|
|
Post by AdminSTB on Nov 7, 2016 18:27:11 GMT
We thought the general election would result in a hung parliament.
We thought Remain would win the EU referendum.
I think you know where I'm going with this. I've made my prediction of Clinton 323, Trump 215 - but could we all be in for a huge shock?
|
|
Richard Allen
Banned
Four time loser in VUKPOTY finals
Posts: 19,052
|
Post by Richard Allen on Nov 7, 2016 18:28:51 GMT
It can't be ruled out 100% but in my view it is very, very unlikely.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Nov 7, 2016 18:33:29 GMT
Never say never, even though unlikely
|
|
Sibboleth
Labour
'Sit on my finger, sing in my ear, O littleblood.'
Posts: 16,035
|
Post by Sibboleth on Nov 7, 2016 18:35:43 GMT
It isn't impossible, but then neither is a shockingly large win for Clinton. It's one of those funny elections in which the range of possible outcomes is very wide. Rationally speaking we have to assume it will be roughly in the middle of possibilities, but that doesn't mean it will be. 'But Brexit' is, however, a very bad reason to wonder about this.
|
|
jamie
Top Poster
Posts: 7,056 
Member is Online
|
Post by jamie on Nov 7, 2016 18:48:21 GMT
To reiterate a point I've made on other threads, Brexit is a poor comparison. Brexit was closer than this race (despite the punditry saying otherwise). Most importantly, the undecided voters looked like Brexit voters, while the undecideds voters here overall look more like Clinton voters.
A Trump victory is possible, but it has to be a massive balls up by the pollsters as opposed to shy Trump voters or anything like that.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Nov 7, 2016 19:17:32 GMT
Not 100% but he would need at least one serious shock in his favour plus everything else marginal going his way. It would need systematic and comprehensive polling error across a number of states.
|
|
Crimson King
Lib Dem
Be nice to each other and sing in tune
Posts: 9,845
|
Post by Crimson King on Nov 7, 2016 21:28:59 GMT
obviously no, not 100%
|
|
Jack
Reform Party
Posts: 8,699
Member is Online
|
Post by Jack on Nov 7, 2016 21:48:24 GMT
Not 100%, but maybe 99.9%.
|
|
|
|
Post by mrhell on Nov 7, 2016 22:08:29 GMT
Trump needs to defy the polls in the rust belt and/or do better than expected in strong Hispanic states.
|
|
|
|
Post by Andrew_S on Nov 7, 2016 22:24:03 GMT
|
|
jamie
Top Poster
Posts: 7,056
Member is Online
|
Post by jamie on Nov 7, 2016 22:37:40 GMT
RCP are a right wing leaning website who include dubious/biased polls while ignoring many polls favourable to Democrats. FiveThirtyEight have Clinton up 3.5% even when they include polls from Remington et al. so her polling lead is closer to 5%. |
|
|
|
Post by East Anglian Lefty on Nov 7, 2016 22:41:20 GMT
Trump doesn't have a particularly good polling day operation in any one battleground state, so for him to win then one or more of the assumptions of the bulk of polls will have to be wrong, so that they're systematically out by a few percent. Either the white share of the electorate needs to be particularly large (in fact it seems Hispanic turnout is sharply up), or Trump has to perform above expectations with whites (which would require a shy Trump effect amongst women or the college-educated), or he has to get non-negligible minority support. I personally do not see any of these as likely.
|
|
|
|
Post by froome on Nov 7, 2016 22:45:46 GMT
It's not likely, but Swan's opener is important. The only thing that can be predicted with certainty at the moment is the unpredictability of any election.
|
|
|
|
Post by Adam in Stroud on Nov 7, 2016 23:09:04 GMT
To reiterate a point I've made on other threads, Brexit is a poor comparison. Brexit was closer than this race (despite the punditry saying otherwise). Most importantly, the undecided voters looked like Brexit voters, while the undecideds voters here overall look more like Clinton voters. A Trump victory is possible, but it has to be a massive balls up by the pollsters as opposed to shy Trump voters or anything like that. That seems sound to me. I thought Leave would win the Referendum and I couldn't understand why the pundits were so sure they wouldn't. In this case, I can see why pundits expect an Electoral College victory for Clinton. Either way, thank God for the checks and balances in the US constitution. |
|
|
|
Post by finsobruce on Nov 8, 2016 0:03:46 GMT
I don't think it can be ruled out.
