I think you've just become HYUFD's favourite pollster
Not really the full Blog is generally positive for Biden.
I still think he will win but it was enough to make me cash out in at least 2 states.
A very interesting read. I follow Ralstons Nevada EV blog closely and he's pretty bullish now about Biden winning there though perhaps by not much more than Clinton did. Generally Ralston downplays confidence until he's pretty sure. NC is interesting and does seem as if Trump has a better than even chance of holding on there, I did have that in his column but this post confirms my view. FL too close to call yes but expecting Trump to hold on there too. AZ I think is flipping and the figures here support that somewhat. Still not convinced Iowa will flip so I'm keeping that in my red column too.
The Archbishop of Canterbury has written to Church of England clergy about the new lockdown provisions in England, which allow churches to remain open but ban communal acts of worship.
In his letter, Justin Welby says he is grateful that the new provisions allow churches to remain open for private prayer.
But he says worshippers not having access to the sacraments is a huge loss and that the Church was not consulted about the new provisions.
He says he intends to speak to the government about why certain exemptions have been made, but not others, and adds that the sacramental life of the Church is not an optional extra.
But he adds "we will abide by the law".
The letter follows a statement from the leader of Catholics in England and Wales yesterday calling for the government to provide evidence as to why communal worship should be banned and a similar call from the Muslim Council of Britain.
.....if they don't know by now why this is a problem, God help them.
The government, much as everyone loves to criticise, have got this one right.
They're happy to keep large indoor spaces such as churches open, but don't want organised events taking place inside where many people congregate at the same time. The alternative is to shut them, alongside pubs and shops.
It will be interesting to see what list of exceptions there is in these regulations, when the proposals are published. There were so many in the last lot that, with a bit of imagination, almost anything could have been made legal.
Do we know when the regulations (ie laws, as opposed to the guidance) will be published?
No, but they weren’t there when I looked an hour or so ago. Doubtless some officials had wrecked weekends and are still working on it. judging from last performance we may not see anything until Weds morning
Sounds like Amber Heard and Johnny Depp beat seven shades of shit out of each other. Libel action lost because it's factually overwhemingly likely he did beat her, it's also factually overwhemingly likely she attacked him - but that's nothing to do with the libel case. What a train wreck of a relationship.
Yay! Mrs RP reports that two of her teaching colleagues now have Covid. Does make me laugh how people are flapping about "lockdown" when its nothing of the sort
I still remember the Daly Mail/Daily express comments from July/August about how safe it would be in schools when the pupils went back. I remember them saying that students don't transmit to each other so it's all right. When challenged about infecting teachers they said "the jury is still out on that one". Outrageous. How many teachers, parents and grannies have died due to this.
Farage has judged this wrong in terms of mass appeal among those that don't hate his guts. Oldies are scared and want to be locked away. It is mostly the youngsters, who believe he is a massive racist, that are willing to entertain the idea of just getting on it, as lockdowns are destroying their jobs and prospects.
Really the LibDems should be on this one. I don’t see they have anything to gain by joining Starmer in pushing the government to lockdown sooner and further
As pragmatists, the Lib Dems prefer to be to be on the side of reason. They have nothing at all to gain by jumping on a populist bandwagon.
As no lockdown has been announced it is difficult to argue against one.
There's been the thought around for some time that Labour has a handle on Boris, who's slowly self-destructing anyway. Rishi, it's felt, is an entirely different kettle of fish and will have amassed vast amounts of political capital by the time he ascends to the leadership, rendering him virtually invincible. It makes sense for Sir Keir to taint the Wonder Kid with some effluence of the Boris era before that happens.
Labour seem to want to focus on Rishi as he may succeed Boris sooner than later
Also notice Starmer did not reference his 8% proposed corporation tax increase to the CBI
It's a possibility, though it seems like a decent attack line anyway.
Rishi is about the only senior Conservative with a good public image right now. Partly, he does seem genuinely good for a young'un, partly because his shiny image is still shiny (though I can see him becoming more than faintly absurd if he keeps the Insta-vibe going much longer). But mostly because, he hosed money around like a drunken sailor. The right thing to do, but also easy popularity.
