Friday 29 November 2019

Damn the Polls

So, there have been a bunch of bad polls out for Labour, and even the Leftie's friend, Survation, have recently given the Conservatives a rip-snorting 11% lead.  You Gov's much vaunted MRP poll - which pretty much nailed the result in 2015 - is currently predicting a comfortable majority for the Conservatives.

People have such touching faith in the opinion polls. I think the polls are wildly incorrect, on a 2017+ scale.

They may be right, they may be wrong.  Usually, they are wrong, often significantly so.  Remember, most pollsters had the conservatives winning a comfortable majority in 2017 ... and said there would be a hung parliament in 2015 ... and the Tories would win decisively in 2010.  Not forgetting the Brexit referendum and the Soccish independence referendum (which pollsters were predicting would be close).

In 2017 I had trust in Survation; this year I've lost faith in them. It's a bit like atheism - they say Christian's don't believe in 99.99% of all gods that have historically been worshipped, atheists feel the same, but add one more to the list. My hunch / hope is that Survation have messed up the BXP redistribution. Since the BXP pulled out of contesting Conservative seats, Survation have started showing large Conservative leads.

The bottom line is  - I'm finding it hard to believe that Boris Johnson is as popular as Blair in his heyday. 

I can not comprehend widescale enthusiasm for a blustering fool, liar, cheat and blowhard who gets booed wherever he goes. (But that's enough about me ...)

Nor can I imagine Labour's exciting and ambitious manifesto not catching on.

So inspite of the polls I'm feeling pretty cocky at the moment. At the start of the election I predicted Labour would finish as largest party with about 290 seats. Obviously, the opinion polls are saying that isn't going to happen, but we know they are bollocks. I may have been a bit optimistic, but we'll see. I think another stalemate - perhaps even staler than the last one - is a distinct possibility. And Labour finishing ahead is not out of the question.

Also bear in mind this wisdom from Tom Clark, considering the You Gov MRP poll:

Lesson 1 of 2015 is that it’s no use having more data if you’re data is biased. Back then, inspired by the cult of Nate Silver & his state-by-state stuff, we had loads of constituency polls and seat-by-seat projections that turned out to be tosh

https://twitter.com/prospect_clark/status/1199828481095479296

Good thread, by the way.

I think Labour will rally, and the Conservatives will sink, as people realise how insipid their manifesto is and Johnson continues to come across as an over-promoted dolt. It will be like a crapper version of 2019, with neither side managing to whip up the enthusiasm of two years ago.

I think the final result will be:
37% Con
35% Lab
13% Lib Dem
5% BXP
4% SNP
2.5% GRN
2% Irish
0.5% PC
1% Others
And with numbers like that (depending on how they translate into seats) it is hard to see how Johnson can get into Downing Street.

But I am worried about people feeling demoralised and despairing, which may make the polls self-fulfilling

No comments:

Unsurprising

 From the Guardian : The  Observer  understands that as well as backing away from its £28bn a year commitment on green investment (while sti...