Voters seem enthused on both sides and just like here with the Referendum people like us deplored the nature of the campaign....but it has got people who don't usually pay much attention to politics very fired up indeed.
I'm not sure how much of a factor it is in reality but both Johnson and to a lesser extent, Stein were apparently polling quite well a couple of weeks ago but have now faded badly presumably (?) with left and right voters going back to the "mainstream" choice in each case.
|
|
|
|
Post by johnloony on Nov 8, 2016 2:00:00 GMT
No. I'm scared, I'm nervous, I'm on edge. I have been for weeks if not months. I have always believed that Clinton will win, but the mere fact that a Trump victory is even a slight possibility is scary. I won't be settled until Clinton 's victory is definite, and I'm hoping for a Clinton landslide. This is one case where a definite early result is preferable to a protracted psephologists' numberwang.
|
|
|
|
Post by Andrew_S on Nov 8, 2016 2:19:59 GMT
RCP are a right wing leaning website who include dubious/biased polls while ignoring many polls favourable to Democrats. FiveThirtyEight have Clinton up 3.5% even when they include polls from Remington et al. so her polling lead is closer to 5%. The figure isn't that much larger with 538, unless that's also a GOP-leaning website. |
|
jamie
Top Poster
Posts: 7,056
Member is Online
|
Post by jamie on Nov 8, 2016 7:44:22 GMT
RCP are a right wing leaning website who include dubious/biased polls while ignoring many polls favourable to Democrats. FiveThirtyEight have Clinton up 3.5% even when they include polls from Remington et al. so her polling lead is closer to 5%. The figure isn't that much larger with 538, unless that's also a GOP-leaning website. Well it's almost double the lead which I consider pretty different. 538 aren't GOP leaning, I just disagree with including dodgy pollsters (and that works both ways). |
|
Eastwood
Non-Aligned
Politically restricted post
Posts: 2,122
|
Post by Eastwood on Nov 8, 2016 9:11:19 GMT
The figure isn't that much larger with 538, unless that's also a GOP-leaning website. Well it's almost double the lead which I consider pretty different. 538 aren't GOP leaning, I just disagree with including dodgy pollsters (and that works both ways). 538 have a policy of including nearly all pollsters but weighting for reliability and quality and taking into account republican or democratic leaning "house effects". That is probably the correct approach so long as you assume all pollsters are at least trying to predict the election correctly and so their relative movement will be reliable and biases will remain correlated to other pollsters. I suspect jamie 's fear is that there may be some pollsters who are deliberately delivering incorrect results. If they are "just making it up" then we have no way of knowing if they change their relative bias as the election approaches and so even the house effects adjustment may not be any use. |
|
jamie
Top Poster
Posts: 7,056
Member is Online
|
Post by jamie on Nov 8, 2016 9:21:15 GMT
Well it's almost double the lead which I consider pretty different. 538 aren't GOP leaning, I just disagree with including dodgy pollsters (and that works both ways). 538 have a policy of including nearly all pollsters but weighting for reliability and quality and taking into account republican or democratic leaning "house effects". That is probably the correct approach so long as you assume all pollsters are at least trying to predict the election correctly and so their relative movement will be reliable and biases will remain correlated to other pollsters. I suspect jamie 's fear is that there may be some pollsters who are deliberately delivering incorrect results. If they are "just making it up" then we have no way of knowing if they change their relative bias as the election approaches and so even the house effects adjustment may not be any use. I don't necessarily think they are making it up, but I'm struggling to comprehend how pollsters like Trafalgar are getting Trump to 30% (Hillary at 40%) of the black vote. 538 do change the poll based on House effects but it's usually only 1-3 points, so the way out results still lower the average. Also, the problem is mainly recent as we've seen a bunch of GOP leaning pollsters produce quite a few dodgy polls all of a sudden these past couple of weeks e.g.; Trafalgar and Remington, which have really screwed up the averages. But ultimately we will know the results in about 12 hours  |
|
