Westminstenders: The Undecideds

[RedToothBrush]

Apparently the pollsters are nervous.

There are far more undecided than there have been at this stage in recent general election.

This makes it a hugely fine line between a hung parliament and a massive tory majority.

The weather on Thursday isn't expected to be nice and this could affect turnout. The blue corner are particularly nervous about this, but don't forget those postal votes.

Whatever happens on Thursday at least this election campaign is nearly over. And that can't be a bad thing.

And Christmas is in a fortnight so we can all drown our sorrows or celebrate in style.

You always have time to restock the drinks if you end up resort to them on Thursday

(New election special thread on Thursday rather than bunging up Westminstenders BTW)

326 is what is needed for a majority.

339 - 326 = 13. You then double that for your official majority number.

But yes. NI. And don't forget the DUP factor.

Following reaction to the yougov poll on Twitter, and some posters are saying it doesn't take into account younger voters or, of course, tactical voting.
yougov.co.uk/topics/politics/articles-reports/2019/11/27/how-yougovs-2019-general-election-model-works

Be optimistic if you wish
but I no longer am

The poll takes into account those who have decided to tactically vote and stated that to YouGov. They then projected that information based on demographics. So if you have a bunch of LDs who have decided to tactically vote since the last poll then they would show up on the model.

What it doesn't show is people who are still undecided and agonising over whether they should tactically vote or not. Those are the magic voters that pollsters try and guess but have less reliable data for.

So it's not true that YouGov aren't taking into account tactical voting. They are taking into account declared tactical voters and guesstimating how many others they expect will do the same and then projecting it.

"But even these will not be right everywhere: 95% confidence with 632 seats means that we still expect the interval to miss the results in about 30 constituencies."

95% CI means these results could be way off in key marginals.

Ain't over til the fat lady / fella sings

I wouldn't say I'm optimistic, but it's encouraging to see the predicted Tory majority down from 68... I was expecting it might have increased if anything

Would the DUP join up with the Tories again????

I thought they wanted nothing more to do with them.

I must have missed something.

Just adding this in while we all hate the bbc - news from 2016 that’s aged particularly badly:

www.theguardian.com/media/2016/mar/13/government-choose-bbc-board-john-whittingdale

Glad I am not working for the pollsters... I get the feeling that they will all be out one way or another and there will be redundancies within those companies

Looking at it, they expect around 30 seats to be wrong?

So could be a big majority, could be a hung parliament. I'm not sure it really tells us anything new.

I'm finding the methodology discussions really interesting. How many people who are undecided the day before will actually go out and vote?

Be optimistic if you wish
but I no longer am

I'm not hopeful either.

But a cause for comfort is that 54% of voters do not want this Tory government (and don't vote for the Brexit party). That's 54% of fellow citizens who do see through the lies and deceit.

Could anybody kindly post a click link so I can see my constituency?

yougov.co.uk/uk-general-election-2019/

For reference here are the %s

VOTING INTENTION ESTIMATES
SEAT ESTIMATES

Conservative
43%
339
Labour
34%
231
Liberal Democrats
12%
15
Brexit Party
3%
0
Green 3%. 1

SNP
3%
41

Plaid Cymru
0%
4

Other
2%
1

Northern Ireland

18

Quite an exciting 20 minutes just now, saw on Twitter that Boris Johnson was literally 2 minutes down the road from me. Popped down there to have a look (nosey) NOT to heckle, promise.

As I arrived I was stopped by a police motorbike and the 'Get Brexit Done' bus, about 4 range rovers and about 30 police whizzed past!

They don't get my vote, but it was quite exciting all the same!

How many people who are undecided the day before will actually go out and vote?

Again, a big Big question for the pollsters.

@OhYouBadBadKitten we need a weather report

DH has just told me he's seen a forecast of snow on hills for Thursday...

Thanks @thecatfromjapan it looks like I might have to go onto the laptop but I'll do that now!

And this, is why I'm exhausted every day at the moment!

How many people who are undecided the day before will actually go out and vote?

In our marginal the updated model looks much more promising. There is 3% in it compared to about 15% previously. An estimated 10% for Green/Lib Dem so maybe room for a bit more of a squeeze. The Labour candidate has a big drive to get the vote out on the day. Fingers crossed...

"But even these will not be right everywhere: 95% confidence with 632 seats means that we still expect the interval to miss the results in about 30 constituencies."

It’s a bit more complex than that.... The polls will have a margin of error (3% for polls of 1,000). I’m
not sure what it would be for an MRP poll, but would guess between 3 and 5% at constituency level. The 95% relates to the confidence that the poll is within 3-5% of the actual vote. So even if 5% out, the poll is deemed consistent with expected levels of accuracy.

If people want Brexit and decide to vote Conservative to get Brexit done, then I am going to have little sympathy for them when the magical transformation they hope to achieve comes to naught and the Tory loving media ignores them and goes back to their bubble again.

People want Brexit? Then they get to own it. I really hope it works out for them - I would love to be wrong about it, but if people really want it, then they can fucking have it and all the shit and aggravation that comes with it. People can't say we haven't tried to warn them about the perils.

There's a fair number of Welsh seats which are toss ups.

That's interesting as the school of thought for a Conservative breakthrough is they have to make gains in Wales and the NW.

Oakenbeach that is quoted from yougov's own MRP methodology. Yes, 5% is within the margin of error. The issue with polling an election is that the result is often dependent on a small number of target marginals and as such could have a huge impact on the outcome.
TL:DR the poll tells us little, but at least it doesn't show a bigger Tory majority than last time!

Eeek we've switched! From a lean Conservative to a lean Labour! Perhaps the tide is turning ever so slightly!

Squeaky bum time for Raaaaaab.

Good.