Macron is right. Enemies of western liberal democracy are our enemies.
Macron would get my vote, if I were French.
He's showing exactly the sort of steel I wish all Western leaders would.
He is standing up against extremism and for free speech yes but I think Trudeau also had a point there is no point using free speech as an excuse to act irresponsibly, we know the reaction publishing cartoons of the Prophet Muhammed will have in the Muslim world as those mass anti France demos in Bangladesh show, so why do it just to get that reaction?
I think it's very offensive to Muslims who view God as too great to be drawn and the prophet as sacred. Plus, I think many anti-religious types are rude pompous and unfunny liberals who just mock other people's sincerely held beliefs but couldn't tell a good joke in a hundred years.
But, that's not the point. Once people start getting killed for saying something extremely offensive then you have to defend them and the right to do so, or admit you can be censored by violence.
Maybe but that does not change the fact we know cartoons like this will increase the risk of terrorism, you can condemn terrorism and those who resort to violence but it will lead to that reaction and further anti West feeling in the Muslim world and it is easier not to start fires in the first place rather than constantly having to put them out
The man who wants to send the tanks in to Scotland advocates caving in to the demands of terrorists.
There are 1.8 billion Muslims worldwide, only 1.6 million Scots voted Yes to independence in 2014 in a once in a generation referendum.
It is not just terrorists wanted to ban the cartoons of the Prophet Muhammed as the mass demonstrations across the Muslim world when they are published show
Ah, so because its harder to stand up to the numbers who want to ban it, we should give in and have unofficial bans?
I would not necessarily ban it but I am generally not a great fan of cartoons mocking Jesus for the sake of it either, for Muslims they just take it even more seriously, provocation for the sake of provocation just because you can is not a sensible idea
Indeed. @Casino_Royale was saying much the same about the mockery of Magma Carter guy, in the last thread.
But there’s a distinction between a reluctance to be offensive, and accepting the curtailing of freedom of speech.
But @Casino_Royale was not suggesting (and would not suggest) attacking the person doing the mockery or killing him. Whereas all too many Muslims do think violence the right response to mockery. And with such a view there can - and must be - no compromise. Macron is absolutely right on this.
Well Macron has said the right things, but is he going to back it up with any action?
I don't think tolerant Western liberals are capable of doing anything to combat this.
OK. High turnout leads to high uncertainty. Just who are these extra voters? Therefore I'll call it. Either they are for Biden leading to a landslide. Or for Trump leading to a surprise second term. Or they are pretty much even leading to a mess. Pretty sure you'll find out I was right all along!
I think someone should give HY an award for his continual efforts to get out the vote by HYping a tight finish to a contest effectively dead for weeks.
A famous day for PB on the sun soaked lawn of the White House, as the US president pins the Medal of Honor on our own, HY.
But behind the applause a distinct ‘pop’. Followed by the excruciating sound of a whoopee cushion trying desperately hard not to deflate.
I wouldn't back Biden at any shorter than 1.5 myself.
All the possible gains are very close except Wisconsin and Michigan. He really needs either Florida or Arizona as all the rest are more likely to be narrow misses IMO.
Obviously PA is the back stop but that is going to be legal challenge city if the EC depends on that.
OK. High turnout leads to high uncertainty. Just who are these extra voters? Therefore I'll call it. Either they are for Biden leading to a landslide. Or for Trump leading to a surprise second term. Or they are pretty much even leading to a mess. Pretty sure you'll find out I was right all along!
I know that is how I'll be betting. Thanks for the tips.
OK. High turnout leads to high uncertainty. Just who are these extra voters? Therefore I'll call it. Either they are for Biden leading to a landslide. Or for Trump leading to a surprise second term. Or they are pretty much even leading to a mess. Pretty sure you'll find out I was right all along!
I know that is how I'll be betting. Thanks for the tips.
Well FWIW this is my prediction, narrow misses for Biden in the Sunbelt, Trump wins Florida by 1% and Texas by 3%. All comes down to postal votes in Pennsylvania. Obviously I could be quite a way off with this! Can't see IA, TX, OH going blue; Trump wins thanks to large "shy Trump" vote, plus any post-election legal manoeuvres will I think be more likely to benefit the GOP. Of Sunbelt, maybe Arizona is best chance of a flip for Biden?
Anyway good luck to everyone betting and looking forward to the Zoom chat!
That's far from impossible...
But it also requires Biden to have extraordinarily inefficient votes.
Yes, it's not impossible, but I would make it about a 10/1 shot. That's about the price at which I would back Trump.
I think Biden holds everything Clinton held, including Minnesota and Nevada. I think he gains Michigan and Wisconsin. In those four states, Democrats are both governor and secretary of state except Nevada. So I assume attempts at massive fraud to favour Trump will be at least guarded against. If you believe the above, Biden has something like an outside straight draw, because he needs one of the following states: Pennsylvania, Ohio, North Carolina, Georgia, Florida, Texas. He needs 2+ to make a legal fight pointless. As you move the first four states in the opening assumptions to Trump, obviously Biden needs more of the big ones. Arizona doesn't do the job alone but it would make up for losing one of the first four. I think he wins one of Pennsylvania and Ohio with at least 50% probability, one of North Carolina, Georgia and Florida with at least 50% probability, somewhat but not 100% correlated with the first two. He wins Texas with some low probability that I can't bring myself to believe is 30% but is maybe 5-10%? and again not totally correlated with the first bunch because it would be won by a somewhat different coalition relying more heavily on Hispanic voters? So even with shenanigans I find it hard to argue that Trump is above 20% for a win. If I still believe this tomorrow, time to bet then.
Well FWIW this is my prediction, narrow misses for Biden in the Sunbelt, Trump wins Florida by 1% and Texas by 3%. All comes down to postal votes in Pennsylvania. Obviously I could be quite a way off with this! Can't see IA, TX, OH going blue; Trump wins thanks to large "shy Trump" vote, plus any post-election legal manoeuvres will I think be more likely to benefit the GOP. Of Sunbelt, maybe Arizona is best chance of a flip for Biden?
Anyway good luck to everyone betting and looking forward to the Zoom chat!
That's far from impossible...
But it also requires Biden to have extraordinarily inefficient votes.
Yes, it's not impossible, but I would make it about a 10/1 shot. That's about the price at which I would back Trump.
Thanks for the comments on the map, it's not based on the same level of analysis as the likes of Richard Nabavi and JackW (kudos to them and great to have Jack back!) but more of a "gut feel".
I think Texas, Ohio, Iowa are all out of reach for Biden, and feel that Florida will narrowly go for Trump, so if the map is to be avoided then NC or GA, or Arizona, need to be blue. Interestingly the map plus AZ blue would give 270-268 to Biden!
My "Golden Rule" of polling is that the polls aren't always wrong, but if they are, it's usually in the same direction, ie understating the right, and I feel that the number of "shy Trump" voters will probably exceed 2016.
Well FWIW this is my prediction, narrow misses for Biden in the Sunbelt, Trump wins Florida by 1% and Texas by 3%. All comes down to postal votes in Pennsylvania. Obviously I could be quite a way off with this! Can't see IA, TX, OH going blue; Trump wins thanks to large "shy Trump" vote, plus any post-election legal manoeuvres will I think be more likely to benefit the GOP. Of Sunbelt, maybe Arizona is best chance of a flip for Biden?
