Redfield & Wilton

[clyde1998]

May 30, 2022 18:30:52 GMT aargauer said:

May 30, 2022 18:27:52 GMT mattbewilson said:

I'm old enough to remember the daily yougov polls in 2015

Sometimes you just get a feeling the polls are wrong on the day. I had that in 2015. Also in 2019 when they weren't mind (although that was because I was in an exceptional seat).

I felt that in 2016. There was just a sense things were different to what the polls were suggesting; it might be because I'm in an area that was close to the national average or just my gut feeling as to how both campaigns were going.

[andrew111]

May 30, 2022 19:32:33 GMT mattbewilson said:

May 30, 2022 19:25:14 GMT batman said:

I don't think that the Liberal Democrats have lost about a third of their support in one week.

tbf a third of their support sounds larger than a 4% drop

That is maths.. Two quite different things, not generally appreciated by people commenting on polls. At the end of the day it is a certain number of people in successive samples "switching" from one party to another, so talking about % of sample is generally more sensible.

[mattbewilson]

May 31, 2022 14:15:15 GMT andrew111 said:

May 30, 2022 19:32:33 GMT mattbewilson said:

tbf a third of their support sounds larger than a 4% drop

That is maths.. Two quite different things, not generally appreciated by people commenting on polls. At the end of the day it is a certain number of people in successive samples "switching" from one party to another, so talking about % of sample is generally more sensible.

I did realise that lol

I wasn't suggesting a third was larger than 4%. I know it's sometimes difficult to read how something is meant rather than hearing it said out loud

[timmullen1]

[andrew111]

May 30, 2022 21:51:17 GMT aargauer said:

May 30, 2022 19:25:14 GMT batman said:

I don't think that the Liberal Democrats have lost about a third of their support in one week.

14 and 10 Just noise around 12.

Correct!

[timmullen1]

[aargauer]

Can someone remind me how far Ed Miliband was up at this point in 2012?

[The Bishop]

When there was a significant party to the right of the Tories costing them poll support.

And when the pollsters underlying methodology was *much* more favourable to Labour than it is now.

You are better than this disingenuous cope.

[aargauer]

Jun 6, 2022 17:00:44 GMT The Bishop said:

When there was a significant party to the right of the Tories costing them poll support.

And when the pollsters underlying methodology was *much* more favourable to Labour than it is now.

You are better than this disingenuous cope.

But that party of the right continued to poll well until the election. And now we don't have one with the same momentum, brand or sense of purpose. So I think that's very much a legitimate real headwind for labour rather than a data artefact.

How much has polling methodologies reduced Labours VI?

[The Bishop]

Can't quantify exactly, but significantly so in some cases. The 2010-15 polling has been "retrospectively" audited using current methodology and it apparently showed Labour would have been clearly ahead for only relatively short periods (rather than unbroken from spring 2012 to late 2014 as was actually the case) and Tories would have clearly had the edge going into the 2015 GE - which might have made it a rather different affair, of course (that's a hypothetical for you)

This is maybe especially acute with pollsters (like Opinium) who now do what are at least as much quasi-predictions as old style polling. Of course, it was this variety of surveys that came especially embarrassingly unstuck in 2017 and there must be at least a chance that will happen again.

[mattbewilson]

It is worth saying that UKIP took votes from Labour as well as the Tories.

Much like the Greens are now taking votes from the Tories as well as labour.

[The Bishop]

And apart from VI, nearly all polling indicators are in the toilet for the Tories and have been for a while now.

[batman]

Yes this is a bit of a surprise. I suspect it will turn out to be a minor outlier.

[timmullen1]

You’re going to love this (not):

[london(ex)tory]

Jun 7, 2022 11:47:00 GMT timmullen1 said:

You’re going to love this (not):

If they can come up with a reasonable selection of key constituencies and then poll the same set consistently over time then I could probably forgive them for using the “red wall” name.

[Tony Otim]

Jun 7, 2022 11:47:00 GMT timmullen1 said:

You’re going to love this (not):

So much to dislike in just one tweet...

[timmullen1]

Jun 7, 2022 15:55:10 GMT Tony Otim said:

Jun 7, 2022 11:47:00 GMT timmullen1 said:

You’re going to love this (not):

So much to dislike in just one tweet...

I told you you wouldn’t like it 😂

[woollyliberal]

More polling? What's not to like? Personally, I don't mind if they use The Red Wall as the band of usually Labour seats loosely between Rhyl, Leeds and Sheffield, or if they use it to mean any seats in England and Wales that have been Labour in recent years. More polling is good and polling of a currently interesting type of seat is good.

[aargauer]

Jun 7, 2022 16:48:20 GMT woollyliberal said:

More polling? What's not to like? Personally, I don't mind if they use The Red Wall as the band of usually Labour seats loosely between Rhyl, Leeds and Sheffield, or if they use it to mean any seats in England and Wales that have been Labour in recent years. More polling is good and polling of a currently interesting type of seat is good.

Theres some seats we gained in the north or north midlands that don't demographically fit with the general pattern. Stockton South, Gedling for instance.

[European Lefty]

Well I think we can safely remove R&W from our list of credible pollsters