Different pollster, the pollster with Trump ahead was also the only pollster who had Trump ahead in Michigan in 2016, Trafalgar Group
Nate Silver's site gives a very low rating to the Trafalgar Group.
Well the election 2016 result gives a very high rating to Trafalgar Group as it was the only pollster who correctly had Trump ahead in Michigan and Pennsylvania and without Trump winning those states and Wisconsin Hillary would have won the EC and now be President
They also overestimated Trump by 2% in Michigan though, though that is within margin of error at least unlike Florida and no other pollster did well in Michigan.
Pennsylvania was the one result they got right.
Yes. You tell him.
So, not to divert, but on PT the scores on the doors for our Moderates Panel (which you insisted on) opining on whether Labour 2015 was a moderate offering is 4-1 for Yes.
Should we call that now as Yes? Or do you want to wait for another 7 opinions from other moderates to get the full 12?
(ps: forgot lady g and nick p as non-moderates).
I never insisted upon a panel nor did I agree to 12.
If you want a panel you need to have it be impartial: anyone who voted for it would have to recuse themselves.
Ok, so we should call it. Sensible decision. All are sick and tired of it.
Labour's 2015 offering WAS - per an independent panel of PB Moderates - moderate.
Back to Biden vs the ghastly Trump ...
No it wasn't and you didn't assemble an independent panel of PB Moderates.
Who were the "independent panel of PB Moderates"? I'm assuming you're not counting people who voted Labour, who went for such a non-moderate 2015 manifesto?
GOP congressional candidates performed worse than Trump in 2018, they only did better in 2016 because they got Trump voters voting down ballot for them too after voting for Trump first, those voters did not turn out for them again in 2018 with Trump not on the ballot
Trump wasn't on the ballot in 2018! They can't perform worse than someone not in the ballot.
Your arguement makes no sense. Trump was massively popular in 2016, so popular that more people voted for GOP congressional candidates than voted for Trump.
That doesn't make sense.
They voted for Trump then down ballot for the GOP though with little enthusiasm while the GOP also got the traditional GOP vote a few of whom voted for Hillary, in 2018 without as many Trump voters the traditional GOP vote was not enough and they did worse than Trump in 2016
Alistair's point, which you haven't answered, is that Republicans won the popular vote in the House elections in 2016 by 1.1% at the same time Trump was losing by 2.1%.
So there were more people passing on Trump but voting GOP in other contests on the ballot than the other way around.
Your point that turnout was lower in 2018 is utterly facile - it always is in midterm elections, not just when Trump is the President.
Given your liking for cherry picking, here's one for you. More people in Wisconsin voted for Romney in his defeat in 2012 than Trump in his great triumph in 2016. It doesn't really mean all that much - but it illustrates Clinton was poor at driving turnout in an important state, and that Trump wasn't quite engaging a whole new electorate in the way you assert.
Which I have answered and you have deliberately ignored because you do not want to believe there is such a thing as a Trump Democrat. The fact Trump got a higher voteshare in 2016 than the GOP candidates did in 2018 only confirms that, in 2016 the GOP candidates got Trump voters and traditional GOP voters (a few of whom yes did not vote Trump for president), in 2016 they only got traditional GOP voters, Trump voters did not turn out.
More people voted for Trump than Romney in Iowa, Ohio, Michigan and Pennsylvania and Florida ie all the Obama states in 2012 that Trump won in 2016, Wisconsin was the only exception to that, though Trump still got a higher voteshare than Romney did there.
Yet in 2019 Survation were showing a big Tory lead in their final poll unlike their neck and neck final 2017 poll.
Trafalgar Group were the only pollster who correctly predicted the EC win for Trump in 2016 and their 2020 state polls are still showing a Trump EC win
The whole essence of your argument seems to be because Trafalgar got aspects of the election right in 2016 they are going to get it right in 2020.
Maybe - they don't publish their crosstabs so we don't know about their sampling and methodology. As has also been pointed out, not everybody views Trafalgar (and presumably Rasmussen) as favourably and Trafalgar's record isn't uniformly strong. As has been noted, they have a curious tendency for overstating the Republican vote (well, there's a thing).
The other factor is the much diminished number of undecided voters compared to this stage in 2016. Most people have mostly decided as witnessed by another pollster giving a Trump lead in a marginal State:
As for last December, I'd argue Opinium were better than Survation and Mori not that bad. In the last 2-3 days every pollster bar Panelbase had an 11-13 % Conservative lead so the pollsters basically got it right.
In 2019 every pollster had a big Tory lead, there were no polling discrepancies as there are in the US with Trafalgar and Rasmussen for example.
In 2008 even Rasmussen had a big Obama lead so if Biden was heading for a landslide he should have a big lead of 5 to 10% with them too, he doesn't
Different pollster, the pollster with Trump ahead was also the only pollster who had Trump ahead in Michigan in 2016, Trafalgar Group
Nate Silver's site gives a very low rating to the Trafalgar Group.
Well the election 2016 result gives a very high rating to Trafalgar Group as it was the only pollster who correctly had Trump ahead in Michigan and Pennsylvania and without Trump winning those states and Wisconsin Hillary would have won the EC and now be President
We've been through this before, but you do have a tendency to cherry pick.
In 2018, a few days out from the election Trafalgar had McSally (GOP) winning Arizona by 2% (she lost by 2%), Heller (GOP) winning Nevada by 3% (he lost by 5%), Cruz (GOP) winning Texas by 9% (squeaked in by under 2%).
That's cherry picking on my part too, as they had some better ones that night (they called it the right way in Florida, for example, albeit overstating GOP lead). But if you're thinking about the quality of a pollster, you need to look more broadly.
Yes, the pollster with the most extreme effect in the direction of a particular party will sometimes look good by calling wins others don't, but they will also have false alarms. In 2018, you'd have certainly been better listening to other pollsters.
All irrelevant as Trump was not on the ballot in 2018 so many Trump 2016 voters stayed home in 2018.
Not irrelevant at all. A good pollster has a good record in all elections. Your argument is one which tends towards "never mind what my watch says now - it's absolutely perfect twice a day."
No completely irrelevant. You are forgetting the fact that many Trump 2016 voters will not have voted for a Republican candidate for President since Bush Snr in 1988, the last Republican to win Michigan and Pennsylvania, or Reagan in 1984, the last Republican to win Wisconsin (with 1988 effectively Reagan's third term, they switched to Bill Clinton in 1992).
They are in effect white working class Trump Democrats much like the Reagan Democrats of the 1980s, they will only vote for Trump, they will not bother to turn out to vote for other Republicans. Trafalgar were the only pollster to pick them up in 2016
Citation needed. Your are acting like the Dems has a 50 pt lead in those states prior to 2016.
And you still haven't explained why Trump was such a mega draw when the GOP congressional candidates outperformed him in 2016.
No citation needed, Trump was the first Republican to win Wisconsin, Pennsylvania and Michigan since the 1980s. He also got the highest number of votes for any Republican candidate in Pennsylvania since 1988.
GOP congressional candidates performed worse than Trump in 2018, they only did better in 2016 because they got Trump voters voting down ballot for them too after voting for Trump first, those voters did not turn out for them again in 2018 with Trump not on the ballot
In 2018 wasnt the thing that did for the Republicans that the Trumpers did turn out, but that there was a far bigger Democrat turnout than usual?
In 2019, I was rubbished for saying I didn't think it would be close, and that the Tories would win a big majority.
Based on current polls (and they may of course change between now and November 3rd), I would say the same about the US Presidential election.
If the polls given Biden a seven point lead the night before, it's not going to be 270-268 or anything like that, it's going to be at a 100 vote Electoral College margin.
Current forecast (with the proviso that polls change) big Biden win.
I didn't think it would be close last December either. The writing was on the wall the night the ComRes poll only Johnson of the Conservative leadership candidates winning an overall majority and pulling the bulk of the Brexit Party vote to the Conservatives.
Had Hunt or Javid prevailed rather than Johnson, Farage and the Brexit Party would have continued polling strongly in the high teens/low twenties making a GE impossible for the Conservatives - indeed, the question would have been whether there would have been defections from the Conservatives to the Brexit party.
As for this year's Presidential election, the Democrats in 2020 are going to be like Labour in 1997 and it may be that in concentrating on the marginal states they might miss winning more unlikely targets but I'm doubtful.
The question for me is the point at which the Trump vote will begin to realise the game is up and melt away leading to a bigger Biden win.
Yet in 2019 Survation were showing a big Tory lead in their final poll unlike their neck and neck final 2017 poll.
Trafalgar Group were the only pollster who correctly predicted the EC win for Trump in 2016 and their 2020 state polls are still showing a Trump EC win
Trafalgar have Biden picking up Pennsylvania but otherwise unchanged from 2016 in their latest polling
Do Trafalgar have you in their pay roll? You bring them up in most US threads and ignore their not so great 2018 results when other pollsters did much better after changing their methodology.
It should also be pointed out they haven't released polls in some battleground states since June/July (or at all) but your are happy to say they predict a EC win for trump even though we are in September.
They might be right, but their track record in 2018 and being the outliers against pollsters who did better in 2018, I would not be putting any money down purely based on their polls.
Different pollster, the pollster with Trump ahead was also the only pollster who had Trump ahead in Michigan in 2016, Trafalgar Group
Nate Silver's site gives a very low rating to the Trafalgar Group.
Well the election 2016 result gives a very high rating to Trafalgar Group as it was the only pollster who correctly had Trump ahead in Michigan and Pennsylvania and without Trump winning those states and Wisconsin Hillary would have won the EC and now be President
We've been through this before, but you do have a tendency to cherry pick.
In 2018, a few days out from the election Trafalgar had McSally (GOP) winning Arizona by 2% (she lost by 2%), Heller (GOP) winning Nevada by 3% (he lost by 5%), Cruz (GOP) winning Texas by 9% (squeaked in by under 2%).