Anyway good luck to everyone betting and looking forward to the Zoom chat!
That's far from impossible...
But it also requires Biden to have extraordinarily inefficient votes.
Yes, it's not impossible, but I would make it about a 10/1 shot. That's about the price at which I would back Trump.
Well FWIW this is my prediction, narrow misses for Biden in the Sunbelt, Trump wins Florida by 1% and Texas by 3%. All comes down to postal votes in Pennsylvania. Obviously I could be quite a way off with this! Can't see IA, TX, OH going blue; Trump wins thanks to large "shy Trump" vote, plus any post-election legal manoeuvres will I think be more likely to benefit the GOP. Of Sunbelt, maybe Arizona is best chance of a flip for Biden?
Anyway good luck to everyone betting and looking forward to the Zoom chat!
This is almost my prediction, except I think Biden will just about win Pennsylvania, and therefore the election overall.
Jezza bending the covid rules and recommendations yet again.
Which rules?
Actually I am wrong. The inconsistent stupid government rules, which say on one hand if in Tier 2 you should avoid all unnecessary travel, especially to lower Tier areas....which is the bit I read first....but on the other, there is a section for holidays, which says sure, you are free to go on holiday to any Tier.
Which is frankly bonkers. It should be no unnecessary travel outside your Tier, full stop.
Jeremy Corbyn's brother is Pier Corbyn, and Piers Corbyn is friends with David Icke who lives on the Isle of Wight. Probably completely unconnected facts.
Jezza bending the covid rules and recommendations yet again.
Which rules?
Actually I am wrong. The inconsistent stupid government rules, which say on one hand if in Tier 2 you should avoid all unnecessary travel, especially to lower Tier areas....which is the bit I read first....but on the other, there is a section for holidays, which says sure, you are free to go on holiday to any Tier.
Which is frankly bonkers. It should be no unnecessary travel outside your Tier, full stop.
Well FWIW this is my prediction, narrow misses for Biden in the Sunbelt, Trump wins Florida by 1% and Texas by 3%. All comes down to postal votes in Pennsylvania. Obviously I could be quite a way off with this! Can't see IA, TX, OH going blue; Trump wins thanks to large "shy Trump" vote, plus any post-election legal manoeuvres will I think be more likely to benefit the GOP. Of Sunbelt, maybe Arizona is best chance of a flip for Biden?
Anyway good luck to everyone betting and looking forward to the Zoom chat!
That's far from impossible...
But it also requires Biden to have extraordinarily inefficient votes.
Yes, it's not impossible, but I would make it about a 10/1 shot. That's about the price at which I would back Trump.
Well FWIW this is my prediction, narrow misses for Biden in the Sunbelt, Trump wins Florida by 1% and Texas by 3%. All comes down to postal votes in Pennsylvania. Obviously I could be quite a way off with this! Can't see IA, TX, OH going blue; Trump wins thanks to large "shy Trump" vote, plus any post-election legal manoeuvres will I think be more likely to benefit the GOP. Of Sunbelt, maybe Arizona is best chance of a flip for Biden?
Anyway good luck to everyone betting and looking forward to the Zoom chat!
This is almost my prediction, except I think Biden will just about win Pennsylvania, and therefore the election overall.
Well FWIW this is my prediction, narrow misses for Biden in the Sunbelt, Trump wins Florida by 1% and Texas by 3%. All comes down to postal votes in Pennsylvania. Obviously I could be quite a way off with this! Can't see IA, TX, OH going blue; Trump wins thanks to large "shy Trump" vote, plus any post-election legal manoeuvres will I think be more likely to benefit the GOP. Of Sunbelt, maybe Arizona is best chance of a flip for Biden?
Anyway good luck to everyone betting and looking forward to the Zoom chat!
That's far from impossible...
But it also requires Biden to have extraordinarily inefficient votes.
Yes, it's not impossible, but I would make it about a 10/1 shot. That's about the price at which I would back Trump.
I'd back him at 7 or 8 to 1, but no less.
Can I take your odds please?
No. Because I'm betting in the market.
Why would I take your money when I can sell at 2-1?
Well FWIW this is my prediction, narrow misses for Biden in the Sunbelt, Trump wins Florida by 1% and Texas by 3%. All comes down to postal votes in Pennsylvania. Obviously I could be quite a way off with this! Can't see IA, TX, OH going blue; Trump wins thanks to large "shy Trump" vote, plus any post-election legal manoeuvres will I think be more likely to benefit the GOP. Of Sunbelt, maybe Arizona is best chance of a flip for Biden?
Anyway good luck to everyone betting and looking forward to the Zoom chat!
This is almost my prediction, except I think Biden will just about win Pennsylvania, and therefore the election overall.
Yes, Biden can take Pennsylvania which is enough to see him in the White House even if Trump holds on to Florida and Texas, which is by no means certain.
One interesting demographic factoid to have emerged from the polling is that Trump is attracting college-educated Latino voters, though I'm not really sure where this gets him.
Well FWIW this is my prediction, narrow misses for Biden in the Sunbelt, Trump wins Florida by 1% and Texas by 3%. All comes down to postal votes in Pennsylvania. Obviously I could be quite a way off with this! Can't see IA, TX, OH going blue; Trump wins thanks to large "shy Trump" vote, plus any post-election legal manoeuvres will I think be more likely to benefit the GOP. Of Sunbelt, maybe Arizona is best chance of a flip for Biden?
Anyway good luck to everyone betting and looking forward to the Zoom chat!
That's far from impossible...
But it also requires Biden to have extraordinarily inefficient votes.
Yes, it's not impossible, but I would make it about a 10/1 shot. That's about the price at which I would back Trump.
I'd back him at 7 or 8 to 1, but no less.
4/1 for me
The way I look at this is very simple.
It's only a 50/50 chance that the polling error is in Trump's favour. (And given these things tend to oscillate, maybe even less.)
The Morning Consult data I linked to earlier is very interesting. It's a poll of 7,350 early voters.
There's things in there for both candidates: Trump fans will like that 60% of early voters are HS educated or less. But they'll hate the 14 point gap between men and women at this stage. The rural percentage is also down on last time, ticking down two points - although this could of course, change on the day.
And it didn't seem to be lining up with the actual data and here's why:
Now Daniel Howdon, a research fellow at the Leeds Institute of Health Sciences, and Carl Heneghan, professor of Evidence-Based Medicine at the University of Oxford, have discovered that the Cambridge-PHE graph dates from three weeks ago. The pair have analysed the data and deduced that it predicted that deaths would be running at 1,000 a day by 1 November — i.e. yesterday. In fact, deaths have not turned out to be at anything like that level. The average for the past seven days (which smooths out the ‘weekend effect’) is 214 deaths per day.
The other graphs presented on Saturday, according to Howdon and Heneghan, are also at least three weeks old. The second most frightening projection came from Imperial College, which showed deaths peaking at just over 2,500 a day by January. This scenario also showed that deaths would be running at 486 a day by 1 November — more than twice as high as has happened in practice. The London School and Hygiene and Tropical Medicine and University of Warwick projections, which showed deaths peaking at around 2,000 a day predicted 266 and 234 deaths a day by 1 November respectively.
