Political polling

[killer b]

That's how the Lib Dems used to fund their by election campaigns. Spot a by election where they didn't look like they had a chance. Drop a ton of cash on them winning it at the bookies. Watch themselves become the "bookies' favourite". Put that 'fact' repeatedly on election leaflets throughout the campaign. Win the campaign as a result. Bank a load of cash to help pay off the campaign. I understand others have applied the same approach since too, with some degrees of success.

Name one campaign this happened in.

 

[mauvais]

I heard the Lib Dems used to get their money by advertising dildos, massive ones, latest and greatest, six different functions, right, cheques payable to something discreet like Paddy's Parts or whatever, but then when sent any money they would return it saying 'out of stock' and that cheque would be written in the name, 'The Liberal Democrats'

 

[killer b]

Comres is my new best friend.

 

[prunus]

Comres is my new best friend.

For the Telegraph. This to ensure that the Tory vote gets out.

 

[killer b]

oh god shut up.

 

[Plumdaff]

I find it hard to believe that there are any Telegraph readers who haven't already postal voted, because I find it hard to believe that there are any readers under 70.

 

[brogdale]

Everyone knows that the PV agents don't just watch the envelope opening...and there's a role for those with photographic memories & counting systems...but everyone ought to know that blabbing about the data they've gathered in the week before polling day is totally wrong & breaches electoral law.

 

[Hollis]

Email from Labour suggesting I go out tomorrow in Finchley and Golders Green:

"Heading out to a key seat will make the biggest difference tomorrow."

That MRP poll shows them in 3rd on 23:

Finchley and Golders Green 49 23 28

 

[ska invita]

Email from Labour suggesting I go out tomorrow in Finchley and Golders Green:

"Heading out to a key seat will make the biggest difference tomorrow."

That MRP poll shows them in 3rd on 23:

Finchley and Golders Green 49 23 28

a friend has been canvassing out there - she's got burned hard.

 

[killer b]

Everyone knows that the PV agents don't just watch the envelope opening...and there's a role for those with photographic memories & counting systems...but everyone ought to know that blabbing about the data they've gathered in the week before polling day is totally wrong & breaches electoral law.

this article from the other day is good (I've only just read it and it contradicts some of my earlier definitive posts on the topic ): Go postal | Lewis Baston | The Critic Magazine

 

[ignatious]

Email from Labour suggesting I go out tomorrow in Finchley and Golders Green:

"Heading out to a key seat will make the biggest difference tomorrow."

That MRP poll shows them in 3rd on 23:

Finchley and Golders Green 49 23 28

I very much doubt they see Finchley and Golders Green as a ‘key seat’. I think they probably mean ‘don’t waste your time canvassing locally because it’s a lost cause, go and get involved somewhere else’.

 

[brogdale]

this article from the other day is good (I've only just read it and it contradicts some of my earlier definitive posts on the topic ): Go postal | Lewis Baston | The Critic Magazine

It happens...if local parties have people smart enough to develop their own counting/recording systems.

 

[killer b]

You dont actually need to see that many voting slips to get an decent sample tbh. I did some counting over the shoulders of the tellers at a count for a council election last year and predicted a win in two seats we weren't expecting to win, correctly.

 

[brogdale]

You dont actually need to see that many voting slips to get an decent sample tbh. I did some counting over the shoulders of the tellers at a count for a council election last year and predicted a win in two seats we weren't expecting to win, correctly.

PV verification?

 

[killer b]

No just the normal count. They were in the process of changing the boundaries and it was useful to know where the vote was within each ward. I added them all up at the end and predicted the two wins within a few % an hour before they announced

 

[brogdale]

No just the normal count. They were in the process of changing the boundaries and it was useful to know where the vote was within each ward. I added them all up at the end and predicted the two wins within a few % an hour before they announced

Yeah; done plenty of count sampling in my time. As with any sampling, the bigger the better tbh.
The PV verification ‘sampling’ is a whole different level though..they have to do it covertly with no means of immediately recording what they’re counting. To be done effectively enough to allow agents to extrapolate predicted outcomes requires a great deal of concentration and ability to memorise. Not many activists can do it well tbh.

E2a: also depends on EROs; some are more realistic about what’s going on than others.

 

[cupid_stunt]

Round-up of all of the polls that were published yesterday...

 

[William of Walworth]

That 6% Tory lead one from Comres looks like a complete outlier, sadly

 

[Supine]

And so it begins. The BIG poll opens at 7am

 

[kabbes]

Round-up of all of the polls that were published yesterday...

View attachment 192693

Strong consensus around 10%. Now let’s see what reality is...

 

[bimble]

How does a percentage lead translate into seats or does it just not work that way ? Found this but no idea if it's a nonsense. Hope it is.

 

[cupid_stunt]

How does a percentage lead translate into seats or does it just not work that way ? Found this but no idea if it's a nonsense. Hope it is.

Not a great track record TBH.

2017 -

The headline prediction for the June 2017 election was not accurate. The final prediction was for a Conservative majority government with a majority of 66. The actual result was that the Conservatives were short eight seats of a majority.
Track Record: 2017 Errors

2015 -

The headline prediction for the May 2015 election was not accurate. The final prediction was for a hung parliament with Labour/SNP as the largest bloc. The actual result was a small Conservative majority.
Track Record: 2015 Errors

 

[bimble]

good! Massive errors in seats not so much in percentages so presumably it's hard to predict seats from uk wide polls.

 

[cupid_stunt]

good! Massive errors in seats not so much in percentages so presumably it's hard to predict seats from uk wide polls.

The YouGov MRP poll/survey was fairly accurate in 2017, IIRC the first time MRP was used, calling a hung Parliament, despite their last 'normal' poll having the Tories on a 7% lead, compared to the 2.5% achieved.

Their MRP released on Tuesday gave the Tories a 28 seat majority and a 9% lead, the model is based on a massive sample, over 100k compared to the usual 1-2k, and mapped to each seat.

 

[Supine]

Interesting read now the polls have stopped

https://www.craigmurray.org.uk/archives/2019/12/the-largest-vote-swings-in-british-general-election-history-censored-out-by-the-bbc-and-mainstream-media

 

[rekil]

How many times must it be pointed out that Craig Murray is a loon, part of the crankosphere that needs to be persecuted and destroyed with extreme predge.

 

[Poi E]

How many times must it be pointed out that Craig Murray is a loon, part of the crankosphere that needs to be persecuted and destroyed with extreme predge.

How does that distinguish him from the bulk of media in the UK?

 

[Proper Tidy]

How does that distinguish him from the bulk of media in the UK?

The bulk of the media in the UK is a product of its environment and the class and culture from which they come. Whereas Craig Murray is a crank conspiracist assadist dickhead

 

[danny la rouge]

Fuck Craig Murray. Seriously fuck him and David Icke and Vanessa Beeley.

 

[chilango]

Fuck Craig Murray. Seriously fuck him and David Icke and Vanessa Beeley.

This. ^