That's cherry picking on my part too, as they had some better ones that night (they called it the right way in Florida, for example, albeit overstating GOP lead). But if you're thinking about the quality of a pollster, you need to look more broadly.
Yes, the pollster with the most extreme effect in the direction of a particular party will sometimes look good by calling wins others don't, but they will also have false alarms. In 2018, you'd have certainly been better listening to other pollsters.
All irrelevant as Trump was not on the ballot in 2018 so many Trump 2016 voters stayed home in 2018.
Not irrelevant at all. A good pollster has a good record in all elections. Your argument is one which tends towards "never mind what my watch says now - it's absolutely perfect twice a day."
No completely irrelevant. You are forgetting the fact that many Trump 2016 voters will not have voted for a Republican candidate for President since Bush Snr in 1988, the last Republican to win Michigan and Pennsylvania, or Reagan in 1984, the last Republican to win Wisconsin (with 1988 effectively Reagan's third term, they switched to Bill Clinton in 1992).
They are in effect white working class Trump Democrats much like the Reagan Democrats of the 1980s, they will only vote for Trump, they will not bother to turn out to vote for other Republicans. Trafalgar were the only pollster to pick them up in 2016
Citation needed. Your are acting like the Dems has a 50 pt lead in those states prior to 2016.
And you still haven't explained why Trump was such a mega draw when the GOP congressional candidates outperformed him in 2016.
No citation needed, Trump was the first Republican to win Wisconsin, Pennsylvania and Michigan since the 1980s. He also got the highest number of votes for any Republican candidate in Pennsylvania since 1988.
GOP congressional candidates performed worse than Trump in 2018, they only did better in 2016 because they got Trump voters voting down ballot for them too after voting for Trump first, those voters did not turn out for them again in 2018 with Trump not on the ballot
In 2018 wasnt the thing that did for the Republicans that the Trumpers did turn out, but that there was a far bigger Democrat turnout than usual?
In 2016 Trump got 62,984,828 votes, in 2018 the Republicans got only 50,861,970 votes, a huge gap of 12 million votes.
By contrast the Democrats got 60,572,245 votes in 2018 so far closer to the 65,853,514 votes Hillary got in 2016 and a small gap of only 5 million votes.
So Trump voters did not turn out in 2018 but Democrats did
"My main bet is now laying Trump which covers me if something happened to Biden and Kamala Harris became the flag-carrier."
There's also the possibility that something might happen to Trump. (E.g. either on the current military story or something else, what if he insists and insists that he didn't do something, asks "what kind of animal would do that?" etc. - as he has done with this story - and then a tape or clip appears that proves unambiguously that he did do it? A Republican who isn't Trump might still run away with this. Mike Pence is at 390. The others are all at 1000.
I don't think there's time to put another Republican on the ticket. Don't forget candidates have already been filed in pretty much all states, and early voting has even begun.
The only scenario I could see that is even "possible" is for Trump to step down due to medical issues (or a truly horrendous scandal) and Pence to ascend to the Presidency.
Ascending to the presidency doesn't keep Mike Pence in it after 20 January. If there's a way for him to be in the job after that date, then the same way is open to another Republican replacement for Trump.
In 2019 every pollster had a big Tory lead, there were no polling discrepancies as there are in the US with Trafalgar and Rasmussen for example.
In 2008 even Rasmussen had a big Obama lead so if Biden was heading for a landslide he should have a big lead of 5 to 10% with them too, he doesn't
The last Rasmussen poll at the end of August had Biden up 49-45 (their previous poll had Biden up 46-45).
The one interesting aspect is Rasmussen poll 2500 Likely voters so their MoE is 2% so theoretically it could be a tie or Biden could be up by 6.
Most of the polling sample sizes are much smaller - the Harris poll has 2834 Registered Voters and they have Biden up 46-40 on a 1.8% MoE.
As a counter, the CNN poll has only 997 Likely Voters and an MoE of 4% so the 8-point Biden lead 51-43 could also be a tie (47-47 like Rasmussen) or it could be 55-39 which would be a Biden landslide.
Again, simply taking headline poll numbers and weaponising them to make a political point is grossly misleading.
What is the sample size, what is the methodology, what is the margin of error?
These are the questions we should be asking of EVERY poll.
If he hadn't bunked off the Trial in Sweden, he would have finished his sentence by now.
Although I agree he shouldn't have breached bail in 2012 by holing up in the Ecuadorian Embassy, his concern has always been extradition to the US - Sweden is a sideshow.
Indeed, he may well not have served any time at all in Sweden. That case has now been dropped - albeit Swedish investigators claim due to the passage of time making memories less reliable rather than a fundamental problem with their case. So it's questionable if there would have been a conviction.
"My main bet is now laying Trump which covers me if something happened to Biden and Kamala Harris became the flag-carrier."
There's also the possibility that something might happen to Trump. (E.g. either on the current military story or something else, what if he insists and insists that he didn't do something, asks "what kind of animal would do that?" etc. - as he has done with this story - and then a tape or clip appears that proves unambiguously that he did do it? A Republican who isn't Trump might still run away with this. Mike Pence is at 390. The others are all at 1000.
I don't think there's time to put another Republican on the ticket. Don't forget candidates have already been filed in pretty much all states, and early voting has even begun.
The only scenario I could see that is even "possible" is for Trump to step down due to medical issues (or a truly horrendous scandal) and Pence to ascend to the Presidency.
Ascending to the presidency doesn't keep Mike Pence in it after 20 January. If there's a way for him to be in the job after that date, then the same way is open to another Republican replacement for Trump.
And how are they going to remove Trump from the ticket? He would literally have to be dead before he lets go of being the candidate.
A tape saying all veterans are losers and cowards wouldn't cut it. Not even close.
In 2019 every pollster had a big Tory lead, there were no polling discrepancies as there are in the US with Trafalgar and Rasmussen for example.
In 2008 even Rasmussen had a big Obama lead so if Biden was heading for a landslide he should have a big lead of 5 to 10% with them too, he doesn't
The last Rasmussen poll at the end of August had Biden up 49-45 (their previous poll had Biden up 46-45).
The one interesting aspect is Rasmussen poll 2500 Likely voters so their MoE is 2% so theoretically it could be a tie or Biden could be up by 6.
Most of the polling sample sizes are much smaller - the Harris poll has 2834 Registered Voters and they have Biden up 46-40 on a 1.8% MoE.
As a counter, the CNN poll has only 997 Likely Voters and an MoE of 4% so the 8-point Biden lead 51-43 could also be a tie (47-47 like Rasmussen) or it could be 55-39 which would be a Biden landslide.
Again, simply taking headline poll numbers and weaponising them to make a political point is grossly misleading.
What is the sample size, what is the methodology, what is the margin of error?
These are the questions we should be asking of EVERY poll.
This.
Compared to UK polling, US polling is all over the place. There’s often small sample sizes, subsamples used for headlines, no common rules on publishing data, no record of the full set of questions nor their order, very partisan commissioners of polling etc etc.
In 2019 every pollster had a big Tory lead, there were no polling discrepancies as there are in the US with Trafalgar and Rasmussen for example.
In 2008 even Rasmussen had a big Obama lead so if Biden was heading for a landslide he should have a big lead of 5 to 10% with them too, he doesn't
The last Rasmussen poll at the end of August had Biden up 49-45 (their previous poll had Biden up 46-45).
The one interesting aspect is Rasmussen poll 2500 Likely voters so their MoE is 2% so theoretically it could be a tie or Biden could be up by 6.
Most of the polling sample sizes are much smaller - the Harris poll has 2834 Registered Voters and they have Biden up 46-40 on a 1.8% MoE.
As a counter, the CNN poll has only 997 Likely Voters and an MoE of 4% so the 8-point Biden lead 51-43 could also be a tie (47-47 like Rasmussen) or it could be 55-39 which would be a Biden landslide.
Again, simply taking headline poll numbers and weaponising them to make a political point is grossly misleading.
What is the sample size, what is the methodology, what is the margin of error?
These are the questions we should be asking of EVERY poll.
And Trump could still eke out an EC win, just, even with Biden 4% ahead in the popular vote. A 5% or higher Biden popular vote lead and Biden should win the EC regardless.
An increasingly desperate Boris Johnson has ordered his staff to step up personal attacks on the Labour leader Keir Starmer and his record as a lawyer, as confidence in the prime minister’s leadership collapses among Tory party members.
The Observer has been told that Johnson was so furious after last Wednesday’s prime minister’s questions – where he was asked to withdraw comments he made about the Labour leader and the IRA by the Speaker, Lindsay Hoyle – that he turned on his staff for leaving him under-prepared, and asked them to come up with more attack lines on the Labour leader’s career as a lawyer.
“He was furious,” said a well-placed source. “He told his team and people at CCHQ [Conservative campaign headquarters] that he wanted them to go after Starmer’s legal record and double down on the attacks on him.”
An increasingly desperate Boris Johnson has ordered his staff to step up personal attacks on the Labour leader Keir Starmer and his record as a lawyer, as confidence in the prime minister’s leadership collapses among Tory party members.
The Observer has been told that Johnson was so furious after last Wednesday’s prime minister’s questions – where he was asked to withdraw comments he made about the Labour leader and the IRA by the Speaker, Lindsay Hoyle – that he turned on his staff for leaving him under-prepared, and asked them to come up with more attack lines on the Labour leader’s career as a lawyer.
“He was furious,” said a well-placed source. “He told his team and people at CCHQ [Conservative campaign headquarters] that he wanted them to go after Starmer’s legal record and double down on the attacks on him.”
Maybe he should just concentrate on answering the question rather than look for cheap one liners. I thought the big file they have contains the answers.
An increasingly desperate Boris Johnson has ordered his staff to step up personal attacks on the Labour leader Keir Starmer and his record as a lawyer, as confidence in the prime minister’s leadership collapses among Tory party members.