And it didn't seem to be lining up with the actual data and here's why:
Now Daniel Howdon, a research fellow at the Leeds Institute of Health Sciences, and Carl Heneghan, professor of Evidence-Based Medicine at the University of Oxford, have discovered that the Cambridge-PHE graph dates from three weeks ago. The pair have analysed the data and deduced that it predicted that deaths would be running at 1,000 a day by 1 November — i.e. yesterday. In fact, deaths have not turned out to be at anything like that level. The average for the past seven days (which smooths out the ‘weekend effect’) is 214 deaths per day.
The other graphs presented on Saturday, according to Howdon and Heneghan, are also at least three weeks old. The second most frightening projection came from Imperial College, which showed deaths peaking at just over 2,500 a day by January. This scenario also showed that deaths would be running at 486 a day by 1 November — more than twice as high as has happened in practice. The London School and Hygiene and Tropical Medicine and University of Warwick projections, which showed deaths peaking at around 2,000 a day predicted 266 and 234 deaths a day by 1 November respectively.
Three week old predictions, sorry 'scenarios', which have already been proved wrong.
Is nobody in Downing Street capable of any data analysis or is able to ask questions ?
Slide 4 and 5 were a huge give away that I pointed out at the time. Anybody with some knowledge of mathematical models should be raising their hand and asking a series of questions about them.
And it didn't seem to be lining up with the actual data and here's why:
Now Daniel Howdon, a research fellow at the Leeds Institute of Health Sciences, and Carl Heneghan, professor of Evidence-Based Medicine at the University of Oxford, have discovered that the Cambridge-PHE graph dates from three weeks ago. The pair have analysed the data and deduced that it predicted that deaths would be running at 1,000 a day by 1 November — i.e. yesterday. In fact, deaths have not turned out to be at anything like that level. The average for the past seven days (which smooths out the ‘weekend effect’) is 214 deaths per day.
The other graphs presented on Saturday, according to Howdon and Heneghan, are also at least three weeks old. The second most frightening projection came from Imperial College, which showed deaths peaking at just over 2,500 a day by January. This scenario also showed that deaths would be running at 486 a day by 1 November — more than twice as high as has happened in practice. The London School and Hygiene and Tropical Medicine and University of Warwick projections, which showed deaths peaking at around 2,000 a day predicted 266 and 234 deaths a day by 1 November respectively.
Three week old predictions, sorry 'scenarios', which have already been proved wrong.
Is nobody in Downing Street capable of any data analysis or is able to ask questions ?
Is three weeks ago when Keir Starmer was asking for circuit breakers? Though in any case, it is more the shape of the curve that matters, not whether you slide it left or right on the date axis. There have been reports yesterday that ambulance services and hospitals up north are feeling the strain. As to the quality of ministers...
And it didn't seem to be lining up with the actual data and here's why:
Now Daniel Howdon, a research fellow at the Leeds Institute of Health Sciences, and Carl Heneghan, professor of Evidence-Based Medicine at the University of Oxford, have discovered that the Cambridge-PHE graph dates from three weeks ago. The pair have analysed the data and deduced that it predicted that deaths would be running at 1,000 a day by 1 November — i.e. yesterday. In fact, deaths have not turned out to be at anything like that level. The average for the past seven days (which smooths out the ‘weekend effect’) is 214 deaths per day.
The other graphs presented on Saturday, according to Howdon and Heneghan, are also at least three weeks old. The second most frightening projection came from Imperial College, which showed deaths peaking at just over 2,500 a day by January. This scenario also showed that deaths would be running at 486 a day by 1 November — more than twice as high as has happened in practice. The London School and Hygiene and Tropical Medicine and University of Warwick projections, which showed deaths peaking at around 2,000 a day predicted 266 and 234 deaths a day by 1 November respectively.
Three week old predictions, sorry 'scenarios', which have already been proved wrong.
Is nobody in Downing Street capable of any data analysis or is able to ask questions ?
Is three weeks ago when Keir Starmer was asking for circuit breakers? Though in any case, it is more the shape of the curve that matters, not whether you slide it left or right on the date axis. There have been reports yesterday that ambulance services and hospitals up north are feeling the strain. As to the quality of ministers...
A policy that SAGE came up with where they couldn't say within 2 orders of magnitude how effective it might be. People demanding action based on that advice without requesting the modellers to go back and rework their model are also numerically illiterate.
OGH asked what other PBers are doing. As someone who reads these columns a lot, and rarely posts (as I am better at betting than writing) I thought I would share my betting “book” on this election. I am pretty heavily invested on the Betfair markets (having done over 2% of the total volumes). My markets are below (with the share of my money risked, may not add to 100% due to rounding). Biden for President 18% Biden Florida 6% Democrats Winning Party 3% Biden Popular Vote 13% Next President not to lose popular vote 13% Biden Maine 2% Biden Ohio 1% Biden Arizona 1% Biden N. Carolina 1% Senate NOM 1% Biden New Hampshire 1% Georgia negligible House Dem Majority 1% Minnesota Biden 2% Wisconsin Biden negligible (most closed out) Nevada Biden negligible (most closed out) Biden Pennsylvania 2% Biden Michigan 1% Texas Biden negligible (most closed out) Biden 100.5 handicap 2% Biden 48.5 handicap 1% Biden New Mexico 4% Trump lowish electoral votes 16% Biden highish electoral votes 11% Trump Missouri 1% Trump Montana 1% Biden Oregon 1% Biden New Jersey 1%
And it didn't seem to be lining up with the actual data and here's why:
Now Daniel Howdon, a research fellow at the Leeds Institute of Health Sciences, and Carl Heneghan, professor of Evidence-Based Medicine at the University of Oxford, have discovered that the Cambridge-PHE graph dates from three weeks ago. The pair have analysed the data and deduced that it predicted that deaths would be running at 1,000 a day by 1 November — i.e. yesterday. In fact, deaths have not turned out to be at anything like that level. The average for the past seven days (which smooths out the ‘weekend effect’) is 214 deaths per day.
The other graphs presented on Saturday, according to Howdon and Heneghan, are also at least three weeks old. The second most frightening projection came from Imperial College, which showed deaths peaking at just over 2,500 a day by January. This scenario also showed that deaths would be running at 486 a day by 1 November — more than twice as high as has happened in practice. The London School and Hygiene and Tropical Medicine and University of Warwick projections, which showed deaths peaking at around 2,000 a day predicted 266 and 234 deaths a day by 1 November respectively.
Three week old predictions, sorry 'scenarios', which have already been proved wrong.
Is nobody in Downing Street capable of any data analysis or is able to ask questions ?
Is three weeks ago when Keir Starmer was asking for circuit breakers? Though in any case, it is more the shape of the curve that matters, not whether you slide it left or right on the date axis. There have been reports yesterday that ambulance services and hospitals up north are feeling the strain. As to the quality of ministers...
Clearly the rapid rise curves haven't happened and the increase is much more gradual.
And we've been hearing the 'grim up north' tales for a month:
Well FWIW this is my prediction, narrow misses for Biden in the Sunbelt, Trump wins Florida by 1% and Texas by 3%. All comes down to postal votes in Pennsylvania. Obviously I could be quite a way off with this! Can't see IA, TX, OH going blue; Trump wins thanks to large "shy Trump" vote, plus any post-election legal manoeuvres will I think be more likely to benefit the GOP. Of Sunbelt, maybe Arizona is best chance of a flip for Biden?