But was that the real Rishi? He is, after all, a rich man who worked for a hedge fund. The government has been considerably less generous during the time of tiers. I don't think he's denied that he doesn't like the cost of Lockdown 2. If the opposition can convert Dishy Rishi into Stingy Sunak in the public's mind, that seems worth their while.
Farage has judged this wrong in terms of mass appeal among those that don't hate his guts. Oldies are scared and want to be locked away. It is mostly the youngsters, who believe he is a massive racist, that are willing to entertain the idea of just getting on it, as lockdowns are destroying their jobs and prospects.
Really the LibDems should be on this one. I don’t see they have anything to gain by joining Starmer in pushing the government to lockdown sooner and further
Arguing for more deaths and more economic damage might weigh a bit more heavily on their conscience than producing dodgy bar charts though. No, this is one for the far left and right.
They're in Opposition. They can just hammer the government over and over on its failure to isolate the infectious and say the lockdown is the price of that failure.
Farage has judged this wrong in terms of mass appeal among those that don't hate his guts. Oldies are scared and want to be locked away. It is mostly the youngsters, who believe he is a massive racist, that are willing to entertain the idea of just getting on it, as lockdowns are destroying their jobs and prospects.
Really the LibDems should be on this one. I don’t see they have anything to gain by joining Starmer in pushing the government to lockdown sooner and further
As pragmatists, the Lib Dems prefer to be to be on the side of reason. They have nothing at all to gain by jumping on a populist bandwagon.
Clearly it’s neither populist nor a bandwagon right now.
Yay! Mrs RP reports that two of her teaching colleagues now have Covid. Does make me laugh how people are flapping about "lockdown" when its nothing of the sort
I still remember the Daly Mail/Daily express comments from July/August about how safe it would be in schools when the pupils went back. I remember them saying that students don't transmit to each other so it's all right. When challenged about infecting teachers they said "the jury is still out on that one". Outrageous. How many teachers, parents and grannies have died due to this.
Just to add to the fun my eldest starts his work experience tomorrow in a primary school...
Quick shout out to @HYUFD . (1) Yes we’ve seen the Trafalgar Michigan Poll, please don’t embed a tweet in a post as it slows the site from loading, (2) no, no one with a brain is saying Trump can’t win and (3) we take Nate Silver seriously because he deals in odds and probabilities and this is a gambling site he’s not a pollster - even he takes Trafalgar into account. Hope that clears things up for you.
Farage has judged this wrong in terms of mass appeal among those that don't hate his guts. Oldies are scared and want to be locked away. It is mostly the youngsters, who believe he is a massive racist, that are willing to entertain the idea of just getting on it, as lockdowns are destroying their jobs and prospects.
Really the LibDems should be on this one. I don’t see they have anything to gain by joining Starmer in pushing the government to lockdown sooner and further
As pragmatists, the Lib Dems prefer to be to be on the side of reason. They have nothing at all to gain by jumping on a populist bandwagon.
Clearly it’s neither populist nor a bandwagon right now.
I think you've just become HYUFD's favourite pollster
Not really the full Blog is generally positive for Biden.
I still think he will win but it was enough to make me cash out in at least 2 states.
Positive how? He's losing in FL and NC, two states that he'd hoped to flip.
Without one of those, his path to the white house is very much narrower.
The rust belt is what matters in this fight. I don't see how Trump manages those same small margin victories again facing a candidate who doesn't turn voters off in droves.
Instagram index on Depp - everyone thinks the judge has been taken in by Heard fluttering her eyelashes at him.
I haven't followed this super closely, but neither individual appeared particular nice.
Oh, I suspect that's probably true,
But, of course, Heard wasn't the one who brought a disastrous libel case. The focus is on Depp's conduct because the newspaper allegations were about him, and his conduct. The judge implies in the judgment that Heard probably was sleeping around, for example, but correctly points out that he doesn't need to make a determination on that because it wouldn't justify Depp's conduct even if she was.
That's the danger of libel cases. The simple rule is, if you have skeletons in the closet, even if you feel you've got right on your side, don't even think about risking it.
With all the talk about models this morning, I did a quick'n'dirty simplistic model to try to work out how many deaths are already baked in.