The Observer has been told that Johnson was so furious after last Wednesday’s prime minister’s questions – where he was asked to withdraw comments he made about the Labour leader and the IRA by the Speaker, Lindsay Hoyle – that he turned on his staff for leaving him under-prepared, and asked them to come up with more attack lines on the Labour leader’s career as a lawyer.
“He was furious,” said a well-placed source. “He told his team and people at CCHQ [Conservative campaign headquarters] that he wanted them to go after Starmer’s legal record and double down on the attacks on him.”
Maybe he should just concentrate on answering the question rather than look for cheap one liners. I thought the big file they have contains the answers.
Johnson looking to blame someone else again, when he can't be arsed to do the basic homework.
The Tory party have really bought themselves a poundshop PM.
Different pollster, the pollster with Trump ahead was also the only pollster who had Trump ahead in Michigan in 2016, Trafalgar Group
Nate Silver's site gives a very low rating to the Trafalgar Group.
Well the election 2016 result gives a very high rating to Trafalgar Group as it was the only pollster who correctly had Trump ahead in Michigan and Pennsylvania and without Trump winning those states and Wisconsin Hillary would have won the EC and now be President
We've been through this before, but you do have a tendency to cherry pick.
In 2018, a few days out from the election Trafalgar had McSally (GOP) winning Arizona by 2% (she lost by 2%), Heller (GOP) winning Nevada by 3% (he lost by 5%), Cruz (GOP) winning Texas by 9% (squeaked in by under 2%).
That's cherry picking on my part too, as they had some better ones that night (they called it the right way in Florida, for example, albeit overstating GOP lead). But if you're thinking about the quality of a pollster, you need to look more broadly.
Yes, the pollster with the most extreme effect in the direction of a particular party will sometimes look good by calling wins others don't, but they will also have false alarms. In 2018, you'd have certainly been better listening to other pollsters.
All irrelevant as Trump was not on the ballot in 2018 so many Trump 2016 voters stayed home in 2018.
Not irrelevant at all. A good pollster has a good record in all elections. Your argument is one which tends towards "never mind what my watch says now - it's absolutely perfect twice a day."
No completely irrelevant. You are forgetting the fact that many Trump 2016 voters will not have voted for a Republican candidate for President since Bush Snr in 1988, the last Republican to win Michigan and Pennsylvania, or Reagan in 1984, the last Republican to win Wisconsin (with 1988 effectively Reagan's third term, they switched to Bill Clinton in 1992).
They are in effect white working class Trump Democrats much like the Reagan Democrats of the 1980s, they will only vote for Trump, they will not bother to turn out to vote for other Republicans. Trafalgar were the only pollster to pick them up in 2016
Citation needed. Your are acting like the Dems has a 50 pt lead in those states prior to 2016.
And you still haven't explained why Trump was such a mega draw when the GOP congressional candidates outperformed him in 2016.
No citation needed, Trump was the first Republican to win Wisconsin, Pennsylvania and Michigan since the 1980s. He also got the highest number of votes for any Republican candidate in Pennsylvania since 1988.
GOP congressional candidates performed worse than Trump in 2018, they only did better in 2016 because they got Trump voters voting down ballot for them too after voting for Trump first, those voters did not turn out for them again in 2018 with Trump not on the ballot
In 2018 wasnt the thing that did for the Republicans that the Trumpers did turn out, but that there was a far bigger Democrat turnout than usual?
In 2016 Trump got 62,984,828 votes, in 2018 the Republicans got only 50,861,970 votes, a huge gap of 12 million votes.
By contrast the Democrats got 60,572,245 votes in 2018 so far closer to the 65,853,514 votes Hillary got in 2016 and a small gap of only 5 million votes.
So Trump voters did not turn out in 2018 but Democrats did
Andrew Neil is now doing exactly the thing he was criticising the Co-op for. He's "no platforming" them because the person running their social media account (some poor sod who may well lose their job) was fairly clear they didn't like the Spectator's editorial line.
The only logical position for Neil and the Spectator is "the Co-op can advertise with us if they want or not if they don't - it's entirely up to them but won't alter our editorial position one jot either way".
Andrew Neil is now doing exactly the thing he was criticising the Co-op for. He's "no platforming" them because the person running their social media account (some poor sod who may well lose their job) was fairly clear they didn't like the Spectator's editorial line.
The only logical position for Neil and the Spectator is "the Co-op can advertise with us if they want or not if they don't - it's entirely up to them but won't alter our editorial position one jot either way".
Does the Co-Op actually advertise in the Spectator?
In 2019 every pollster had a big Tory lead, there were no polling discrepancies as there are in the US with Trafalgar and Rasmussen for example.
In 2008 even Rasmussen had a big Obama lead so if Biden was heading for a landslide he should have a big lead of 5 to 10% with them too, he doesn't
The last Rasmussen poll at the end of August had Biden up 49-45 (their previous poll had Biden up 46-45).
The one interesting aspect is Rasmussen poll 2500 Likely voters so their MoE is 2% so theoretically it could be a tie or Biden could be up by 6.
Most of the polling sample sizes are much smaller - the Harris poll has 2834 Registered Voters and they have Biden up 46-40 on a 1.8% MoE.
As a counter, the CNN poll has only 997 Likely Voters and an MoE of 4% so the 8-point Biden lead 51-43 could also be a tie (47-47 like Rasmussen) or it could be 55-39 which would be a Biden landslide.
Again, simply taking headline poll numbers and weaponising them to make a political point is grossly misleading.
What is the sample size, what is the methodology, what is the margin of error?
These are the questions we should be asking of EVERY poll.
And Trump could still eke out an EC win, just, even with Biden 4% ahead in the popular vote. A 5% or higher Biden popular vote lead and Biden should win the EC regardless.
In 2019 every pollster had a big Tory lead, there were no polling discrepancies as there are in the US with Trafalgar and Rasmussen for example.
In 2008 even Rasmussen had a big Obama lead so if Biden was heading for a landslide he should have a big lead of 5 to 10% with them too, he doesn't
The last Rasmussen poll at the end of August had Biden up 49-45 (their previous poll had Biden up 46-45).
The one interesting aspect is Rasmussen poll 2500 Likely voters so their MoE is 2% so theoretically it could be a tie or Biden could be up by 6.
Most of the polling sample sizes are much smaller - the Harris poll has 2834 Registered Voters and they have Biden up 46-40 on a 1.8% MoE.
As a counter, the CNN poll has only 997 Likely Voters and an MoE of 4% so the 8-point Biden lead 51-43 could also be a tie (47-47 like Rasmussen) or it could be 55-39 which would be a Biden landslide.
Again, simply taking headline poll numbers and weaponising them to make a political point is grossly misleading.
What is the sample size, what is the methodology, what is the margin of error?
These are the questions we should be asking of EVERY poll.
This.
Compared to UK polling, US polling is all over the place.
An increasingly desperate Boris Johnson has ordered his staff to step up personal attacks on the Labour leader Keir Starmer and his record as a lawyer, as confidence in the prime minister’s leadership collapses among Tory party members.
The Observer has been told that Johnson was so furious after last Wednesday’s prime minister’s questions – where he was asked to withdraw comments he made about the Labour leader and the IRA by the Speaker, Lindsay Hoyle – that he turned on his staff for leaving him under-prepared, and asked them to come up with more attack lines on the Labour leader’s career as a lawyer.
“He was furious,” said a well-placed source. “He told his team and people at CCHQ [Conservative campaign headquarters] that he wanted them to go after Starmer’s legal record and double down on the attacks on him.”
If there was something significant then BoZo would have used it already.
People as mendacious as BoZo really struggle to comprehend that not everyone else is as bent as they are. Starmer is a bit square, but that is about it.
In 2019 every pollster had a big Tory lead, there were no polling discrepancies as there are in the US with Trafalgar and Rasmussen for example.
In 2008 even Rasmussen had a big Obama lead so if Biden was heading for a landslide he should have a big lead of 5 to 10% with them too, he doesn't
The last Rasmussen poll at the end of August had Biden up 49-45 (their previous poll had Biden up 46-45).
The one interesting aspect is Rasmussen poll 2500 Likely voters so their MoE is 2% so theoretically it could be a tie or Biden could be up by 6.
Most of the polling sample sizes are much smaller - the Harris poll has 2834 Registered Voters and they have Biden up 46-40 on a 1.8% MoE.
As a counter, the CNN poll has only 997 Likely Voters and an MoE of 4% so the 8-point Biden lead 51-43 could also be a tie (47-47 like Rasmussen) or it could be 55-39 which would be a Biden landslide.
Again, simply taking headline poll numbers and weaponising them to make a political point is grossly misleading.
What is the sample size, what is the methodology, what is the margin of error?
These are the questions we should be asking of EVERY poll.
And Trump could still eke out an EC win, just, even with Biden 4% ahead in the popular vote. A 5% or higher Biden popular vote lead and Biden should win the EC regardless.
I posted this several times earlier today and yesterday because it's so important IMO.
Though apart from Hillary, who won the PV by 2.1% yet lost, the only other person (in recent times) to win the PV but lose the EV was Al Gore, winning the PV by 0.5%, but losing by a few thousand hanging chads.
Of course there can be a PV/EV disparity, but it is a far from a fixed part of the US electoral system.
The [Biden] campaign expects the contest to tighten in key swing states, acknowledged Jennifer O’Malley Dillon, Mr. Biden’s campaign manager, as she described the team’s efforts to chart multiple paths to the 270 electoral votes needed to win and nodded to growing efforts to expand on-the-ground campaigning.
Different pollster, the pollster with Trump ahead was also the only pollster who had Trump ahead in Michigan in 2016, Trafalgar Group
Nate Silver's site gives a very low rating to the Trafalgar Group.
Well the election 2016 result gives a very high rating to Trafalgar Group as it was the only pollster who correctly had Trump ahead in Michigan and Pennsylvania and without Trump winning those states and Wisconsin Hillary would have won the EC and now be President
We've been through this before, but you do have a tendency to cherry pick.