Anyway good luck to everyone betting and looking forward to the Zoom chat!
That's far from impossible...
But it also requires Biden to have extraordinarily inefficient votes.
Yes, it's not impossible, but I would make it about a 10/1 shot. That's about the price at which I would back Trump.
I'd back him at 7 or 8 to 1, but no less.
Can I take your odds please?
No. Because I'm betting in the market.
Why would I take your money when I can sell at 2-1?
Obviously tongue in cheek.
My wider point though is that 8 or 10-1 just seem ludicrous in what is a two horse race.
As I think someone mentioned on here the other day, if you truly believe that a Biden win is so certain, you should be betting everything you have (if you can, I know you are in the US) on a Biden win. It's a 50% guaranteed return in a few days.
Jezza bending the covid rules and recommendations yet again.
Which rules?
Actually I am wrong. The inconsistent stupid government rules, which say on one hand if in Tier 2 you should avoid all unnecessary travel, especially to lower Tier areas....which is the bit I read first....but on the other, there is a section for holidays, which says sure, you are free to go on holiday to any Tier.
Which is frankly bonkers. It should be no unnecessary travel outside your Tier, full stop.
I absolutely agree with you, so much movement throughout all 2020 could not have been different if Covi19 itself was in charge.
Golf clubs shut. Sections of supermarket shut off. Yet look at the front of the Daily Star today. How does that add up?
And it didn't seem to be lining up with the actual data and here's why:
Now Daniel Howdon, a research fellow at the Leeds Institute of Health Sciences, and Carl Heneghan, professor of Evidence-Based Medicine at the University of Oxford, have discovered that the Cambridge-PHE graph dates from three weeks ago. The pair have analysed the data and deduced that it predicted that deaths would be running at 1,000 a day by 1 November — i.e. yesterday. In fact, deaths have not turned out to be at anything like that level. The average for the past seven days (which smooths out the ‘weekend effect’) is 214 deaths per day.
The other graphs presented on Saturday, according to Howdon and Heneghan, are also at least three weeks old. The second most frightening projection came from Imperial College, which showed deaths peaking at just over 2,500 a day by January. This scenario also showed that deaths would be running at 486 a day by 1 November — more than twice as high as has happened in practice. The London School and Hygiene and Tropical Medicine and University of Warwick projections, which showed deaths peaking at around 2,000 a day predicted 266 and 234 deaths a day by 1 November respectively.
Three week old predictions, sorry 'scenarios', which have already been proved wrong.
Is nobody in Downing Street capable of any data analysis or is able to ask questions ?
Is three weeks ago when Keir Starmer was asking for circuit breakers? Though in any case, it is more the shape of the curve that matters, not whether you slide it left or right on the date axis. There have been reports yesterday that ambulance services and hospitals up north are feeling the strain. As to the quality of ministers...
A policy that SAGE came up with where they couldn't say within 2 orders of magnitude how effective it might be. People demanding action based on that advice without requesting the modellers to go back and rework their model are also numerically illiterate.
This innumeracy business is a bit odd, given so many in government, both ministers and civil servants (and political journalists!) have Oxford PPEs. Economics is quite mathematical and even if PPE includes only introductory economics, there must surely be a lot of looking at graphs. It is the same as when the csv files were too full, there were people even on pb saying the data did not look right, that there must be something wrong even if they could not know what.
Well FWIW this is my prediction, narrow misses for Biden in the Sunbelt, Trump wins Florida by 1% and Texas by 3%. All comes down to postal votes in Pennsylvania. Obviously I could be quite a way off with this! Can't see IA, TX, OH going blue; Trump wins thanks to large "shy Trump" vote, plus any post-election legal manoeuvres will I think be more likely to benefit the GOP. Of Sunbelt, maybe Arizona is best chance of a flip for Biden?
Anyway good luck to everyone betting and looking forward to the Zoom chat!
That's far from impossible...
But it also requires Biden to have extraordinarily inefficient votes.
Yes, it's not impossible, but I would make it about a 10/1 shot. That's about the price at which I would back Trump.
I'd back him at 7 or 8 to 1, but no less.
Can I take your odds please?
No. Because I'm betting in the market.
Why would I take your money when I can sell at 2-1?
Obviously tongue in cheek.
My wider point though is that 8 or 10-1 just seem ludicrous in what is a two horse race.
As I think someone mentioned on here the other day, if you truly believe that a Biden win is so certain, you should be betting everything you have (if you can, I know you are in the US) on a Biden win. It's a 50% guaranteed return in a few days.
No, it's really not.
My logic is really quite simple.
It's 50/50 if Trump is 2.5% behind Biden.
The current polls show him about 8.5 points behind Biden.
There's only a 50% chance the polling error is in Trump's favour (and given the tendency of these errors to oscillate, that's quite generous). And you then have to believe the error is twice the Romney-Obama one, which was in itself the largest in recent times.
That makes Biden the clear favorite, but far from a certainty.
(Put it in context: if Lasik blinded you 12 to 16% of a time would to do it?)
Well FWIW this is my prediction, narrow misses for Biden in the Sunbelt, Trump wins Florida by 1% and Texas by 3%. All comes down to postal votes in Pennsylvania. Obviously I could be quite a way off with this! Can't see IA, TX, OH going blue; Trump wins thanks to large "shy Trump" vote, plus any post-election legal manoeuvres will I think be more likely to benefit the GOP. Of Sunbelt, maybe Arizona is best chance of a flip for Biden?
Anyway good luck to everyone betting and looking forward to the Zoom chat!
That's far from impossible...
But it also requires Biden to have extraordinarily inefficient votes.
Yes, it's not impossible, but I would make it about a 10/1 shot. That's about the price at which I would back Trump.
I'd back him at 7 or 8 to 1, but no less.
Can I take your odds please?
No. Because I'm betting in the market.
Why would I take your money when I can sell at 2-1?
Obviously tongue in cheek.
My wider point though is that 8 or 10-1 just seem ludicrous in what is a two horse race.
As I think someone mentioned on here the other day, if you truly believe that a Biden win is so certain, you should be betting everything you have (if you can, I know you are in the US) on a Biden win. It's a 50% guaranteed return in a few days.
Speaking as a small punter, and a jobless one, I've already invested far more than is sensible on a Biden victory, and am trying to resist the temptation to top up -- always save your bus fare home, as the saying goes. (ETA in 2016, I backed Trump.)
And it didn't seem to be lining up with the actual data and here's why:
Now Daniel Howdon, a research fellow at the Leeds Institute of Health Sciences, and Carl Heneghan, professor of Evidence-Based Medicine at the University of Oxford, have discovered that the Cambridge-PHE graph dates from three weeks ago. The pair have analysed the data and deduced that it predicted that deaths would be running at 1,000 a day by 1 November — i.e. yesterday. In fact, deaths have not turned out to be at anything like that level. The average for the past seven days (which smooths out the ‘weekend effect’) is 214 deaths per day.
The other graphs presented on Saturday, according to Howdon and Heneghan, are also at least three weeks old. The second most frightening projection came from Imperial College, which showed deaths peaking at just over 2,500 a day by January. This scenario also showed that deaths would be running at 486 a day by 1 November — more than twice as high as has happened in practice. The London School and Hygiene and Tropical Medicine and University of Warwick projections, which showed deaths peaking at around 2,000 a day predicted 266 and 234 deaths a day by 1 November respectively.
