YouGov have just updated their projection model for what will presumably be the third-last time (I'm guessing there'll be the usual update tomorrow afternoon, and then a final one after the polls close). Again, the news is mixed - the SNP are on 44 seats, which is two lower than yesterday, but two higher than the day before that. And if you look a little closer, the picture seems to be considerably more reassuring than the update of two days ago, because the SNP's floor (the minimum number of seats they'd be expected to win within a 95% confidence interval) now stands at 30 rather than 17. Their ceiling is now 53, which is only one down on yesterday. The most logical conclusion is that there was a particularly good batch of interviews for Labour about three days ago which brought a large number of central belt seats closer to being competitive, but that the threat has receded somewhat since then.
At UK-wide level, the situation is utterly bizarre. The consensus among pundits and "sources" that Labour are set for a pounding tomorrow is growing ever stronger, and you'd think eventually that would start to be reflected in what YouGov are hearing from their respondents. Instead, today's update shows the Tories slipping to yet another new low - just 33 seats ahead of Labour in a hung parliament.
UK-wide seats projection (YouGov) :
Conservatives 302
Labour 269
SNP 44
Liberal Democrats 12
Plaid Cymru 2
Greens 1
"Others" 2
Northern Ireland Parties 18
HUNG PARLIAMENT : Conservatives short by 24, Labour short by 57
Even the Conservatives' ceiling would only put them on 334 seats, which would be an overall majority of just 18 seats (or perhaps 20 after John Bercow is taken into account) - barely any improvement at all on what David Cameron achieved two years ago. I have to say these numbers aren't really passing the smell test, but we'll find out soon enough.
I have to admire them for sticking by the model when it's getting a caning from the punditocracy and (to some extent) other pollsters.
ReplyDeleteLet's hope it's at least close to the truth.
I just keep thinking about the New York Times' assessment that Clinton was "95 percent" certain to win immediately before the US election.
ReplyDeleteBrexit too.
I hope these analogies are pertinent.
To be clear.... I dont mean trumps victory was a good thing, just that the US commentariat rubbished the possibility of it happening..... Would love to see the UK equivalent with egg all over their faces....
The thought of a tory majority of any kind leaves me feeling despondent... Though perhaps it will boost indy..... Silver lining there....
The US commentariat did not see perfectly civilized middle class people voting for Trump. It happened right here in Raleigh, NC. Now Raleigh is a Democratic city, if you go by the affiliations of those who are in elected office, but that hides a significant number of unrepresented Republican and Unaffiliated voters.
ReplyDeleteOf course, Clinton threw away the Rust Belt. Trump spoke to those who had lost jobs to overseas and they heard him loud and clear. So, who is being heard by the voters in the UK? That is the ultimate question.
I am also reminded of Iain Davidson's comment two years ago, along the lines of "It didn't matter what Labour said on the doorstep. People had stopped listening to us."
Am I the only one who feels keep getting contradictory messages from not just YouGov but other pollsters.
ReplyDeleteI certainly question YouGov seat projection model.
I honestly think that unless there is a recovery in the Fib Doom vote in England then we are looking at another 5 years of a small Tory Majority.
I am not betting on this election at all as the Opinion Pollsters give me no confidence and have been feeding us a Hung Parly narrative like they did in 2015.
I am sniffing well deid fish from the lot of them. Unless Brexit England has suddenly gone OH EFF we better get the Tories out, but I don't get that sense at all.
Would 44 seats out of 59, be considered alright?
ReplyDeleteWhat's our bottom line number of seats in terms of maintaining pressure on U.K. govt for Indy Ref 2?
God knows. The Tories will gloat if the SNP lose one seat, they'll gloat if the SNP lose forty. Everything in between is just a matter of degree - although obviously losing the outright majority (ie. falling below 30) would make the 'mandate' argument much more difficult.
DeleteMore than 30 is a majority. It's plain and simple. The Tories have other problems now and will probably not be the interested in Scotland even if they win by a small amount.
Delete"The Tories have other problems now and will probably not be the interested in Scotland even if they win by a small amount"
DeleteYou jest?
The reason this election was called was to try to spike the next Indy ref.
@juteman, I agree that might have been part time of the reason. Surely though it was mostly because she thought she'd walk it in the rest of the UK?