In 2018, a few days out from the election Trafalgar had McSally (GOP) winning Arizona by 2% (she lost by 2%), Heller (GOP) winning Nevada by 3% (he lost by 5%), Cruz (GOP) winning Texas by 9% (squeaked in by under 2%).
That's cherry picking on my part too, as they had some better ones that night (they called it the right way in Florida, for example, albeit overstating GOP lead). But if you're thinking about the quality of a pollster, you need to look more broadly.
Yes, the pollster with the most extreme effect in the direction of a particular party will sometimes look good by calling wins others don't, but they will also have false alarms. In 2018, you'd have certainly been better listening to other pollsters.
All irrelevant as Trump was not on the ballot in 2018 so many Trump 2016 voters stayed home in 2018.
Not irrelevant at all. A good pollster has a good record in all elections. Your argument is one which tends towards "never mind what my watch says now - it's absolutely perfect twice a day."
No completely irrelevant. You are forgetting the fact that many Trump 2016 voters will not have voted for a Republican candidate for President since Bush Snr in 1988, the last Republican to win Michigan and Pennsylvania, or Reagan in 1984, the last Republican to win Wisconsin (with 1988 effectively Reagan's third term, they switched to Bill Clinton in 1992).
They are in effect white working class Trump Democrats much like the Reagan Democrats of the 1980s, they will only vote for Trump, they will not bother to turn out to vote for other Republicans. Trafalgar were the only pollster to pick them up in 2016
Citation needed. Your are acting like the Dems has a 50 pt lead in those states prior to 2016.
And you still haven't explained why Trump was such a mega draw when the GOP congressional candidates outperformed him in 2016.
No citation needed, Trump was the first Republican to win Wisconsin, Pennsylvania and Michigan since the 1980s. He also got the highest number of votes for any Republican candidate in Pennsylvania since 1988.
GOP congressional candidates performed worse than Trump in 2018, they only did better in 2016 because they got Trump voters voting down ballot for them too after voting for Trump first, those voters did not turn out for them again in 2018 with Trump not on the ballot
In 2018 wasnt the thing that did for the Republicans that the Trumpers did turn out, but that there was a far bigger Democrat turnout than usual?
In 2016 Trump got 62,984,828 votes, in 2018 the Republicans got only 50,861,970 votes, a huge gap of 12 million votes.
By contrast the Democrats got 60,572,245 votes in 2018 so far closer to the 65,853,514 votes Hillary got in 2016 and a small gap of only 5 million votes.
So Trump voters did not turn out in 2018 but Democrats did
Trafalgar got it wrong then.
I would not use Trafalgar for congress votes no, I would use Trafalgar when Trump is on the ballot though for as they showed in 2016 they know where the Trump voters are in the swing states
Does Boris Johnson really want to contest another general election as leader?
I'm almost sure he does. He sees himself as a great statesman, and that's getting closer to being true. As such he can be a little hands-off down the road.
His presumptions are premature though. He's more great whipper-snapper at the moment.
(I know it's not popular but I do in fact think he's doing ok)
Andrew Neil is now doing exactly the thing he was criticising the Co-op for. He's "no platforming" them because the person running their social media account (some poor sod who may well lose their job) was fairly clear they didn't like the Spectator's editorial line.
The only logical position for Neil and the Spectator is "the Co-op can advertise with us if they want or not if they don't - it's entirely up to them but won't alter our editorial position one jot either way".
Does the Co-Op actually advertise in the Spectator?
Yes, that was what started the whole thing off.
A leftist campaigner Tweeted at the Co-Op asking why they were spending money with this evil rag, to which the Co-Op replied with an apology saying they didn’t mean to, it was part of larger ad buy, to which Neil replied no worries, we won’t take your money in future, to which a clearly more senior Co-Op person tried to walk back what the first person had said...
Different pollster, the pollster with Trump ahead was also the only pollster who had Trump ahead in Michigan in 2016, Trafalgar Group
Nate Silver's site gives a very low rating to the Trafalgar Group.
Well the election 2016 result gives a very high rating to Trafalgar Group as it was the only pollster who correctly had Trump ahead in Michigan and Pennsylvania and without Trump winning those states and Wisconsin Hillary would have won the EC and now be President
We've been through this before, but you do have a tendency to cherry pick.
In 2018, a few days out from the election Trafalgar had McSally (GOP) winning Arizona by 2% (she lost by 2%), Heller (GOP) winning Nevada by 3% (he lost by 5%), Cruz (GOP) winning Texas by 9% (squeaked in by under 2%).
That's cherry picking on my part too, as they had some better ones that night (they called it the right way in Florida, for example, albeit overstating GOP lead). But if you're thinking about the quality of a pollster, you need to look more broadly.
Yes, the pollster with the most extreme effect in the direction of a particular party will sometimes look good by calling wins others don't, but they will also have false alarms. In 2018, you'd have certainly been better listening to other pollsters.
All irrelevant as Trump was not on the ballot in 2018 so many Trump 2016 voters stayed home in 2018.
Not irrelevant at all. A good pollster has a good record in all elections. Your argument is one which tends towards "never mind what my watch says now - it's absolutely perfect twice a day."
No completely irrelevant. You are forgetting the fact that many Trump 2016 voters will not have voted for a Republican candidate for President since Bush Snr in 1988, the last Republican to win Michigan and Pennsylvania, or Reagan in 1984, the last Republican to win Wisconsin (with 1988 effectively Reagan's third term, they switched to Bill Clinton in 1992).
They are in effect white working class Trump Democrats much like the Reagan Democrats of the 1980s, they will only vote for Trump, they will not bother to turn out to vote for other Republicans. Trafalgar were the only pollster to pick them up in 2016
Citation needed. Your are acting like the Dems has a 50 pt lead in those states prior to 2016.
And you still haven't explained why Trump was such a mega draw when the GOP congressional candidates outperformed him in 2016.
No citation needed, Trump was the first Republican to win Wisconsin, Pennsylvania and Michigan since the 1980s. He also got the highest number of votes for any Republican candidate in Pennsylvania since 1988.
GOP congressional candidates performed worse than Trump in 2018, they only did better in 2016 because they got Trump voters voting down ballot for them too after voting for Trump first, those voters did not turn out for them again in 2018 with Trump not on the ballot
In 2018 wasnt the thing that did for the Republicans that the Trumpers did turn out, but that there was a far bigger Democrat turnout than usual?
In 2016 Trump got 62,984,828 votes, in 2018 the Republicans got only 50,861,970 votes, a huge gap of 12 million votes.
By contrast the Democrats got 60,572,245 votes in 2018 so far closer to the 65,853,514 votes Hillary got in 2016 and a small gap of only 5 million votes.
So Trump voters did not turn out in 2018 but Democrats did
Trafalgar got it wrong then.
I would not use Trafalgar for congress votes no, I would use Trafalgar when Trump is on the ballot though for as they showed in 2016 they know where the Trump voters are in the swing states
Andrew Neil is now doing exactly the thing he was criticising the Co-op for. He's "no platforming" them because the person running their social media account (some poor sod who may well lose their job) was fairly clear they didn't like the Spectator's editorial line.
The only logical position for Neil and the Spectator is "the Co-op can advertise with us if they want or not if they don't - it's entirely up to them but won't alter our editorial position one jot either way".
No, they aren't advocating other people to stop going there. The co-op and other stupid private companies are signing up to idiotic "anti-hate speech" campaigns which seek to silence opinions they don't agree with. The Speccie is absolutely justified in withdrawing from an advertising relationship with them until the co-op reviews this policy and stops trying to use it's advertising money to influence editorial decisions.
I'm reliably informed that Unilever have had a round of training with all social media teams who have been instructed that Tories buy stuff too and around half of the country is right of centre. The co-op probably needs the same training.
An increasingly desperate Boris Johnson has ordered his staff to step up personal attacks on the Labour leader Keir Starmer and his record as a lawyer, as confidence in the prime minister’s leadership collapses among Tory party members.
The Observer has been told that Johnson was so furious after last Wednesday’s prime minister’s questions – where he was asked to withdraw comments he made about the Labour leader and the IRA by the Speaker, Lindsay Hoyle – that he turned on his staff for leaving him under-prepared, and asked them to come up with more attack lines on the Labour leader’s career as a lawyer.
“He was furious,” said a well-placed source. “He told his team and people at CCHQ [Conservative campaign headquarters] that he wanted them to go after Starmer’s legal record and double down on the attacks on him.”
If there was something significant then BoZo would have used it already.
People as mendacious as BoZo really struggle to comprehend that not everyone else is as bent as they are. Starmer is a bit square, but that is about it.
Particularly as SKS worked alongside and for the previous 2 Tory leaders. Cameron PM and May Home Secretary. Presumably any attack lines would not be a secret from high up Tories. Here's an idea. Get briefed and at least attempt to answer the question. Rather than blurting out the first flowery bollocks that springs to mind. Radical I know.
Andrew Neil is now doing exactly the thing he was criticising the Co-op for. He's "no platforming" them because the person running their social media account (some poor sod who may well lose their job) was fairly clear they didn't like the Spectator's editorial line.
The only logical position for Neil and the Spectator is "the Co-op can advertise with us if they want or not if they don't - it's entirely up to them but won't alter our editorial position one jot either way".
No he's not.
What he's said if you read his Tweets in full is that anyone who attempts to coerce the Spectator to change its Editorial line with threats about advertising gets blacklisted. But they don't then try to get other people to stop using them or taking their adverts, nor do they go on a campaign against them. It's simply saying "we will not be blackmailed, if you attempt to blackmail us into changing our Editorial line we will not work with you again".
Andrew Neil is now doing exactly the thing he was criticising the Co-op for. He's "no platforming" them because the person running their social media account (some poor sod who may well lose their job) was fairly clear they didn't like the Spectator's editorial line.
The only logical position for Neil and the Spectator is "the Co-op can advertise with us if they want or not if they don't - it's entirely up to them but won't alter our editorial position one jot either way".