Three week old predictions, sorry 'scenarios', which have already been proved wrong.
Is nobody in Downing Street capable of any data analysis or is able to ask questions ?
Is three weeks ago when Keir Starmer was asking for circuit breakers? Though in any case, it is more the shape of the curve that matters, not whether you slide it left or right on the date axis. There have been reports yesterday that ambulance services and hospitals up north are feeling the strain. As to the quality of ministers...
A policy that SAGE came up with where they couldn't say within 2 orders of magnitude how effective it might be. People demanding action based on that advice without requesting the modellers to go back and rework their model are also numerically illiterate.
This innumeracy business is a bit odd, given so many in government, both ministers and civil servants (and political journalists!) have Oxford PPEs. Economics is quite mathematical and even if PPE includes only introductory economics, there must surely be a lot of looking at graphs. It is the same as when the csv files were too full, there were people even on pb saying the data did not look right, that there must be something wrong even if they could not know what.
It might seem "a bit odd", but it is demonstrated time and time and time again. It isn't just Boris not understanding this, it goes much much wider, and even worse many don't seem to have educated themselves over the past 6 months.
I don't think we go a week without Max, Malesbury, myself, load of other people, keep pointing out things that aren't correct or have been done in a way that no professional in that space would do.
These are the two scenarios I see as most plausible. Either there is a silent majority effect and Trump will win and possibly flip Nevada, or Biden will win comfortably.
As I think someone mentioned on here the other day, if you truly believe that a Biden win is so certain, you should be betting everything you have (if you can, I know you are in the US) on a Biden win. It's a 50% guaranteed return in a few days.
Not guaranteed, but 90% chance of a 50% return. 10% chance of losing my stake.
It’s a simple value bet. Very rare for the odds to be so divorced from the polls.
What we need is a judge led public inquiry into innumeracy.....
This is my problem with Dominic Cummings. He does identify real problems with government, including institutional innumeracy; it is just his solutions are misguided, self-defeating and in places fascistic.
ETA not to mention cartoonists' uncertainty as to whether he wears glasses.
These are the two scenarios I see as most plausible. Either there is a silent majority effect and Trump will win and possibly flip Nevada, or Biden will win comfortably.
No COVID and I could get behind this large silent majority. All those who were doing well economically, don't particularly like Trump as a person, but when push came to shove quietly vote for self interest / against the more radical Biden platform without admitting to their friends that is what they did and why.
However, COVID has effected everybody, and only the real Trumpsters buy he has done a good job re COVID.
Well FWIW this is my prediction, narrow misses for Biden in the Sunbelt, Trump wins Florida by 1% and Texas by 3%. All comes down to postal votes in Pennsylvania. Obviously I could be quite a way off with this! Can't see IA, TX, OH going blue; Trump wins thanks to large "shy Trump" vote, plus any post-election legal manoeuvres will I think be more likely to benefit the GOP. Of Sunbelt, maybe Arizona is best chance of a flip for Biden?
Anyway good luck to everyone betting and looking forward to the Zoom chat!
This is almost my prediction, except I think Biden will just about win Pennsylvania, and therefore the election overall.
Yes, Biden can take Pennsylvania which is enough to see him in the White House even if Trump holds on to Florida and Texas, which is by no means certain.
One interesting demographic factoid to have emerged from the polling is that Trump is attracting college-educated Latino voters, though I'm not really sure where this gets him.
What we need is a judge led public inquiry into innumeracy.....
This is my problem with Dominic Cummings. He does identify real problems with government, including institutional innumeracy; it is just his solutions are misguided, self-defeating and in places fascistic.
ETA not to mention cartoonists' uncertainty as to whether he wears glasses.
I wouldn't say his solutions are necessarily 100% misguided, but his way of doing business is turn everything into some sort of huge ideological battle against anybody who raises objections and in which he then turns all his attention to fighting this latest war, not actually improving things.
What we need is a judge led public inquiry into innumeracy.....
This is my problem with Dominic Cummings. He does identify real problems with government, including institutional innumeracy; it is just his solutions are misguided, self-defeating and in places fascistic.
ETA not to mention cartoonists' uncertainty as to whether he wears glasses.
I wouldn't say his solutions are necessarily 100% misguided, but his way of doing business is turn everything into some sort of huge ideological battle against anybody who raises objections and in which he then turns all his attention to fighting this latest war, not actually improving things.
Cummings is also too keen on central control, when all his reading should tell him this does not work and we need more localisation with network effects. The idea that all data should flow to a central decision maker was tested by Gordon Brown's premiership and failed then. Indeed the whole point of capitalism is not to answer the Soviet question of who organises the bread supply to London. Even at education, what was billed as autonomy for schools ended up, as the joke had it, with the largest education authority in the country being Michael Gove's desk, complete with rows about the minutiae of the history syllabus. Now he wants to bypass the whole of Whitehall and take all decisions himself from a "war room" in the Cabinet Office.
The closing 538 video is odd. At least two of the four contributors, including Nate Silver, seem strung out as if they have been awake for 72 hours, which leads to almost surreal rambling. There was also, as many will know but had escaped me, a bug in the 538 model which led it yesterday morning to predict a landslide for Trump (caused by a divide by zero error). It's not just Imperial College!
These are the two scenarios I see as most plausible. Either there is a silent majority effect and Trump will win and possibly flip Nevada, or Biden will win comfortably.
No COVID and I could get behind this large silent majority. All those who were doing well economically, don't particularly like Trump as a person, but when push came to shove quietly vote for self interest / against the more radical Biden platform without admitting to their friends that is what they did and why.
However, COVID has effected everybody, and only the real Trumpsters buy he has done a good job re COVID.
That’s where I am too, but I can’t quite rule out the possibility that he will pull off a John Major 1992-style win if people get cold feet about the Democrats.
We are now only about about two and a half hours from the first official results of the 2016 election.
Unlike last time, Hart's Location will not count at midnight, so we have Dixville Notch and Millsfield.
Dixville went 4 Clinton, 2 Trump, 1 Romney, 1 Johnson last time around. I expect Biden to win 5-4 this time around.
Millsfield went 16 Trump, 4 Clinton in 2016. That's a pretty strong result for the President. This time? I reckon Trump will get his 16 again (he did in the Primary earlier this year, after all), and I think Biden will improve on Clinton's 4 - getting 5 or maybe 6 votes at a pinch.
So... small swings to Biden in the (very) early New Hampshire vote.
These are the two scenarios I see as most plausible. Either there is a silent majority effect and Trump will win and possibly flip Nevada, or Biden will win comfortably.
No COVID and I could get behind this large silent majority. All those who were doing well economically, don't particularly like Trump as a person, but when push came to shove quietly vote for self interest / against the more radical Biden platform without admitting to their friends that is what they did and why.
However, COVID has effected everybody, and only the real Trumpsters buy he has done a good job re COVID.
That’s where I am too, but I can’t quite rule out the possibility that he will pull off a John Major 1992-style win if people get cold feet about the Democrats.
That was also, of course, on massively increased turnout.
As an aside, if Trump wins, Obamacare will be for the toilet, and that will be an utter disaster for the Republicans.
According to Wikipedia, there are 255 million people of voting age in the US. The all time record of turnout as a percentage of Voting Age Population was 1960 at 62.8%.
