Options
politicalbetting.com » Blog Archive » NEW PB / Polling Matters podcast: GE17 debrief. Breaking down

This week’s podcast is split into two parts.
0
This discussion has been closed.
Comments
How is it that with so many PBers spread across the country and with thousands of actual punters also spread across the country , hardly anyone picked up any signals regarding the election results.
Notable exception is, of course, David Herdson, very late in the day.
Would I be correct that most people even though they are putting up actual money tend to think the outcome will be what they would like it to be ?
Some of the swings in London and in the South and in Scotland were massive. I find it difficult to believe that a seasoned canvasser would not have picked some of the signals.
To be fair, Nick and Barnesian did say support was holding up. Even our cybernats who are normally not very shy either failed to pick up the huge [ albeit not as big as 2015 ] swings in Scotland.
The Tory swings in the North East of Scotland were anticipated but I cannot remember anyone suggesting Labour could win more than 3 seats there.
Wales. The perceived wisdom was that the Tories will finish on top !! And, the Midlands would be a bloodbath for Labour.
Where did the intelligence go wrong ?
I also said that Lamb was safe, and forecast a net rise in LD seats.
My anecdata was supportive of Labour and I did write on the 7th that while the number of seats for a Lab majority were implausible, a hung parliament was not too tall an order.
I did get herded to predict a Con majority of 76. Like over Brexit, I sensed what was coming, but didn't have the courage of my convictions.
Apparently YouGov were the pollsters that did that, although I cant say for certain if that is true
Look at the HuffPost non headline figures.
Over the campaign Don't Knows broke almost exclusively for Labour. People who had spent the last 2 years bad mouthing Corbyn voted for him on election day and then probably immediately started bad mouthing him as soon as they stepped out of the voting booth.
The Headline poll figures were hugely misleading.
Nothing else that I envisaged then happened, but I was right about where I live. Food for thought.
We should have offered £350m per week for the NHS and had Boris driving around a bus all over the country with that written all over it.
This time it was a Horlicks. However, YouGov modelling has to be praised. For accuracy and guts !!!
Canvassing down West the weekend before showed a real enthusiasm amongst labour voters that I had not seen since 1997. I think I also mentioned this
Thing is, most have got me on ignore.
As the tragic events have unfolded throughout the day the experts are pointing to the newly installed external cladding creating a ferocious outside fire that engulfed the whole building. This after the building had received full building regs and fire brigade approval in late 2016.
Experts have also stated that sprinklers would have made no difference
There are similarities to the external cladding on the Dubai sky scrapper and other buildings that have used this external covering and had catastrosphic fires
Everyone needs to give the Authorities the time to find out the cause and in the meantime say a prayer for all those who have suffered the most dreadful day of their lives
On the main Con/Lab battle, I was completely confused by the end, so I closed my spread bets. My central forecast was a big but not overwhelming Con majority, but I wasn't confident of it.
I remember some rather excitable LibDems were talking about Bermondsey going yellow. I did say though the Labour majority would be over 10000. Kingston and Surbiton also went yellow comfortably.
And, what about the UKIP capital city, Lincoln ?
Let's face it, collectively, we failed miserably across all political opinions.
Note as usual the response rate was almost 100% , the voters fill in a secret mock ballot paper .
* To be fair nobody is, as forecasting in general is awful across essentially all fields.
I wasn't canvassing though, and lost money on the election, so what do I know?
Right until the end they were campaigning defensively.
The odds being offered on Labour in London were ridiculously long based on the local demographics.
I changed tack after that and sought value on Labour, and against the Lib Dems and SNP. I just prayed the Tory malaise didn't spread to Scotland !
In the end Ynys Mons, East Lothian and Leeds NW were all Labour bets at stupid prices well won - which paid and then some for the daft Tory gets in the Gower and so forth.
The exit poll was better for my pocket than I first realised I think.
https://twitter.com/GuyVerhofstadt/status/874894288462761984
Really not acceptable. Not acceptable at all.
Going hard on the anti-democracy ticket is a brave move after that.
I knew that was a serious mistake and would cost votes.
Likewise the Conservatives made numerous other mistakes and all the lost votes added up.
But the media aren't interested. For now.
@AndrewCooper__: Pre-referendum, only 23% thought Leave vote would mean UK leaving EU & the single market. 50% thought we'd leave EU & stay in single market
@AndrewCooper__: @DAaronovitch Research during the referendum found most voters definitely didn't think we'd be outside the single market; thought it was 'project fear'.
@AndrewCooper__: @DAaronovitch Focus groups said over & again "they need us more than we need them". Many felt this meant EU would have to let us stay in single market.
@AndrewCooper__: @DAaronovitch Many voters rejected out of hand the whole idea that the UK might face trade-offs such as single market vs. control of free movement.
Chancellor Philip Hammond joined forces with Home Secretary Amber Rudd to demand the weakened PM prioritise jobs over tough immigration control
https://www.thesun.co.uk/news/3802303/chancellor-and-other-top-tories-urge-theresa-may-to-put-jobs-before-immigration-in-talks/
And the imagery will remind many oldies about the worst features of the 1970s and 1980s.