No, they aren't advocating other people to stop going there. The co-op and other stupid private companies are signing up to idiotic "anti-hate speech" campaigns which seek to silence opinions they don't agree with. The Speccie is absolutely justified in withdrawing from an advertising relationship with them until the co-op reviews this policy and stops trying to use it's advertising money to influence editorial decisions.
I'm reliably informed that Unilever have had a round of training with all social media teams who have been instructed that Tories buy stuff too and around half of the country is right of centre. The co-op probably needs the same training.
Michael Jordan famously said that “Republicans buy sneakers too”, as a way of turning down involvement in a Democrat campaign ad.
Consumer brands trying to do politics doesn’t usually end well.
Andrew Neil is now doing exactly the thing he was criticising the Co-op for. He's "no platforming" them because the person running their social media account (some poor sod who may well lose their job) was fairly clear they didn't like the Spectator's editorial line.
The only logical position for Neil and the Spectator is "the Co-op can advertise with us if they want or not if they don't - it's entirely up to them but won't alter our editorial position one jot either way".
No, they aren't advocating other people to stop going there. The co-op and other stupid private companies are signing up to idiotic "anti-hate speech" campaigns which seek to silence opinions they don't agree with. The Speccie is absolutely justified in withdrawing from an advertising relationship with them until the co-op reviews this policy and stops trying to use it's advertising money to influence editorial decisions.
I'm reliably informed that Unilever have had a round of training with all social media teams who have been instructed that Tories buy stuff too and around half of the country is right of centre. The co-op probably needs the same training.
Michael Jordan famously said that “Republicans buy sneakers too”, as a way of turning down involvement in a Democrat campaign ad.
Consumer brands trying to do politics doesn’t usually end well.
Yes, most of the time you don't do enough for the campaigner types and you still manage to piss everyone else off in the process. It's a no-win situation.
I enjoyed Facebook posting record as revenues while they were being boycotted by woke companies who are now all back on board after seeing just how ineffective advertising off Facebook and Instagram is.
In 2019 every pollster had a big Tory lead, there were no polling discrepancies as there are in the US with Trafalgar and Rasmussen for example.
In 2008 even Rasmussen had a big Obama lead so if Biden was heading for a landslide he should have a big lead of 5 to 10% with them too, he doesn't
The last Rasmussen poll at the end of August had Biden up 49-45 (their previous poll had Biden up 46-45).
The one interesting aspect is Rasmussen poll 2500 Likely voters so their MoE is 2% so theoretically it could be a tie or Biden could be up by 6.
Most of the polling sample sizes are much smaller - the Harris poll has 2834 Registered Voters and they have Biden up 46-40 on a 1.8% MoE.
As a counter, the CNN poll has only 997 Likely Voters and an MoE of 4% so the 8-point Biden lead 51-43 could also be a tie (47-47 like Rasmussen) or it could be 55-39 which would be a Biden landslide.
Again, simply taking headline poll numbers and weaponising them to make a political point is grossly misleading.
What is the sample size, what is the methodology, what is the margin of error?
These are the questions we should be asking of EVERY poll.
And Trump could still eke out an EC win, just, even with Biden 4% ahead in the popular vote. A 5% or higher Biden popular vote lead and Biden should win the EC regardless.
I posted this several times earlier today and yesterday because it's so important IMO.
Though apart from Hillary, who won the PV by 2.1% yet lost, the only other person (in recent times) to win the PV but lose the EV was Al Gore, winning the PV by 0.5%, but losing by a few thousand hanging chads.
Of course there can be a PV/EV disparity, but it is a far from a fixed part of the US electoral system.
Different pollster, the pollster with Trump ahead was also the only pollster who had Trump ahead in Michigan in 2016, Trafalgar Group
Nate Silver's site gives a very low rating to the Trafalgar Group.
Well the election 2016 result gives a very high rating to Trafalgar Group as it was the only pollster who correctly had Trump ahead in Michigan and Pennsylvania and without Trump winning those states and Wisconsin Hillary would have won the EC and now be President
We've been through this before, but you do have a tendency to cherry pick.
In 2018, a few days out from the election Trafalgar had McSally (GOP) winning Arizona by 2% (she lost by 2%), Heller (GOP) winning Nevada by 3% (he lost by 5%), Cruz (GOP) winning Texas by 9% (squeaked in by under 2%).
That's cherry picking on my part too, as they had some better ones that night (they called it the right way in Florida, for example, albeit overstating GOP lead). But if you're thinking about the quality of a pollster, you need to look more broadly.
Yes, the pollster with the most extreme effect in the direction of a particular party will sometimes look good by calling wins others don't, but they will also have false alarms. In 2018, you'd have certainly been better listening to other pollsters.
All irrelevant as Trump was not on the ballot in 2018 so many Trump 2016 voters stayed home in 2018.
Not irrelevant at all. A good pollster has a good record in all elections. Your argument is one which tends towards "never mind what my watch says now - it's absolutely perfect twice a day."
No completely irrelevant.
You are forgetting the fact that many Trump 2016 voters will not have voted for a Republican candidate for President since Bush Snr in 1988, the last Republican to win Michigan and Pennsylvania, or Reagan in 1984, the last Republican to win Wisconsin (with 1988 effectively Reagan's third term, they switched to Bill Clinton in 1992).
They are in effect white working class Trump Democrats much like the Reagan Democrats of the 1980s, they will only vote for Trump, they will not bother to turn out to vote for other Republicans. Trafalgar were the only pollster to pick them up in 2016
Humphrey failed to carry Wisconsin against Nixon in 1968 despite coming from the neighbouring state of Minnesota.
Different pollster, the pollster with Trump ahead was also the only pollster who had Trump ahead in Michigan in 2016, Trafalgar Group
Nate Silver's site gives a very low rating to the Trafalgar Group.
Well the election 2016 result gives a very high rating to Trafalgar Group as it was the only pollster who correctly had Trump ahead in Michigan and Pennsylvania and without Trump winning those states and Wisconsin Hillary would have won the EC and now be President
We've been through this before, but you do have a tendency to cherry pick.
In 2018, a few days out from the election Trafalgar had McSally (GOP) winning Arizona by 2% (she lost by 2%), Heller (GOP) winning Nevada by 3% (he lost by 5%), Cruz (GOP) winning Texas by 9% (squeaked in by under 2%).
That's cherry picking on my part too, as they had some better ones that night (they called it the right way in Florida, for example, albeit overstating GOP lead). But if you're thinking about the quality of a pollster, you need to look more broadly.
Yes, the pollster with the most extreme effect in the direction of a particular party will sometimes look good by calling wins others don't, but they will also have false alarms. In 2018, you'd have certainly been better listening to other pollsters.
All irrelevant as Trump was not on the ballot in 2018 so many Trump 2016 voters stayed home in 2018.
Not irrelevant at all. A good pollster has a good record in all elections. Your argument is one which tends towards "never mind what my watch says now - it's absolutely perfect twice a day."
No completely irrelevant.
You are forgetting the fact that many Trump 2016 voters will not have voted for a Republican candidate for President since Bush Snr in 1988, the last Republican to win Michigan and Pennsylvania, or Reagan in 1984, the last Republican to win Wisconsin (with 1988 effectively Reagan's third term, they switched to Bill Clinton in 1992).
They are in effect white working class Trump Democrats much like the Reagan Democrats of the 1980s, they will only vote for Trump, they will not bother to turn out to vote for other Republicans. Trafalgar were the only pollster to pick them up in 2016
Humphrey failed to carry Wisconsin against Nixon in 1968 despite coming from the neighbouring state of Minnesota.
Since WW2 yes Nixon, Eisenhower, Reagan and Trump are the only Republican presidential candidates to have won Wisconsin (Nixon won it in 1960 too)
Speaking as a liberal conservative - no, there's a lot of us around.
Though given Parris seems to define liberal conservativism as being pro-EU for some reason, I doubt he'd recognise me as a liberal conservative. Since when liberal meant pro-EU I don't know, but liberal conservativism shouldn't be about the EU.
Speaking as a liberal conservative - no, there's a lot of us around.
Though given Parris seems to define liberal conservativism as being pro-EU for some reason, I doubt he'd recognise me as a liberal conservative. Since when liberal meant pro-EU I don't know, but liberal conservativism shouldn't be about the EU.
All these labels are a little meaningless, for two main reasons:
1) there is no real consensus on what is liberal and what is not, what is conservative and not, etc. 2) whatever the definitions, many people will be liberal on some issues, authoritarian on others, conservative on some, progressive on others, etc. etc.
The more interesting question is whether Remainer Conservatives - capital C - are now history. Probably not, but currently lying low.
If Biden can maintain his advantage among Stein and Johnson voters in Michigan, Pennsylvania, and Wisconsin, then those states will return to the Democratic column in 2020 and Biden will be elected president.
Different pollster, the pollster with Trump ahead was also the only pollster who had Trump ahead in Michigan in 2016, Trafalgar Group
Nate Silver's site gives a very low rating to the Trafalgar Group.
Well the election 2016 result gives a very high rating to Trafalgar Group as it was the only pollster who correctly had Trump ahead in Michigan and Pennsylvania and without Trump winning those states and Wisconsin Hillary would have won the EC and now be President
We've been through this before, but you do have a tendency to cherry pick.
In 2018, a few days out from the election Trafalgar had McSally (GOP) winning Arizona by 2% (she lost by 2%), Heller (GOP) winning Nevada by 3% (he lost by 5%), Cruz (GOP) winning Texas by 9% (squeaked in by under 2%).
That's cherry picking on my part too, as they had some better ones that night (they called it the right way in Florida, for example, albeit overstating GOP lead). But if you're thinking about the quality of a pollster, you need to look more broadly.
Yes, the pollster with the most extreme effect in the direction of a particular party will sometimes look good by calling wins others don't, but they will also have false alarms. In 2018, you'd have certainly been better listening to other pollsters.