If you apply the same percentage this time around, you get turnout of 160 million.
To get the 170 million that some have suggested would require turnout of 66.7% which would be well above the record. Possible, sure. But far from certain.
What we need is a judge led public inquiry into innumeracy.....
This is my problem with Dominic Cummings. He does identify real problems with government, including institutional innumeracy; it is just his solutions are misguided, self-defeating and in places fascistic.
ETA not to mention cartoonists' uncertainty as to whether he wears glasses.
I wouldn't say his solutions are necessarily 100% misguided, but his way of doing business is turn everything into some sort of huge ideological battle against anybody who raises objections and in which he then turns all his attention to fighting this latest war, not actually improving things.
Cummings is also too keen on central control, when all his reading should tell him this does not work and we need more localisation with network effects. The idea that all data should flow to a central decision maker was tested by Gordon Brown's premiership and failed then. Indeed the whole point of capitalism is not to answer the Soviet question of who organises the bread supply to London. Even at education, what was billed as autonomy for schools ended up, as the joke had it, with the largest education authority in the country being Michael Gove's desk, complete with rows about the minutiae of the history syllabus. Now he wants to bypass the whole of Whitehall and take all decisions himself from a "war room" in the Cabinet Office.
I sometimes wonder what Gorbachev would make of Cummings, because the feeling in Cummings's oeuvre is very much "If I'd been in charge of perestroika, it would have worked!"
According to Wikipedia, there are 255 million people of voting age in the US. The all time record of turnout as a percentage of Voting Age Population was 1960 at 62.8%.
If you apply the same percentage this time around, you get turnout of 160 million.
To get the 170 million that some have suggested would require turnout of 66.7% which would be well above the record. Possible, sure. But far from certain.
You could get there by assuming that the percentage of RV voting will be at previous levels. If 81% of 211 million RV turn out, then you get over 170m.
In 2016, 136.8 million of 157.6 million RV voted, i.e. 86%. So similar RV turnout this time would get us to 183.2 million voters.
These are the two scenarios I see as most plausible. Either there is a silent majority effect and Trump will win and possibly flip Nevada, or Biden will win comfortably.
No COVID and I could get behind this large silent majority. All those who were doing well economically, don't particularly like Trump as a person, but when push came to shove quietly vote for self interest / against the more radical Biden platform without admitting to their friends that is what they did and why.
However, COVID has effected everybody, and only the real Trumpsters buy he has done a good job re COVID.
That’s where I am too, but I can’t quite rule out the possibility that he will pull off a John Major 1992-style win if people get cold feet about the Democrats.
That was also, of course, on massively increased turnout.
As an aside, if Trump wins, Obamacare will be for the toilet, and that will be an utter disaster for the Republicans.
The real battle will be in Congress (House of Reps especially), I dont sense a Republican overall victory across house of Reps or Senate (I could be wrong) but if the Republican party doesnt win them, then the battle over Obamacare will grind on for years and years....
According to my wife, I have not checked, The Lincoln Project is claiming that 13% of early voting registered Republicans have voted for Biden. No idea where they got this number from
According to my wife, I have not checked, The Lincoln Project is claiming that 13% of early voting registered Republicans have voted for Biden. No idea where they got this number from
They can't know for sure, it must just be estimates.
According to Wikipedia, there are 255 million people of voting age in the US. The all time record of turnout as a percentage of Voting Age Population was 1960 at 62.8%.
If you apply the same percentage this time around, you get turnout of 160 million.
To get the 170 million that some have suggested would require turnout of 66.7% which would be well above the record. Possible, sure. But far from certain.
Number of voters at recent presidential elections:
According to Wikipedia, there are 255 million people of voting age in the US. The all time record of turnout as a percentage of Voting Age Population was 1960 at 62.8%.
If you apply the same percentage this time around, you get turnout of 160 million.
To get the 170 million that some have suggested would require turnout of 66.7% which would be well above the record. Possible, sure. But far from certain.
Number of voters at recent presidential elections:
According to Wikipedia, there are 255 million people of voting age in the US. The all time record of turnout as a percentage of Voting Age Population was 1960 at 62.8%.
If you apply the same percentage this time around, you get turnout of 160 million.
To get the 170 million that some have suggested would require turnout of 66.7% which would be well above the record. Possible, sure. But far from certain.
Number of voters at recent presidential elections:
I can't see the number this time being more than 150m, but I'd be happy to be proved wrong tomorrow.
If it does happen, it will be because there has been a massive increase in voter registration - 53 million additional RV* - since 2016. It will take less than 25% of these newly registered voters to actually vote to take the number over 150m, assuming other voters are as eager as 2016.
* i.e. 6.5 million more than the entire British electorate at the Brexit referendum
According to Wikipedia, there are 255 million people of voting age in the US. The all time record of turnout as a percentage of Voting Age Population was 1960 at 62.8%.
If you apply the same percentage this time around, you get turnout of 160 million.
To get the 170 million that some have suggested would require turnout of 66.7% which would be well above the record. Possible, sure. But far from certain.
Number of voters at recent presidential elections:
I can't see the number this time being more than 150m, but I'd be happy to be proved wrong tomorrow.
If it does happen, it will be because there has been a massive increase in voter registration - 53 million additional RV* - since 2016. It will take less than 25% of these newly registered voters to actually vote to take the number over 150m, assuming other voters are as eager as 2016.
* i.e. 6.5 million more than the entire British electorate at the Brexit referendum
I can buy a 15% increase in the number of votes, which gets you to 160m votes. That would be - by some margin - the biggest increase in turnout percentage in modern history. But more than that is a real struggle.
Ryanair still operating flights in November even though passengers are not allowed to travel. Just so that they don't have to refund their tickets.
Not that I have much sympathy for anyone who has booked a flight in the past 6 months, mind.
Why not?
Putting aside the recklessness of international travel, people knew that the travel rules could change at the drop of a hat. So they can't really complain when exactly that happens.
Lots of people have very good reasons to travel. I haven’t done so myself because I’m not in that position.
You strike me as having absolutely no empathy with those who aren’t in your privileged position of being able to cocoon themselves inside for months on end while lecturing others on their “selfishness”.
I have no empathy for those who think they have a God-given right to spend a fortnight in Benidorm, Covid or no Covid.
It's not just holidays though. Many of my colleagues have family in other countries, and often elderly parents to visit, or even spouses and children. A flying ban is a major reduction in their way of life.
Not to mention business travel.
Not to mention the tiny detail that shedloads of jobs, and hence lives and livelihoods, depend on travel. You might equally say that eating out is an indulgence, yet if people gave it up, how many would you be putting out of work?
Those of us who have the financial security to be able to withstand these lockdowns ought to have the awareness to consider the impact on others before pontificating about what people should and shouldn’t do from our armchairs.
According to Wikipedia, there are 255 million people of voting age in the US. The all time record of turnout as a percentage of Voting Age Population was 1960 at 62.8%.
If you apply the same percentage this time around, you get turnout of 160 million.
To get the 170 million that some have suggested would require turnout of 66.7% which would be well above the record. Possible, sure. But far from certain.
Number of voters at recent presidential elections:
I can't see the number this time being more than 150m, but I'd be happy to be proved wrong tomorrow.