First graph: Using ONS stats for infections per day in England, take an IFR of 0.65%, take date of death as infection day + 19, multiply by 1.2 for all of UK (matches existing stats pretty well), and extrapolate forward. - Red bars are deaths already recorded - Translucent orange bars are against the existing ONS stats (baked in) - Translucent grey bars are extrapolating the ONS rate of increase between the last two surveys (questionable, but the best prior we can get, really. If rate of increase has fallen, will over-predict. If spread has pushed further into vulnerable demographics, will under-predict. I expect they'll cancel out to some extent).
That's another 5,600 deaths until the lockdown kicks in. Peaking at over 900 deaths per day.
Piling assumption on assumption ("all models are wrong, but some are useful"), assuming an R of 0.9 during the new lockdown, and extrapolating forwards to the end of lockdown (which gives deaths on Christmas Eve), and with the same colour scheme:
Quite depressing. That's a total of 36,000 deaths between now and Christmas.
I really hope this is an overly pessimistic extrapolation.
Are we staying within hospital capacity? Otherwise that IFR will go up significantly.
I think so. The peak is very close to the March/April peak, and we stayed (just?) within hospital capacity then. It does imply that they acted not a moment too soon. Another week (if - and it's a big if - the virus spread was continuing at that rate) and we'd have gone over.
Will be interesting to see how reality compares to this prediction. Can we expect updates?
Sure. I can do it daily, but the death figures are usually fairly laggy. Might make more sense to update it on Thursdays and Mondays. And I'll be hoping the model is overpredicting.
Friday (when the ONS survey is updated) and Tuesday (when the highest number of backdated deaths are normally reported) would seem the obvious days if possible.
Good point. I'll update it every Thursday after the ONS survey is updated. This will cause the translucent grey bars to alter; in order to have the ability to measure against the earlier versions, I'll actually add a separate series for "following new ONS survey data," when it comes in.
With all the talk about models this morning, I did a quick'n'dirty simplistic model to try to work out how many deaths are already baked in.
First graph: Using ONS stats for infections per day in England, take an IFR of 0.65%, take date of death as infection day + 19, multiply by 1.2 for all of UK (matches existing stats pretty well), and extrapolate forward. - Red bars are deaths already recorded - Translucent orange bars are against the existing ONS stats (baked in) - Translucent grey bars are extrapolating the ONS rate of increase between the last two surveys (questionable, but the best prior we can get, really. If rate of increase has fallen, will over-predict. If spread has pushed further into vulnerable demographics, will under-predict. I expect they'll cancel out to some extent).
That's another 5,600 deaths until the lockdown kicks in. Peaking at over 900 deaths per day.
Piling assumption on assumption ("all models are wrong, but some are useful"), assuming an R of 0.9 during the new lockdown, and extrapolating forwards to the end of lockdown (which gives deaths on Christmas Eve), and with the same colour scheme:
Quite depressing. That's a total of 36,000 deaths between now and Christmas.
I really hope this is an overly pessimistic extrapolation.
Are we staying within hospital capacity? Otherwise that IFR will go up significantly.
I think so. The peak is very close to the March/April peak, and we stayed (just?) within hospital capacity then. It does imply that they acted not a moment too soon. Another week (if - and it's a big if - the virus spread was continuing at that rate) and we'd have gone over.
Will be interesting to see how reality compares to this prediction. Can we expect updates?
Sure. I can do it daily, but the death figures are usually fairly laggy. Might make more sense to update it on Thursdays and Mondays. And I'll be hoping the model is overpredicting.
Friday (when the ONS survey is updated) and Tuesday (when the highest number of backdated deaths are normally reported) would seem the obvious days if possible.
Good point. I'll update it every Thursday after the ONS survey is updated. This will cause the translucent grey bars to alter; in order to have the ability to measure against the earlier versions, I'll actually add a separate series for "following new ONS survey data," when it comes in.
How does it feel to be the new owner of the "example scenario"?