All irrelevant as Trump was not on the ballot in 2018 so many Trump 2016 voters stayed home in 2018.
Not irrelevant at all. A good pollster has a good record in all elections. Your argument is one which tends towards "never mind what my watch says now - it's absolutely perfect twice a day."
No completely irrelevant.
You are forgetting the fact that many Trump 2016 voters will not have voted for a Republican candidate for President since Bush Snr in 1988, the last Republican to win Michigan and Pennsylvania, or Reagan in 1984, the last Republican to win Wisconsin (with 1988 effectively Reagan's third term, they switched to Bill Clinton in 1992).
They are in effect white working class Trump Democrats much like the Reagan Democrats of the 1980s, they will only vote for Trump, they will not bother to turn out to vote for other Republicans. Trafalgar were the only pollster to pick them up in 2016
Humphrey failed to carry Wisconsin against Nixon in 1968 despite coming from the neighbouring state of Minnesota.
Since WW2 yes Nixon, Eisenhower, Reagan and Trump are the only Republican presidential candidates to have won Wisconsin (Nixon won it in 1960 too)
Indeed - so for much of the post-World War 2 period it has really been a swing state. Only since 1988 has Wisconsin normally been in the Democrat column.
The state, the percentage of the vote Clinton got, the percentage of the vote Biden gets, Trump gets and how much more or less Biden gets than Clinton.
This leaves Biden taking the Clinton states minus Minnesota despite polling nationally 50.4% vs 43% against Trump.
It is an exercise best left to the reader to spot the flaws in this analysis (hint, it is the third parties)
In 2019 every pollster had a big Tory lead, there were no polling discrepancies as there are in the US with Trafalgar and Rasmussen for example.
In 2008 even Rasmussen had a big Obama lead so if Biden was heading for a landslide he should have a big lead of 5 to 10% with them too, he doesn't
The last Rasmussen poll at the end of August had Biden up 49-45 (their previous poll had Biden up 46-45).
The one interesting aspect is Rasmussen poll 2500 Likely voters so their MoE is 2% so theoretically it could be a tie or Biden could be up by 6.
Most of the polling sample sizes are much smaller - the Harris poll has 2834 Registered Voters and they have Biden up 46-40 on a 1.8% MoE.
As a counter, the CNN poll has only 997 Likely Voters and an MoE of 4% so the 8-point Biden lead 51-43 could also be a tie (47-47 like Rasmussen) or it could be 55-39 which would be a Biden landslide.
Again, simply taking headline poll numbers and weaponising them to make a political point is grossly misleading.
What is the sample size, what is the methodology, what is the margin of error?
These are the questions we should be asking of EVERY poll.
And Trump could still eke out an EC win, just, even with Biden 4% ahead in the popular vote. A 5% or higher Biden popular vote lead and Biden should win the EC regardless.
I posted this several times earlier today and yesterday because it's so important IMO.
Though apart from Hillary, who won the PV by 2.1% yet lost, the only other person (in recent times) to win the PV but lose the EV was Al Gore, winning the PV by 0.5%, but losing by a few thousand hanging chads.
Of course there can be a PV/EV disparity, but it is a far from a fixed part of the US electoral system.
In 2019 every pollster had a big Tory lead, there were no polling discrepancies as there are in the US with Trafalgar and Rasmussen for example.
In 2008 even Rasmussen had a big Obama lead so if Biden was heading for a landslide he should have a big lead of 5 to 10% with them too, he doesn't
The last Rasmussen poll at the end of August had Biden up 49-45 (their previous poll had Biden up 46-45).
The one interesting aspect is Rasmussen poll 2500 Likely voters so their MoE is 2% so theoretically it could be a tie or Biden could be up by 6.
Most of the polling sample sizes are much smaller - the Harris poll has 2834 Registered Voters and they have Biden up 46-40 on a 1.8% MoE.
As a counter, the CNN poll has only 997 Likely Voters and an MoE of 4% so the 8-point Biden lead 51-43 could also be a tie (47-47 like Rasmussen) or it could be 55-39 which would be a Biden landslide.
Again, simply taking headline poll numbers and weaponising them to make a political point is grossly misleading.
What is the sample size, what is the methodology, what is the margin of error?
These are the questions we should be asking of EVERY poll.
And Trump could still eke out an EC win, just, even with Biden 4% ahead in the popular vote. A 5% or higher Biden popular vote lead and Biden should win the EC regardless.
I posted this several times earlier today and yesterday because it's so important IMO.
Though apart from Hillary, who won the PV by 2.1% yet lost, the only other person (in recent times) to win the PV but lose the EV was Al Gore, winning the PV by 0.5%, but losing by a few thousand hanging chads.
Of course there can be a PV/EV disparity, but it is a far from a fixed part of the US electoral system.
So where does PB think we are tonight with the Trump military story?
Yesterday, the collective view seemed to be this would sink his candidacy.
Today, it doesn’t feel like it has gained traction.
I’m guessing the collective answer will be too early to tell...
Barring a bigger surprise, Trump is sunk.
A bigger surprise could yet come out, but it would be a shocker. I'd say Biden has a 90% chance of victory, Trump 10% - and that's possibly generous to Trump.
So where does PB think we are tonight with the Trump military story?
Yesterday, the collective view seemed to be this would sink his candidacy.
Today, it doesn’t feel like it has gained traction.
I’m guessing the collective answer will be too early to tell...
Barring a bigger surprise, Trump is sunk.
A bigger surprise could yet come out, but it would be a shocker. I'd say Biden has a 90% chance of victory, Trump 10% - and that's possibly generous to Trump.
You are far more optimistic than me!!
I worry the polls aren't picking up silent Trumpers.
So where does PB think we are tonight with the Trump military story?
Yesterday, the collective view seemed to be this would sink his candidacy.
Today, it doesn’t feel like it has gained traction.
I’m guessing the collective answer will be too early to tell...
It won't harm Trump. His base will either just dismiss it or - as we've seen on previous occasions - perform mental gymnastics in order to conclude that he might have a point after all. Remember how Russia was public enemy number one amongst Republicans until it transpired Trump might have conspired with them? Suddenly it became the great and benign culture that brought us Tchaikovsky.
So where does PB think we are tonight with the Trump military story?
Yesterday, the collective view seemed to be this would sink his candidacy.
Today, it doesn’t feel like it has gained traction.
I’m guessing the collective answer will be too early to tell...
The story has made Trump's job a tiny bit harder. The world is rarely full of step functions, and this isn't one. But the path he has to climb - and he has only 58 days left - has got a little steeper.
Let's assume he wins the electoral college if he's 3 points behind or less. And let's assume the the polls overstate Biden's lead by two points.
That means Trump needs to take two points off biden in the next 58 days. Possible, absolutely. But far from certain. And this also relies on the - possibly fallacious assumption - that Trump is understated in the polls.
Speaking as a liberal conservative - no, there's a lot of us around.
Though given Parris seems to define liberal conservativism as being pro-EU for some reason, I doubt he'd recognise me as a liberal conservative. Since when liberal meant pro-EU I don't know, but liberal conservativism shouldn't be about the EU.
I wouldn't define it in those terms either. There was a clear convergence from 2008-10 between the Conservatives under Cameron and the Liberal Democrats under Clegg. Whether that was entirely a philosophical convergence I'm uncertain though the Orange Bookers and the Cameroons clearly had some things in common.
What drove both was the response to a decade or more of Labour Government and the Global Financial Crash. From the former came a mood of decentralisation and a loosening of the dead hand of Whitehall and with the latter came the presumption the State had become too large and needed to be reduced as part of the mechanism for restoring public finances.
Beyond that, there were areas such as pension reform and environmental policies where there was a degree of commonality of approach. Whether that presaged the Coalition is moot now but at the time it made something which had seemed unlikely possible.
The Atlantic reported first, and the Associated Press, Washington Post, and even Fox News have confirmed the story. The smart thing is to let Trump deny it, let Trump's team back him up, and then get people to go on record. I don't know if people will, as Trump will retaliate and they will get death threats. I'm certain the story is broadly correct whatever Trump says, as it ties up with a pattern of behaviour and statements that have been reported before, particularly things that Rex Tillerson dealt with.
So where does PB think we are tonight with the Trump military story?
Yesterday, the collective view seemed to be this would sink his candidacy.
Today, it doesn’t feel like it has gained traction.
I’m guessing the collective answer will be too early to tell...
It won't harm Trump. His base will either just dismiss it or - as we've seen on previous occasions - perform mental gymnastics in order to conclude that he might have a point after all. Remember how Russia was public enemy number one amongst Republicans until it transpired Trump might have conspired with them? Suddenly it became the great and benign culture that brought us Tchaikovsky.
Like GOP view of US debt. Terrible thing, tantamount to the work of the devil when Obama was POTUS.
Now, it couldn't be added to fast enough for these folk.
In 2019 every pollster had a big Tory lead, there were no polling discrepancies as there are in the US with Trafalgar and Rasmussen for example.
In 2008 even Rasmussen had a big Obama lead so if Biden was heading for a landslide he should have a big lead of 5 to 10% with them too, he doesn't
The last Rasmussen poll at the end of August had Biden up 49-45 (their previous poll had Biden up 46-45).
The one interesting aspect is Rasmussen poll 2500 Likely voters so their MoE is 2% so theoretically it could be a tie or Biden could be up by 6.
Most of the polling sample sizes are much smaller - the Harris poll has 2834 Registered Voters and they have Biden up 46-40 on a 1.8% MoE.
As a counter, the CNN poll has only 997 Likely Voters and an MoE of 4% so the 8-point Biden lead 51-43 could also be a tie (47-47 like Rasmussen) or it could be 55-39 which would be a Biden landslide.
Again, simply taking headline poll numbers and weaponising them to make a political point is grossly misleading.
What is the sample size, what is the methodology, what is the margin of error?
These are the questions we should be asking of EVERY poll.