If it does happen, it will be because there has been a massive increase in voter registration - 53 million additional RV* - since 2016. It will take less than 25% of these newly registered voters to actually vote to take the number over 150m, assuming other voters are as eager as 2016.
* i.e. 6.5 million more than the entire British electorate at the Brexit referendum
I can buy a 15% increase in the number of votes, which gets you to 160m votes. That would be - by some margin - the biggest increase in turnout percentage in modern history. But more than that is a real struggle.
Yep, in my exchange with JackW, I said I expected around 150m, but the voter registration numbers are truly astounding (for the US) - basically 88% of VEP.
Dixville Notch will probably be 100% Democrat this year.
It was 4 Clinton, 2 Trump, 1 Romney, 1 Johnson in 2016. My guess is that it will go Biden 5, Trump 2-3.
Millsfield was 16-4 to Trump last time around, and Trump got 16 votes in the Republican Primary this time around, so I'd be staggered if he doesn't get 16 (or 17 if the other Registered Republican goes for him). If Biden is lucky he'll pick up a Republican and get 5, perhaps 6 at a stretch. I'd say 16-5 is the most likely result.
According to Wikipedia, there are 255 million people of voting age in the US. The all time record of turnout as a percentage of Voting Age Population was 1960 at 62.8%.
If you apply the same percentage this time around, you get turnout of 160 million.
To get the 170 million that some have suggested would require turnout of 66.7% which would be well above the record. Possible, sure. But far from certain.
Number of voters at recent presidential elections:
I can't see the number this time being more than 150m, but I'd be happy to be proved wrong tomorrow.
If it does happen, it will be because there has been a massive increase in voter registration - 53 million additional RV* - since 2016. It will take less than 25% of these newly registered voters to actually vote to take the number over 150m, assuming other voters are as eager as 2016.
* i.e. 6.5 million more than the entire British electorate at the Brexit referendum
Can you explain how the number of RVs has gone up so much?
Is this entirely due to individuals all taking their own separate decision to choose to register? Or have there been large public campaigns? Or something else?
The increase is so vast that it seems hard to believe without some very good explanation.
What we need is a judge led public inquiry into innumeracy.....
This is my problem with Dominic Cummings. He does identify real problems with government, including institutional innumeracy; it is just his solutions are misguided, self-defeating and in places fascistic.
ETA not to mention cartoonists' uncertainty as to whether he wears glasses.
Identifying and diagnosing problems is way easier than producing solutions that will actually resolve the problems you have identified without creating equally damaging problems by way of consequences, and that are acceptable and capable of being implemented in the real world, rather than the fantasy world in your head. Politics is actually difficult, as Cummo is doubtless discovering.
According to Wikipedia, there are 255 million people of voting age in the US. The all time record of turnout as a percentage of Voting Age Population was 1960 at 62.8%.
If you apply the same percentage this time around, you get turnout of 160 million.
To get the 170 million that some have suggested would require turnout of 66.7% which would be well above the record. Possible, sure. But far from certain.
Number of voters at recent presidential elections:
I can't see the number this time being more than 150m, but I'd be happy to be proved wrong tomorrow.
If it does happen, it will be because there has been a massive increase in voter registration - 53 million additional RV* - since 2016. It will take less than 25% of these newly registered voters to actually vote to take the number over 150m, assuming other voters are as eager as 2016.
* i.e. 6.5 million more than the entire British electorate at the Brexit referendum
Can you explain how the number of RVs has gone up so much?
Is this entirely due to individuals all taking their own separate decision to choose to register? Or have there been large public campaigns? Or something else?
The increase is so vast that it seems hard to believe without some very good explanation.
Many deliberate campaigns. 1. The GOP has had a sustained 4-year campaign to register people it thinks are Republican voters as Republicans. Some of this will have resulted in formerly Democrat-registered voters changing their party affiliation, but a lot of it is new registrations. 2. The Democratic party has also had ambitious new voter-registration programmes
Between the parties, for example, in TX alone they have added 1.85 million RV to the electoral rolls.
3. BLM have had as a core programme registering more of their communities to vote, and getting them to vote black representatives to local government positions. 4. Individual, more local but nonetheless impressive initiatives, like Stacey Abrams Georgia-based voter-registration drive and vote protection advocacy following her close loss in the gubernatorial race.
My final prediction is Biden 279, Trump 259. Biden to take Michigan, Wisconsin, Pennsylvania and Nebraska 2nd.
I think Michigan and Wisconsin are going to be fairly easy Biden pickups. There are then about half a dozen states which could go either way (Florida, Georgia, Pennsylvania, Arizona, North Carolina, perhaps Texas and Iowa). And I guess I don't think Trump gets lucky twice: at least one and probably more of these fall.
Comments
Therefore I'll call it. Either they are for Biden leading to a landslide. Or for Trump leading to a surprise second term. Or they are pretty much even leading to a mess.
Pretty sure you'll find out I was right all along!
A famous day for PB on the sun soaked lawn of the White House, as the US president pins the Medal of Honor on our own, HY.
But behind the applause a distinct ‘pop’. Followed by the excruciating sound of a whoopee cushion trying desperately hard not to deflate.
I wouldn't back Biden at any shorter than 1.5 myself.
All the possible gains are very close except Wisconsin and Michigan. He really needs either Florida or Arizona as all the rest are more likely to be narrow misses IMO.
Obviously PA is the back stop but that is going to be legal challenge city if the EC depends on that.
Yours Biden bedwetter
Pennsylvania
Trump 48%
Biden 46%
Ohio
Trump 49%
Biden 44%
Michigan
Trump 48%
Biden 46%
North Carolina
Trump 49%
Biden 47%
Nevada
Biden 49%
Trump 47%
Arizona
Trump 49%
Biden 46%
Florida
Trump 50%
Biden 47%
Minnesota
Biden 48%
Trump 45%
Wisconsin
Biden 48%
Trump 47%
Louisiana
Trump 54%
Biden 36%
Missouri
Trump 52%
Biden 41%
Texas
Trump 49%
Biden 43%
Georgia
Trump 50%
Biden 43%
And so would I.
It to be so close in Wisconsin no chance.
PA Trump to win unlikely without SCOTUS help.
All others they could have the right winners but exaggerated margins and I hope either Arizona or Florida scrape through as blue.
Its Election day
Yeahhhh
I think Texas, Ohio, Iowa are all out of reach for Biden, and feel that Florida will narrowly go for Trump, so if the map is to be avoided then NC or GA, or Arizona, need to be blue. Interestingly the map plus AZ blue would give 270-268 to Biden!
My "Golden Rule" of polling is that the polls aren't always wrong, but if they are, it's usually in the same direction, ie understating the right, and I feel that the number of "shy Trump" voters will probably exceed 2016.
Which is frankly bonkers. It should be no unnecessary travel outside your Tier, full stop.
Anyway Corbyn is yesterday's man.
Onwards.
Why would I take your money when I can sell at 2-1?
One interesting demographic factoid to have emerged from the polling is that Trump is attracting college-educated Latino voters, though I'm not really sure where this gets him.
It's only a 50/50 chance that the polling error is in Trump's favour. (And given these things tend to oscillate, maybe even less.)
You then need to have probably a four point miss.
I think that's a 12-18% chance all-in.