So, in short, I started out a big skeptic of the Con landslide and ended up a small skeptic after reading PB. I still made a couple of months salary.
Ordinarily a Shadow Chancellor who has just failed to be part of an election-winning cabinet calling for civil unrest would be cause for alarm in all quarters. Can't win through the ballot boxes so just take the to the street - that is dangerous talk. But I would be amazed if this rates much comment.
I said that the LibDems' tactical vote was unwinding back to Labour; and that Labour voters were loud and proud this time. But it would have been impossible to discern the national result from the SW. Even Plymouth Moor View was safe, because of a very strong local candidate, whose seat really should have fallen if Labour were going to advance as they did.
https://twitter.com/officiaIwinemom/status/874702862730833921
Long may they continue to do so!
I'm assuming that inciting the overthrow of a democratically elected government is somehow against the rules?
Q1) "Would I be correct that most people even though they are putting up actual money tend to think the outcome will be what they would like it to be ?"
A1) Yes, or thereabouts. Most people in the real world who bet on politics do so as an affirmation of support for their party or to win an argument. Thankfully there are some (not enough!) gamblers and analysts on PB who don't allow their partisanship to override their healthy pursuit of profit. BlackRook and AndyJS stand out, tho the former has to be nudged every once in a while, and there are others.
Q2) "How is it that with so many PBers spread across the country and with thousands of actual punters also spread across the country , hardly anyone picked up any signals regarding the election results."
A2) Partisan shouting drowned out nonpartisan analysis. The analysis of the YouGov model was robust but intended to dismiss not assess, and not matched by a similar robust analysis of the other polls and models. I (and a couple of others, Casino_Royale was one I think) were sufficiently cognizant of it to steer clear of Con absolute majority, tho CR changed his mind in the latter stages. Plus my usual touchstone (postal vote sampling) didn't bark: there were no reports of unusual patterns. Canvassing reports were either pro-Con (that Labour Uncut guy) or were dismissed if pro-Lab (David Herdson?)
In a genuine spirit of inquiry, if you have any information source from prior to election day that a) pointed to the results, b) wasn't a product of fantasy and c) can be used again, I'd be grateful if you could make it known
He is a Trot.
After that, I made a small (and dumb) error, buying Conservative seats at 368. Fortunately, SpreadEx limited me to £2/seat, and SPIN wouldn't take my money at all.
I did take some money from those who thought the LibDems were in with a shout in Vauxhall. (Disclaimer: they weren't.) And I lost a tiny amount on Argyll & Bute (which I still need to pay).
My big call was that the LDs would do better in seat terms than people thought. On the day of the election I called 12 seats. (Someone joked that I'd made that prediction in 2015, and was just hoping it would be right this time around.) However, full disclosure, is that I thought the LDs would take 12 seats on a c. 10% vote share. I correctly predicted tactical voting would return, but I did not expect the LDs would go backwards in national vote share.
Of course, I like to think that I've been broadly right (and counter PB-consensus) on the LDs three elections in a row: 2015, 2016 Holyrood, and now 2017. Given that, I will make no more LD forecasts so as to ensure my record remains unsullied.
I also suspect there was a few people who voted Labour thinking they were being ironic and that it would be good fun to see more of Diane Abbott. In the same way that entertainment acts which are so-shite-they're-hilarious get fans.
Was it here or on another forum I said "Whoever wins, it'll be by a three figure majority"?
A significant portion of the electorate doesn't care if the LotO supports terrorist murderers, as long as he gives them other people's money.
https://twitter.com/mrjamesob/status/875021436615839746
The gradual swing to Labour would have been picked up. But ICM and ComRes by pooh-poohing Yougov and Survation successfully stopped that.
Apart from May, Martin Boon is the biggest loser of this election.
I imagine that some Labour supporters from the sane wing of the party will have made a similar calculation this time.
[Fwiw he thought the manifesto was dire and that May was not helpful)
They want to collape the minority government of headless chickens, for new elections not a revolution.
Victoria Derbyshire this morning was simply appalling. Full of emotive language and opinion-giving, very little news reporting.
The coverage has been very poor - compared to the immediate aftermath of the London Bridge events. Today they have gone with every rumour or ill-informed speculation rather than waiting for real facts.
The problem was wildly differing swings in different places.
Maybe we need an interpretist study, with a correspondent in each constituency?
Your prediction of certain Labour defeats in Cambridge and Hampstead may have helped me get 5/1 on Labour.
I would like to feel rather proud that after carefully analysing the demographics I discovered that Labour to hold Westminster N at 7/2 was incredible value.
But as the swing in Westminster North was similar to that in nearby seats I'm not sure my analysis had any merit after all.
I live in the city of Lincoln and I can assure you that it's about as far from being "UKIP capital city" as you can imagine. UKIP have never even taken a single council seat in the city of Lincoln let alone ever looked like doing anything here at a general election... thankfully. Even in 2015 they only reached 12% here. Lincoln is not at all like Boston or any of the other places in more rural Lincolnshire where UKIP have actual history of success.
https://grenfellactiongroup.wordpress.com/2016/11/20/kctmo-playing-with-fire/
Reading it now it is horribly prescient.