by lancastrian on Dec 10, 2019 23:25:47 GMT
Labour have very quietly built a lot of momentum during this campaign. I expect to see at least one poll showing a lead less than six points (ie smaller than any poll up to now) tomorrow.
|
|
|
YouGov
Dec 10, 2019 23:35:53 GMT
Post by Andrew_S on Dec 10, 2019 23:35:53 GMT
It appears there has been a swing to the Tories amongst certain demographics. For example the model now has Labour ahead by just 1% in Lincoln compared to 3% previously.
Also, Gedling has a 1% Labour lead now, compared to 3% before.
|
|