There's things in there for both candidates: Trump fans will like that 60% of early voters are HS educated or less. But they'll hate the 14 point gap between men and women at this stage. The rural percentage is also down on last time, ticking down two points - although this could of course, change on the day.
https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/931775/Slides_to_accompany_coronavirus_press_conference-_CSA-__31_October_2020.pdf
And it didn't seem to be lining up with the actual data and here's why:
Now Daniel Howdon, a research fellow at the Leeds Institute of Health Sciences, and Carl Heneghan, professor of Evidence-Based Medicine at the University of Oxford, have discovered that the Cambridge-PHE graph dates from three weeks ago. The pair have analysed the data and deduced that it predicted that deaths would be running at 1,000 a day by 1 November — i.e. yesterday. In fact, deaths have not turned out to be at anything like that level. The average for the past seven days (which smooths out the ‘weekend effect’) is 214 deaths per day.
The other graphs presented on Saturday, according to Howdon and Heneghan, are also at least three weeks old. The second most frightening projection came from Imperial College, which showed deaths peaking at just over 2,500 a day by January. This scenario also showed that deaths would be running at 486 a day by 1 November — more than twice as high as has happened in practice. The London School and Hygiene and Tropical Medicine and University of Warwick projections, which showed deaths peaking at around 2,000 a day predicted 266 and 234 deaths a day by 1 November respectively.
https://www.spectator.co.uk/article/the-problem-with-downing-street-s-covid-projections
Three week old predictions, sorry 'scenarios', which have already been proved wrong.
Is nobody in Downing Street capable of any data analysis or is able to ask questions ?
My markets are below (with the share of my money risked, may not add to 100% due to rounding).
Biden for President 18%
Biden Florida 6%
Democrats Winning Party 3%
Biden Popular Vote 13%
Next President not to lose popular vote 13%
Biden Maine 2%
Biden Ohio 1%
Biden Arizona 1%
Biden N. Carolina 1%
Senate NOM 1%
Biden New Hampshire 1%
Georgia negligible
House Dem Majority 1%
Minnesota Biden 2%
Wisconsin Biden negligible (most closed out)
Nevada Biden negligible (most closed out)
Biden Pennsylvania 2%
Biden Michigan 1%
Texas Biden negligible (most closed out)
Biden 100.5 handicap 2%
Biden 48.5 handicap 1%
Biden New Mexico 4%
Trump lowish electoral votes 16%
Biden highish electoral votes 11%
Trump Missouri 1%
Trump Montana 1%
Biden Oregon 1%
Biden New Jersey 1%
And we've been hearing the 'grim up north' tales for a month:
https://www.theguardian.com/world/2020/oct/08/hospitals-in-north-of-england-to-run-out-of-covid-beds-within-a-week
Now there's 'boy cries wolf' possibilities but UK hospital usage seems to be far lower than it is in other European countries.
My wider point though is that 8 or 10-1 just seem ludicrous in what is a two horse race.
As I think someone mentioned on here the other day, if you truly believe that a Biden win is so certain, you should be betting everything you have (if you can, I know you are in the US) on a Biden win. It's a 50% guaranteed return in a few days.
Golf clubs shut. Sections of supermarket shut off. Yet look at the front of the Daily Star today. How does that add up?
https://storify.com/services/proxy/2/K8pZQ_zD6gqLxChAcGRoKg/https/media.fyre.co/Ph0cwzvcRcyN0FoxCl9s_ds03p01_1604357236_001.png
My logic is really quite simple.
It's 50/50 if Trump is 2.5% behind Biden.
The current polls show him about 8.5 points behind Biden.
There's only a 50% chance the polling error is in Trump's favour (and given the tendency of these errors to oscillate, that's quite generous). And you then have to believe the error is twice the Romney-Obama one, which was in itself the largest in recent times.
That makes Biden the clear favorite, but far from a certainty.
(Put it in context: if Lasik blinded you 12 to 16% of a time would to do it?)
I don't think we go a week without Max, Malesbury, myself, load of other people, keep pointing out things that aren't correct or have been done in a way that no professional in that space would do.
It’s a simple value bet. Very rare for the odds to be so divorced from the polls.
I’m reinvesting a fair chunk of my 2016 winnings.
David Wasserman"
https://cookpolitical.com/analysis/national/national-politics/what-one-florida-county-could-tell-us-about-gray-revolt
ETA not to mention cartoonists' uncertainty as to whether he wears glasses.
However, COVID has effected everybody, and only the real Trumpsters buy he has done a good job re COVID.
(What will everybody be drinking on election night?)
Unlike last time, Hart's Location will not count at midnight, so we have Dixville Notch and Millsfield.
Dixville went 4 Clinton, 2 Trump, 1 Romney, 1 Johnson last time around. I expect Biden to win 5-4 this time around.
Millsfield went 16 Trump, 4 Clinton in 2016. That's a pretty strong result for the President. This time? I reckon Trump will get his 16 again (he did in the Primary earlier this year, after all), and I think Biden will improve on Clinton's 4 - getting 5 or maybe 6 votes at a pinch.
So... small swings to Biden in the (very) early New Hampshire vote.
As an aside, if Trump wins, Obamacare will be for the toilet, and that will be an utter disaster for the Republicans.
If you apply the same percentage this time around, you get turnout of 160 million.
To get the 170 million that some have suggested would require turnout of 66.7% which would be well above the record. Possible, sure. But far from certain.
In 2016, 136.8 million of 157.6 million RV voted, i.e. 86%. So similar RV turnout this time would get us to 183.2 million voters.
Don't think it will happen, but ...
1992: 104.6m
1996: 96.4m
2000: 105.6m
2004: 122.3m
2008: 131.4m
2012: 129.2m
2016: 138.8m
I can't see the number this time being more than 150m, but I'd be happy to be proved wrong tomorrow.
* i.e. 6.5 million more than the entire British electorate at the Brexit referendum
Those of us who have the financial security to be able to withstand these lockdowns ought to have the awareness to consider the impact on others before pontificating about what people should and shouldn’t do from our armchairs.
Millsfield was 16-4 to Trump last time around, and Trump got 16 votes in the Republican Primary this time around, so I'd be staggered if he doesn't get 16 (or 17 if the other Registered Republican goes for him). If Biden is lucky he'll pick up a Republican and get 5, perhaps 6 at a stretch. I'd say 16-5 is the most likely result.
Is this entirely due to individuals all taking their own separate decision to choose to register? Or have there been large public campaigns? Or something else?
The increase is so vast that it seems hard to believe without some very good explanation.
Trump's chance of winning PA has moved up slightly to 16% according to 538.
https://projects.fivethirtyeight.com/2020-election-forecast/pennsylvania/
1. The GOP has had a sustained 4-year campaign to register people it thinks are Republican voters as Republicans. Some of this will have resulted in formerly Democrat-registered voters changing their party affiliation, but a lot of it is new registrations.
2. The Democratic party has also had ambitious new voter-registration programmes
Between the parties, for example, in TX alone they have added 1.85 million RV to the electoral rolls.
3. BLM have had as a core programme registering more of their communities to vote, and getting them to vote black representatives to local government positions.
4. Individual, more local but nonetheless impressive initiatives, like Stacey Abrams Georgia-based voter-registration drive and vote protection advocacy following her close loss in the gubernatorial race.
Biden 330 + EC
and the Senate will flip at least 5 to 6 seats.
It will be a crushing victory that will set America on a new path.
New Thread