YouGov

[mattbewilson]

Jun 20, 2024 20:50:52 GMT doktorb🏳️‍🌈🏳️‍⚧️ said:

Jun 19, 2024 16:55:07 GMT michaelarden said:

Exactly - there is no way Galloway is going to get 7% in Rochdale and come 5th for example.

There's something broken about how many players don't include any of the prominent 'outsiders', be it Galloway, Corbyn, even Farage. There will have to be lots of reviews into the way these things work after polling day.

randomly tho in 2017/19 YouGov did register the support for the independent in Devon

[redtony]

The Socialist Campaign Group will provide the opposition from within the PLP

[carolus]

Jun 20, 2024 17:04:37 GMT mattbewilson said:

What is up with people trying to make out pollsters are bubkus atm

You just need to look at them!

The latest polls from various pollsters, all taken within the last week, have the following ranges:

Lab 35-46

Con 18-25

Reform 13-24

LD 9-14

Green 4 -8

These aren't just statistical noise or even a single outlier, these are serious, substantial disagreements between 10+ pollsters. They are simply inconsistent, which means that some, or even most, must be wrong.

At this point it's very clear that both media narrative and electoral campaigns are being shaped by polls, many of which cannot be correct. At the same time, pollsters are incentivised to be the first to show some new trend, or to be the one voice predicting some shock result, or to happily show the result most useful for your client. Combine that with a regulatory regime that it would be generous to describe as "light-touch", and you have a recipe for disaster.

Of course this is made even worse by the pseudo-scientific MRP fad.

[johng]

Jun 20, 2024 21:13:07 GMT carolus said:

These aren't just statistical noise or even a single outlier, these are serious, substantial disagreements between 10+ pollsters. They are simply inconsistent, which means that some, or even most, must be wrong.

At this point it's very clear that both media narrative and electoral campaigns are being shaped by polls, many of which cannot be correct. At the same time, pollsters are incentivised to be the first to show some new trend, or to be the one voice predicting some shock result, or to happily show the result most useful for your client. Combine that with a regulatory regime that it would be generous to describe as "light-touch", and you have a recipe for disaster.

Of course this is made even worse by the pseudo-scientific MRP fad.

I have to agree, the divergence is just too much. We've had a Labour lead of 11 and a Labour lead of 27 in the last few days. Reform have polled as high as 24 and low as 13.

Of course, we can say People Polling is nonsense (Britain's Trafalgar) and we know all pollsters have outliers, but this is beyond anything that could be considered normal.

And this MRP thing... I am now genuinely annoyed when I see people talk about Labour being ahead somewhere and Reform somewhere else. It's an enhanced universal swing calculator. Nothing more, nothing less. They are built upon relatively complex and mainly untested models, and those models may indeed be wrong. Not the gospel that some suggest.

[batman]

I had to tell a Trotskyist friend on Facebook today (he's actually a really good bloke) that there has been no poll showing Lee Anderson ahead in Ashfield, it was just a mixture of guesswork & extrapolation from a national poll. I think he gets it now. He lives in a neighbouring constituency (to Ashfield).

[hullenedge]

[cogload]

A European style **** show.

[bigfatron]

Jun 27, 2024 11:23:43 GMT hullenedge said:

Electoral Calculus (without tactical voting) for the LOLs...

Labour - 441   (majority 232)
Lib Dem - 82
Tory - 56
Reform - 22
SNP - 22
PC - 4
Green - 4

[graham]

Jun 27, 2024 11:23:43 GMT hullenedge said:

The party change in vote share figures shown there are inaccurate. A week ago Yougov came up with -

From the YouGov website for 18th Jun: (change vs 13th Jun)
Con 20% +2
Lab 36% -1
Lib 14% =
SNP 3% = (27% in sub-sample -7 so probably wat too small)
PC 1% =
Ref 18% -1
Grn 7% =
Oth 2% =

The latest figures should be -

Labour NC

COn   -2

Ref    -1

LD    +1

Grn  +1

SNP  NC

[The Bishop]

They had another, though not as publicised, poll in between.

[Strontium Dog]

Jun 27, 2024 11:36:57 GMT bigfatron said:

Jun 27, 2024 11:23:43 GMT hullenedge said:

Electoral Calculus (without tactical voting) for the LOLs...

Labour - 441   (majority 232)
Lib Dem - 82
Tory - 56
Reform - 22
SNP - 22
PC - 4
Green - 4

If the result is anything like that, I will eat Paddy Ashdown's hat.

[jamesdoyle]

3 points between Con and LibDem? Oh, yeah, baby.

[batman]

No other pollster has the Labour vote as low as this. That doesn't necessarily mean it's untrue, they could of course be right & everyone else wrong - but it is striking.

[Tony Otim]

Jun 27, 2024 15:37:01 GMT batman said:

No other pollster has the Labour vote as low as this. That doesn't necessarily mean it's untrue, they could of course be right & everyone else wrong - but it is striking.

Their new methodology does seem to push the Labour vote down, whilst their MRP continues to predict a large majority. Are they planning an update to the MRP before polling day?

[graham]

Jun 27, 2024 14:40:08 GMT The Bishop said:

They had another, though not as publicised, poll in between.

No sign of it on here!

[bridgyboy]

Jun 27, 2024 15:49:15 GMT Tony Otim said:

Jun 27, 2024 15:37:01 GMT batman said:

No other pollster has the Labour vote as low as this. That doesn't necessarily mean it's untrue, they could of course be right & everyone else wrong - but it is striking.

Their new methodology does seem to push the Labour vote down, whilst their MRP continues to predict a large majority. Are they planning an update to the MRP before polling day?

Do they provide figures using their old methodology? I understood this was a form of prediction rather than a snap poll but surely with a week out you must take a view that "don't knows may not actually vote rather than allocate somewhere else?

[Adrian]

Jun 27, 2024 15:37:01 GMT batman said:

No other pollster has the Labour vote as low as this. That doesn't necessarily mean it's untrue, they could of course be right & everyone else wrong - but it is striking.

I assume, until informed otherwise, that this is the same problem (with using a panel) as at previous elections - whatever bias is in the panel has not been effectively accounted for and will remain throughout the election polling. If I'm right, then the other pollsters have the Labour figure more accurate (40%+), which is not exactly good news for the Tories.

[johng]

I've done a fair bit of canvassing and heard more from others. Plus have seen the other polls obviously. Do we really think the two main parties are going to get 54% of the vote? Doesn't seem realistic to me.

Jun 27, 2024 15:49:15 GMT Tony Otim said:

Jun 27, 2024 15:37:01 GMT batman said:

No other pollster has the Labour vote as low as this. That doesn't necessarily mean it's untrue, they could of course be right & everyone else wrong - but it is striking.

Their new methodology does seem to push the Labour vote down, whilst their MRP continues to predict a large majority. Are they planning an update to the MRP before polling day?

Their methodology change really hurt Labour. That's not to say they are wrong.

Polling this year has been rather mixed.

In the last week alone, are Labour 36 or 42? Are Tories 18 or 25? These are relatively big gaps and would lead to two quite different outcomes.

[graham]

Latest YouGov Westminster voting intention (25-27 June 2024)

Con: 20% (+2 from 24-25 June)
Lab: 37% (+1)
Reform UK: 17% (=)
Lib Dem: 13% (-2)
Green: 7% (-1)
SNP: 3% (=)

[manchesterman]

If this were to be the percentages on the night, this would be the lowest combined Con-Lab vote share (57%) since...  ?