And Trump could still eke out an EC win, just, even with Biden 4% ahead in the popular vote. A 5% or higher Biden popular vote lead and Biden should win the EC regardless.
I posted this several times earlier today and yesterday because it's so important IMO.
Though apart from Hillary, who won the PV by 2.1% yet lost, the only other person (in recent times) to win the PV but lose the EV was Al Gore, winning the PV by 0.5%, but losing by a few thousand hanging chads.
Of course there can be a PV/EV disparity, but it is a far from a fixed part of the US electoral system.
Though its worth remembering that in many respects the Republicans of the 19th century would now be Democrats and vice-versa.
Hayes and Harrison would likely now be Democrats, Tilden and Cleveland would likely now be Republicans.
Indeed, in 1888 Cleveland the Democrat candidate and popular vote winner won every southern state, while Harrison the Republican candidate and winner of the electoral college won most Northeastern and Midwest states and every state in the West at that time.
So where does PB think we are tonight with the Trump military story?
Yesterday, the collective view seemed to be this would sink his candidacy.
Today, it doesn’t feel like it has gained traction.
I’m guessing the collective answer will be too early to tell...
Barring a bigger surprise, Trump is sunk.
A bigger surprise could yet come out, but it would be a shocker. I'd say Biden has a 90% chance of victory, Trump 10% - and that's possibly generous to Trump.
You are far more optimistic than me!!
I worry the polls aren't picking up silent Trumpers.
The remarkable thing is how most US presidential elections seem to end up being reasonably close in terms of vote share.
I'm certain the story is broadly correct whatever Trump says, as it ties up with a pattern of behaviour and statements that have been reported before, particularly things that Rex Tillerson dealt with.
So where does PB think we are tonight with the Trump military story?
Yesterday, the collective view seemed to be this would sink his candidacy.
Today, it doesn’t feel like it has gained traction.
I’m guessing the collective answer will be too early to tell...
I am not sure who thought it would sink his candidacy (I skipped PB yesterday) but if Trump's incessant lying, his promotion of conspiracy theories, his woeful mismanagement of Covid-19, his promotion of untested treatments, his racial views, his comments about women, his incitement of violence, his support for white supremacists, his kow-towing to Putin, his track-record of employing criminal has not sunk his candidacy, I doubt if this latest revelation will make much difference.
So where does PB think we are tonight with the Trump military story?
Yesterday, the collective view seemed to be this would sink his candidacy.
Today, it doesn’t feel like it has gained traction.
I’m guessing the collective answer will be too early to tell...
Barring a bigger surprise, Trump is sunk.
A bigger surprise could yet come out, but it would be a shocker. I'd say Biden has a 90% chance of victory, Trump 10% - and that's possibly generous to Trump.
You are far more optimistic than me!!
I worry the polls aren't picking up silent Trumpers.
So far as I can see, Trumpers are far from silent.
So where does PB think we are tonight with the Trump military story?
Yesterday, the collective view seemed to be this would sink his candidacy.
Today, it doesn’t feel like it has gained traction.
I’m guessing the collective answer will be too early to tell...
Barring a bigger surprise, Trump is sunk.
A bigger surprise could yet come out, but it would be a shocker. I'd say Biden has a 90% chance of victory, Trump 10% - and that's possibly generous to Trump.
You are far more optimistic than me!!
I worry the polls aren't picking up silent Trumpers.
The remarkable thing is how most US presidential elections seem to end up being reasonably close in terms of vote share.
in this case we will find around 50% of the voters don't want to live in a democracy anymore.
In 2019 every pollster had a big Tory lead, there were no polling discrepancies as there are in the US with Trafalgar and Rasmussen for example.
In 2008 even Rasmussen had a big Obama lead so if Biden was heading for a landslide he should have a big lead of 5 to 10% with them too, he doesn't
The last Rasmussen poll at the end of August had Biden up 49-45 (their previous poll had Biden up 46-45).
The one interesting aspect is Rasmussen poll 2500 Likely voters so their MoE is 2% so theoretically it could be a tie or Biden could be up by 6.
Most of the polling sample sizes are much smaller - the Harris poll has 2834 Registered Voters and they have Biden up 46-40 on a 1.8% MoE.
As a counter, the CNN poll has only 997 Likely Voters and an MoE of 4% so the 8-point Biden lead 51-43 could also be a tie (47-47 like Rasmussen) or it could be 55-39 which would be a Biden landslide.
Again, simply taking headline poll numbers and weaponising them to make a political point is grossly misleading.
What is the sample size, what is the methodology, what is the margin of error?
These are the questions we should be asking of EVERY poll.
And Trump could still eke out an EC win, just, even with Biden 4% ahead in the popular vote. A 5% or higher Biden popular vote lead and Biden should win the EC regardless.
So. For this to race be 50-50, according to these numbers, requires a little bit more than a 2.5% lead for Biden. (As 2-3% is a 47% chance of a Biden win.)
I have said repeatedly, that I think Biden is a 65-70% chance (although probably closer to 70% now), with the value bet being on him winning by a comfortable margin.
Those numbers from Nate bear me out. A 65-70% chance of winning is a little bit more than a 3% lead. Trump currently trails by 7 points, and that gap is very stable.
As we get closer and closer to election day, then if the polls don't move (and so far they haven't moved), then Trump's only chance is that the polls aren't wrong like 2016 (where they were 1% out), but that they'e 3-4 points out.
Now that's possible. But I remember in 2012, a poster called Stuart Truth who would rip up every pollster, with some excellent detailed analysis, showing that they all depended on excessive black turnout. And that Romney would win.
Mr Truth made his final appearance on the site the day before polling day, and was never seen again.
Banking on pollsters - in aggregate, and at the national level - being more than a percent or two out seems like a mugs game. And assuming it's definitely going to go in one direction (Obama outperformed the polls in 2012) is a double mugs game.
So where does PB think we are tonight with the Trump military story?
Yesterday, the collective view seemed to be this would sink his candidacy.
Today, it doesn’t feel like it has gained traction.
I’m guessing the collective answer will be too early to tell...
Barring a bigger surprise, Trump is sunk.
A bigger surprise could yet come out, but it would be a shocker. I'd say Biden has a 90% chance of victory, Trump 10% - and that's possibly generous to Trump.
You are far more optimistic than me!!
I worry the polls aren't picking up silent Trumpers.
The remarkable thing is how most US presidential elections seem to end up being reasonably close in terms of vote share.
Since WW2 that is not really true, the 1952, 1956, 1964, 1972, 1980, 1984, 1988, 1992, 1996 and 2008 elections were all won by 5% or more.
The 1952, 1956, 1964, 1972, 1984 elections were even won by 10% or more.
It is only since 2000 the elections have been close as the parties have become so polarised other than 2008 when Obama won 9% of Republicans, 20% of conservatives and 52% of Independents.
In 2019 every pollster had a big Tory lead, there were no polling discrepancies as there are in the US with Trafalgar and Rasmussen for example.
In 2008 even Rasmussen had a big Obama lead so if Biden was heading for a landslide he should have a big lead of 5 to 10% with them too, he doesn't
The last Rasmussen poll at the end of August had Biden up 49-45 (their previous poll had Biden up 46-45).
The one interesting aspect is Rasmussen poll 2500 Likely voters so their MoE is 2% so theoretically it could be a tie or Biden could be up by 6.
Most of the polling sample sizes are much smaller - the Harris poll has 2834 Registered Voters and they have Biden up 46-40 on a 1.8% MoE.
As a counter, the CNN poll has only 997 Likely Voters and an MoE of 4% so the 8-point Biden lead 51-43 could also be a tie (47-47 like Rasmussen) or it could be 55-39 which would be a Biden landslide.
Again, simply taking headline poll numbers and weaponising them to make a political point is grossly misleading.
What is the sample size, what is the methodology, what is the margin of error?
These are the questions we should be asking of EVERY poll.
And Trump could still eke out an EC win, just, even with Biden 4% ahead in the popular vote. A 5% or higher Biden popular vote lead and Biden should win the EC regardless.
So. For this to race be 50-50, according to these numbers, requires a little bit more than a 2.5% lead for Biden. (As 2-3% is a 47% chance of a Biden win.)
I have said repeatedly, that I think Biden is a 65-70% chance (although probably closer to 70% now), with the value bet being on him winning by a comfortable margin.
Those numbers from Nate bear me out. A 65-70% chance of winning is a little bit more than a 3% lead. Trump currently trails by 7 points, and that gap is very stable.
As we get closer and closer to election day, then if the polls don't move (and so far they haven't moved), then Trump's only chance is that the polls aren't wrong like 2016 (where they were 1% out), but that they'e 3-4 points out.
Now that's possible. But I remember in 2012, a poster called Stuart Truth who would rip up every pollster, with some excellent detailed analysis, showing that they all depended on excessive black turnout. And that Romney would win.
Mr Truth made his final appearance on the site the day before polling day, and was never seen again.
Banking on pollsters - in aggregate, and at the national level - being more than a percent or two out seems like a mugs game. And assuming it's definitely going to go in one direction (Obama outperformed the polls in 2012) is a double mugs game.
Haha "Mr Truth made his final appearance on the site the day before polling day, and was never seen again." sounds like it would habe been a more appropriate parable for the 2016 election.
Trump is a gobshite, but people do make up crap about him all the time as well.
That's likely true, and it is also possiby the case that because he says such batshit things we are inclined to believe batshit stories about him which are not in fact true, the way we sometimes swallow tales of outlandish cruelty about the North Korean regime which did not happen, or at least not provably happen.
But the one person who cannot complain that others are exagerrating the facts or even making stuff up, is Donald Trump himself. That wouldn't make a lie about him ok, if that happened, but there's any number of non-lies to rely upon, and his own demonstrable behaviour lends credence to other tales and he can have no complaints about that.
Comments
Who were the "independent panel of PB Moderates"? I'm assuming you're not counting people who voted Labour, who went for such a non-moderate 2015 manifesto?
