Categories
politicalbetting.com » Blog Archive » Your regular reminder that hypothetical polls can be as accura
politicalbetting.com » Blog Archive » Your regular reminder that hypothetical polls can be as accurate as an American war film
Westminster voting intention…
0
This discussion has been closed.
Comments
Oh and good afternoon PB
This hypothetical polling prediction is of course as equally valuable as the example in the header.
And what were the VI figures *without* the hypothetical at the time of the hypothetical poll, for comparison?
What a brilliant analogy, articles like this is why I read PB.
Post-race ramble:
https://enormo-haddock.blogspot.com/2019/12/abu-dhabi-post-race-analysis-2019.html
Good and Bad news for the Conservatives.
The bad news is that consumer confidence for November came in at -14, the worst in any election since 1983 apart from 2010 which was at -16. In 2017 it was -10, and Major in 1992 had it at -8.
Consumer confidence being above -10 has correctly predicted government majorities in the past 8 out of 9 elections, 1997 was the only time it got it wrong.
The good news is that their predicted majority with this week's opinion polls is stable, no change with my swingometer at a majority of 24-88 same as past week, and at 347 (up 2) with Electoral Calculus using the average regional subsamples of all pollsters.
This week is the first week that I have the SNP bellow 40% in the average scottish subsamples.
It's CON 25 LAB 19 LD 15 SNP 38 GRN 1 BRX 2 in scotland.
I do think extending without a deal would have been a different story.
But it is close, another point decline will push my swingometer to the 2-64 majority range, and you do have more and more pollsters predicting a Hung Parliament and the really bad consumer confidence figures.
Also there is an error in the Electoral Calculus model that it doesn't account very well for the Brexit Party not standing in Conservative seats, it subtracts from them in Labour seats but does not add them in Conservative seats, it would overestimate conservative losses if the polls swing that far.
I noticed it yesterday playing with it.
Had a LibDem canvasser today, I told him I'd voted LibDem in 2010 and 2015 but wouldn't be doing so this time. As I refused to say who I would vote for he no doubt correctly assumed I will be voting Tory. He didn't ask why they had lost my vote though, which was a shame, but presumably I'm a lost cause now.
On the menu is said 'Haway Pizza' - must have been a special for Geordies.
Did anyone use it in 2017 at the time to predict a HP before everyone else?
Sign of a weak canvass not getting as much useful info as possible...
#ukchampionshipsnooker
It's really now or never for Lib Dem voters (that extra 5%), to decide whether they prefer Corbyn or Brexit
Laura Pidcock would know this. Islington Labour not so much.
Good targeting
Average Tory lead (last 6 polls) 11 days ahead of GE19 = 9.2%
Average Tory lead 11 days ahead of GE17 = 8.8%
That hypothetical poll was not useless. For any hypothetical poll you need to think not just what the question says but what the respondent thinks and what they think they're answering. Many overly simplistic individuals read these polls and thought that this meant that Boris would be destroyed if only Remainers could stall Brexit until after Hallowe'en.
What the polls really showed, which many of us here identified, was that the respondents were fed up of the messing around Parliament had done for months and wanted to "Get Brexit Done". Boris identified that and fed into that narrative.
The hypothetical poll showed a surge for the BXP as people viewed the hypothetical as meaning that the Tories and others were still messing around. By saying he would rather "die in a ditch" and by obsessing over the date and refusing to extend Boris successfully tapped into that feeling himself. Voters aren't idiots, leavers aren't idiots. We could all see the Benn Act, we all knew who was responsible for delaying Brexit again and it wasn't Johnson it was Grieve and Labour and the Lib Dems etc.
The hypothetical implied if you wanted to Get Brexit Done you needed to say BXP. Reality was something else. The poll was not useless - it just needed to be read and understood correctly.
My leaflet count: 2 Tory, 0 anybody else.
On my street: 2 houses with Labour signs in people's gardens (1 has a giant Unite flag too), 2 with Tory ones. Amusingly the Unite/Labour one was first then a few days later the Tory one went up next door but one to the Unite/Labour one.
Plus Mike's letter which makes decent sense here in a LD-Con remainer marginal.
Timmy is now promoting himself as an "Independent Liberal". Here in W&L Postal Voters have had the equivalent of the Mike Smithson letter from Sarah Wallaston - really !
Any bets I should be placing?
It's a complete mess, and we're all guessing and nothing more than that. So I'll stick with my original prediction and say a working Tory majority - just.