DeleteI again (only IF) she wind in the rest of the UK, I think she will have bigger fish to fry. Think Northern Ireland and then the negotiations?
For me, the problem isn't how many seats will be lost, it's who we will be losing. John Nicolson will be a big loss, and it will be a tragedy if we lose Moray
ReplyDeleteJames, other sites are deducting the three to possible seven seats sinn fein wins from total needed.???
ReplyDeleteI don't think that we should be fretting too much re the YouGov projections. As James said, the results don't pass the sniff test. Some of the recent movements don't make any sense, such as the move of Orkney and Shetland from Likely SNP to Likely LibDem. To be honest, I suspect that the LibDems will win this seat - but I don't think that anyone thinks that can be due to a recent upsurge in LibDem support. Rather it reflects the fact that it was always probably a likely (or even safe) LibDem seat. The fact that the seat has moved from one category to another just highlights the fact that the YouGov model appears to be ever so slightly "random"...
ReplyDeleteWhat I would say is that the overall prediction in Scotland is likely to be close to the truth, but figures in individual seats must be subject to a huge amount of caution. However as they interview new people every day the figures for the seats Sholden get more accurate as the sample increases.
DeleteJames, what's your predictions?
ReplyDeleteI make a point of not predicting elections, except when people are missing something blindingly obvious - for example, I confidently predicted last year that Marine Le Pen would not be elected President of France, because the run-off system made it almost impossible for her. So here is my Le Pen-style statement of what should be pretty obvious for tomorrow - the Conservatives will hold Dumfriesshire, Clydesdale and Tweeddale, and will gain Berwickshire, Roxburgh and Selkirk. I also think it's overwhelmingly likely that the Liberal Democrats will hold Orkney and Shetland, although I'll be a bit more circumspect about that prediction because that constituency is a law unto itself.
DeleteAshcroft tweets ICM Poll CON 46% LAB 34% LDEM 7% UKIP 5% GRNS 2%..
ReplyDeleteNot much sign of a hung parliament there
Only a 1% increase in the Tory lead, which suggests the Labour-friendly pollsters may continue pointing to a hung parliament (or something close to it) in their final polls tonight. It's just down to who you believe.
DeleteICM Scottish subsample... SNP 49, Tories 29, Labour 21, LD 2... from the most Tory friendly pollster.
DeleteIf I can be a pedant, it's actually SNP 47, Tories 30, Labour 21 after turnout weighting.
DeleteGet my prediction out of the way:
ReplyDeleteI have some inside info BUT do not read too much into it (lol)
Snp to hold east Dunbartonshire, hold moray and hold perth.
Tories take Aberdeen south, 3 borders, abdn west and east ren.
Labour hold edin south
Libs take edin west, Orkney and Shetland
Snp to get 50 seats.
I don't have any inside information, but I think 50 may be very optimistic. I hope you're right, though. (Although I'd be gutted about Aberdeen South - in theory that should be tougher for the Tories than some of the others.)
DeleteIs the inside info positive or negative, compared to the above?
DeletePretty sure my friend lives in Aberdeen South (Stonehaven). They've been hammering that constituency!
DeleteStonehaven is in West Aberdeenshire & Kincardine (another of the Tories' top targets).
DeleteAgree it is very optimistic given everything that has been said, but doable on the day. 47 - 48 is probably more likely but thought I would go positive. East Lothian and north east fife might go, but Ive heard ne Fife is ok...
DeleteThe inside info is positive regarding the 3 'big beasts' that I have said I think will hold.
If Aberdeen South gets their vote out (SNP obvs) they might do it, but my gut says he might just miss out. The Tories and labour unionists are very motivated up here and I am not sure that Nicolas revelation yesterday helped with that....
Salmond is looking ok, Banff and Buchan since it was pointed out that the fish processing plants are under severe threat from a hard brexit seems to have calmed down.
Might be a surprise in Edinburgh that isnt good for SNP. We shall see.
Thankfully my job is not a pollster!
I have come to the conclusion that the Pollsters (ones that are public at least) are not to be trusted any more than the newspapers.
ReplyDeleteWho votes Tory in Scotland anyway?! What have they ever done for Scotland?
ReplyDeleteComres and icm subs for Scotland look very encouraging for the SNP
ReplyDelete