More people voted for Trump than Romney in Iowa, Ohio, Michigan and Pennsylvania and Florida ie all the Obama states in 2012 that Trump won in 2016, Wisconsin was the only exception to that, though Trump still got a higher voteshare than Romney did there.
In 2008 even Rasmussen had a big Obama lead so if Biden was heading for a landslide he should have a big lead of 5 to 10% with them too, he doesn't
It should also be pointed out they haven't released polls in some battleground states since June/July (or at all) but your are happy to say they predict a EC win for trump even though we are in September.
They might be right, but their track record in 2018 and being the outliers against pollsters who did better in 2018, I would not be putting any money down purely based on their polls.
By contrast the Democrats got 60,572,245 votes in 2018 so far closer to the 65,853,514 votes Hillary got in 2016 and a small gap of only 5 million votes.
So Trump voters did not turn out in 2018 but Democrats did
The one interesting aspect is Rasmussen poll 2500 Likely voters so their MoE is 2% so theoretically it could be a tie or Biden could be up by 6.
Most of the polling sample sizes are much smaller - the Harris poll has 2834 Registered Voters and they have Biden up 46-40 on a 1.8% MoE.
As a counter, the CNN poll has only 997 Likely Voters and an MoE of 4% so the 8-point Biden lead 51-43 could also be a tie (47-47 like Rasmussen) or it could be 55-39 which would be a Biden landslide.
Again, simply taking headline poll numbers and weaponising them to make a political point is grossly misleading.
What is the sample size, what is the methodology, what is the margin of error?
These are the questions we should be asking of EVERY poll.
Indeed, he may well not have served any time at all in Sweden. That case has now been dropped - albeit Swedish investigators claim due to the passage of time making memories less reliable rather than a fundamental problem with their case. So it's questionable if there would have been a conviction.
https://twitter.com/afneil/status/1302323547222740992?s=21
A tape saying all veterans are losers and cowards wouldn't cut it. Not even close.
Compared to UK polling, US polling is all over the place. There’s often small sample sizes, subsamples used for headlines, no common rules on publishing data, no record of the full set of questions nor their order, very partisan commissioners of polling etc etc.
https://twitter.com/NateSilver538/status/1301190941110341632?s=20
https://twitter.com/NateSilver538/status/1301191558734188545?s=20
The Observer has been told that Johnson was so furious after last Wednesday’s prime minister’s questions – where he was asked to withdraw comments he made about the Labour leader and the IRA by the Speaker, Lindsay Hoyle – that he turned on his staff for leaving him under-prepared, and asked them to come up with more attack lines on the Labour leader’s career as a lawyer.
“He was furious,” said a well-placed source. “He told his team and people at CCHQ [Conservative campaign headquarters] that he wanted them to go after Starmer’s legal record and double down on the attacks on him.”
https://www.theguardian.com/politics/2020/sep/05/desperate-boris-johnson-to-step-up-personal-attacks-on-keir-starmer
The Tory party have really bought themselves a poundshop PM.
The only logical position for Neil and the Spectator is "the Co-op can advertise with us if they want or not if they don't - it's entirely up to them but won't alter our editorial position one jot either way".
He remains stuck at 44/45 percent of the vote.
https://cookpolitical.com/analysis/national/national-politics/two-months-go-steady-presidential-race
People as mendacious as BoZo really struggle to comprehend that not everyone else is as bent as they are. Starmer is a bit square, but that is about it.
Of course there can be a PV/EV disparity, but it is a far from a fixed part of the US electoral system.
NYTimes
Simon Rattle, CBSO, 100th anniversary concert.
https://www.youtube.com/watch?v=emZua--3eIs
His presumptions are premature though. He's more great whipper-snapper at the moment.
(I know it's not popular but I do in fact think he's doing ok)
A leftist campaigner Tweeted at the Co-Op asking why they were spending money with this evil rag, to which the Co-Op replied with an apology saying they didn’t mean to, it was part of larger ad buy, to which Neil replied no worries, we won’t take your money in future, to which a clearly more senior Co-Op person tried to walk back what the first person had said...
https://www.thetimes.co.uk/edition/news/extinction-rebellion-protests-under-threat-from-parasitic-hard-left-groups-9mgphfv6b
I'm reliably informed that Unilever have had a round of training with all social media teams who have been instructed that Tories buy stuff too and around half of the country is right of centre. The co-op probably needs the same training.
Presumably any attack lines would not be a secret from high up Tories.
Here's an idea.
Get briefed and at least attempt to answer the question. Rather than blurting out the first flowery bollocks that springs to mind.
Radical I know.
What he's said if you read his Tweets in full is that anyone who attempts to coerce the Spectator to change its Editorial line with threats about advertising gets blacklisted. But they don't then try to get other people to stop using them or taking their adverts, nor do they go on a campaign against them. It's simply saying "we will not be blackmailed, if you attempt to blackmail us into changing our Editorial line we will not work with you again".
That's very different to Cancel Culture.
Good look, two months our from the election.
Consumer brands trying to do politics doesn’t usually end well.
Are liberal conservatives now history?"
https://www.spectator.co.uk/article/are-liberal-conservatives-now-history-
I enjoyed Facebook posting record as revenues while they were being boycotted by woke companies who are now all back on board after seeing just how ineffective advertising off Facebook and Instagram is.
https://en.wikipedia.org/wiki/List_of_United_States_presidential_elections_in_which_the_winner_lost_the_popular_vote
"will do anything to win"...
And Dom is too busy building UK-DARPA in a smelly back bedroom with a really big TV screen.
Though given Parris seems to define liberal conservativism as being pro-EU for some reason, I doubt he'd recognise me as a liberal conservative. Since when liberal meant pro-EU I don't know, but liberal conservativism shouldn't be about the EU.
1) there is no real consensus on what is liberal and what is not, what is conservative and not, etc.
2) whatever the definitions, many people will be liberal on some issues, authoritarian on others, conservative on some, progressive on others, etc. etc.
The more interesting question is whether Remainer Conservatives - capital C - are now history. Probably not, but currently lying low.
https://thehill.com/policy/healthcare/public-global-health/515273-covid-19-cases-will-surge-in-the-fall-peak-after
If Biden can maintain his advantage among Stein and Johnson voters in Michigan, Pennsylvania, and Wisconsin, then those states will return to the Democratic column in 2020 and Biden will be elected president.
The Hill
The state, the percentage of the vote Clinton got, the percentage of the vote Biden gets, Trump gets and how much more or less Biden gets than Clinton.
This leaves Biden taking the Clinton states minus Minnesota despite polling nationally 50.4% vs 43% against Trump.
It is an exercise best left to the reader to spot the flaws in this analysis (hint, it is the third parties)
Yesterday, the collective view seemed to be this would sink his candidacy.
Today, it doesn’t feel like it has gained traction.
I’m guessing the collective answer will be too early to tell...
Hayes and Harrison would likely now be Democrats, Tilden and Cleveland would likely now be Republicans.
A bigger surprise could yet come out, but it would be a shocker. I'd say Biden has a 90% chance of victory, Trump 10% - and that's possibly generous to Trump.
https://www.politico.com/news/2020/09/04/biden-rips-trump-fallen-troops-409082
I worry the polls aren't picking up silent Trumpers.
He really could finally sink Trump.
Let's assume he wins the electoral college if he's 3 points behind or less. And let's assume the the polls overstate Biden's lead by two points.
That means Trump needs to take two points off biden in the next 58 days. Possible, absolutely. But far from certain. And this also relies on the - possibly fallacious assumption - that Trump is understated in the polls.
What drove both was the response to a decade or more of Labour Government and the Global Financial Crash. From the former came a mood of decentralisation and a loosening of the dead hand of Whitehall and with the latter came the presumption the State had become too large and needed to be reduced as part of the mechanism for restoring public finances.
Beyond that, there were areas such as pension reform and environmental policies where there was a degree of commonality of approach. Whether that presaged the Coalition is moot now but at the time it made something which had seemed unlikely possible.
He's far too professional to get involved in an election. Even if Trump is almost goading him to do so.
Now, it couldn't be added to fast enough for these folk.
https://en.wikipedia.org/wiki/1888_United_States_presidential_election
A similar trend was seen in 1876 though the GOP candidate Hayes won 3 southern states and the Democratic candidate Tilden won New York
https://en.wikipedia.org/wiki/1876_United_States_presidential_election
https://twitter.com/ianbremmer/status/1302340391124062212?s=20
Who am I going to believe, the world's biggest liar or the editor of The Atlantic?
I have said repeatedly, that I think Biden is a 65-70% chance (although probably closer to 70% now), with the value bet being on him winning by a comfortable margin.
Those numbers from Nate bear me out. A 65-70% chance of winning is a little bit more than a 3% lead. Trump currently trails by 7 points, and that gap is very stable.
As we get closer and closer to election day, then if the polls don't move (and so far they haven't moved), then Trump's only chance is that the polls aren't wrong like 2016 (where they were 1% out), but that they'e 3-4 points out.
Now that's possible. But I remember in 2012, a poster called Stuart Truth who would rip up every pollster, with some excellent detailed analysis, showing that they all depended on excessive black turnout. And that Romney would win.
Mr Truth made his final appearance on the site the day before polling day, and was never seen again.
Banking on pollsters - in aggregate, and at the national level - being more than a percent or two out seems like a mugs game. And assuming it's definitely going to go in one direction (Obama outperformed the polls in 2012) is a double mugs game.
The 1952, 1956, 1964, 1972, 1984 elections were even won by 10% or more.
It is only since 2000 the elections have been close as the parties have become so polarised other than 2008 when Obama won 9% of Republicans, 20% of conservatives and 52% of Independents.
But the one person who cannot complain that others are exagerrating the facts or even making stuff up, is Donald Trump himself. That wouldn't make a lie about him ok, if that happened, but there's any number of non-lies to rely upon, and his own demonstrable behaviour lends credence to other tales and he can have no complaints about that.