Quick shout out to @HYUFD . (1) Yes we’ve seen the Trafalgar Michigan Poll, please don’t embed a tweet in a post as it slows the site from loading, (2) no, no one with a brain is saying Trump can’t win and (3) we take Nate Silver seriously because he deals in odds and probabilities and this is a gambling site he’s not a pollster - even he takes Trafalgar into account. Hope that clears things up for you.
Thanks. The site is absolutely unusable on the iPhone when people embed tweets.
With all the talk about models this morning, I did a quick'n'dirty simplistic model to try to work out how many deaths are already baked in.
First graph: Using ONS stats for infections per day in England, take an IFR of 0.65%, take date of death as infection day + 19, multiply by 1.2 for all of UK (matches existing stats pretty well), and extrapolate forward. - Red bars are deaths already recorded - Translucent orange bars are against the existing ONS stats (baked in) - Translucent grey bars are extrapolating the ONS rate of increase between the last two surveys (questionable, but the best prior we can get, really. If rate of increase has fallen, will over-predict. If spread has pushed further into vulnerable demographics, will under-predict. I expect they'll cancel out to some extent).
That's another 5,600 deaths until the lockdown kicks in. Peaking at over 900 deaths per day.
Piling assumption on assumption ("all models are wrong, but some are useful"), assuming an R of 0.9 during the new lockdown, and extrapolating forwards to the end of lockdown (which gives deaths on Christmas Eve), and with the same colour scheme:
Quite depressing. That's a total of 36,000 deaths between now and Christmas.
I really hope this is an overly pessimistic extrapolation.
Are we staying within hospital capacity? Otherwise that IFR will go up significantly.
I think so. The peak is very close to the March/April peak, and we stayed (just?) within hospital capacity then. It does imply that they acted not a moment too soon. Another week (if - and it's a big if - the virus spread was continuing at that rate) and we'd have gone over.
Will be interesting to see how reality compares to this prediction. Can we expect updates?
Sure. I can do it daily, but the death figures are usually fairly laggy. Might make more sense to update it on Thursdays and Mondays. And I'll be hoping the model is overpredicting.
Friday (when the ONS survey is updated) and Tuesday (when the highest number of backdated deaths are normally reported) would seem the obvious days if possible.
Good point. I'll update it every Thursday after the ONS survey is updated. This will cause the translucent grey bars to alter; in order to have the ability to measure against the earlier versions, I'll actually add a separate series for "following new ONS survey data," when it comes in.
The ONS survey data comes out on Friday, but yes, that sounds like a good idea.
Instagram index on Depp - everyone thinks the judge has been taken in by Heard fluttering her eyelashes at him.
There is an audio recording of her admitting to being violent with Depp.
And if she was the one suing the Sun for calling her a husband beater, that would be a hell of a lot more relevant.
The court wasn't deciding whether Heard or Depp was the worse person in the relationship... it was deciding whether the allegations in the Sun about Depp, to the effect HE was violent towards HER on several occasions, were substantively true. And, to the civil standard, they were.
Farage has judged this wrong in terms of mass appeal among those that don't hate his guts. Oldies are scared and want to be locked away. It is mostly the youngsters, who believe he is a massive racist, that are willing to entertain the idea of just getting on it, as lockdowns are destroying their jobs and prospects.
Really the LibDems should be on this one. I don’t see they have anything to gain by joining Starmer in pushing the government to lockdown sooner and further
Arguing for more deaths and more economic damage might weigh a bit more heavily on their conscience than producing dodgy bar charts though. No, this is one for the far left and right.
With all the talk about models this morning, I did a quick'n'dirty simplistic model to try to work out how many deaths are already baked in.
First graph: Using ONS stats for infections per day in England, take an IFR of 0.65%, take date of death as infection day + 19, multiply by 1.2 for all of UK (matches existing stats pretty well), and extrapolate forward. - Red bars are deaths already recorded - Translucent orange bars are against the existing ONS stats (baked in) - Translucent grey bars are extrapolating the ONS rate of increase between the last two surveys (questionable, but the best prior we can get, really. If rate of increase has fallen, will over-predict. If spread has pushed further into vulnerable demographics, will under-predict. I expect they'll cancel out to some extent).
That's another 5,600 deaths until the lockdown kicks in. Peaking at over 900 deaths per day.
Piling assumption on assumption ("all models are wrong, but some are useful"), assuming an R of 0.9 during the new lockdown, and extrapolating forwards to the end of lockdown (which gives deaths on Christmas Eve), and with the same colour scheme:
Quite depressing. That's a total of 36,000 deaths between now and Christmas.
I really hope this is an overly pessimistic extrapolation.
Are we staying within hospital capacity? Otherwise that IFR will go up significantly.
I think so. The peak is very close to the March/April peak, and we stayed (just?) within hospital capacity then. It does imply that they acted not a moment too soon. Another week (if - and it's a big if - the virus spread was continuing at that rate) and we'd have gone over.
Will be interesting to see how reality compares to this prediction. Can we expect updates?
Sure. I can do it daily, but the death figures are usually fairly laggy. Might make more sense to update it on Thursdays and Mondays. And I'll be hoping the model is overpredicting.
Friday (when the ONS survey is updated) and Tuesday (when the highest number of backdated deaths are normally reported) would seem the obvious days if possible.
Good point. I'll update it every Thursday after the ONS survey is updated. This will cause the translucent grey bars to alter; in order to have the ability to measure against the earlier versions, I'll actually add a separate series for "following new ONS survey data," when it comes in.
Thanks Andy – can you link the images rather than embedding them though?
Labour seem to want to focus on Rishi as he may succeed Boris sooner than later
Also notice Starmer did not reference his 8% proposed corporation tax increase to the CBI
It's a possibility, though it seems like a decent attack line anyway.
Rishi is about the only senior Conservative with a good public image right now. Partly, he does seem genuinely good for a young'un, partly because his shiny image is still shiny (though I can see him becoming more than faintly absurd if he keeps the Insta-vibe going much longer). But mostly because, he hosed money around like a drunken sailor. The right thing to do, but also easy popularity.
But was that the real Rishi? He is, after all, a rich man who worked for a hedge fund. The government has been considerably less generous during the time of tiers. I don't think he's denied that he doesn't like the cost of Lockdown 2. If the opposition can convert Dishy Rishi into Stingy Sunak in the public's mind, that seems worth their while.
All political parties have members who are good people and members who are not. Rishi stands out from so many of current front line Tories because he is a decent human being. Yes he's a Conservative and is having to filter events through that world view. But he has thrown caution to the wind and thrown oceans of cash at people in ways that were almost inconceivable for a Tory chancellor to do.
Yay! Mrs RP reports that two of her teaching colleagues now have Covid. Does make me laugh how people are flapping about "lockdown" when its nothing of the sort
I still remember the Daly Mail/Daily express comments from July/August about how safe it would be in schools when the pupils went back. I remember them saying that students don't transmit to each other so it's all right. When challenged about infecting teachers they said "the jury is still out on that one". Outrageous. How many teachers, parents and grannies have died due to this.
It would be refreshing if the "lockdown sceptics" ever admitted error.
- There are no excess deaths from covid in March/April - The excess deaths aren't actually that high; they were sort of "owed" already - All the models were wrong, we'd have been fine if we hadn't locked down - Only oldies get it - Loads of us have already had it; we've got close to herd immunity already - Children can't get it or transmit it on - There won't be a second wave - All the positives are false positives
We're going around the cycle again, with no reference to the fact they were totally wrong last time around. As I count it, we've repeated the first five already.
With all the talk about models this morning, I did a quick'n'dirty simplistic model to try to work out how many deaths are already baked in.
First graph: Using ONS stats for infections per day in England, take an IFR of 0.65%, take date of death as infection day + 19, multiply by 1.2 for all of UK (matches existing stats pretty well), and extrapolate forward. - Red bars are deaths already recorded - Translucent orange bars are against the existing ONS stats (baked in) - Translucent grey bars are extrapolating the ONS rate of increase between the last two surveys (questionable, but the best prior we can get, really. If rate of increase has fallen, will over-predict. If spread has pushed further into vulnerable demographics, will under-predict. I expect they'll cancel out to some extent).
That's another 5,600 deaths until the lockdown kicks in. Peaking at over 900 deaths per day.
Piling assumption on assumption ("all models are wrong, but some are useful"), assuming an R of 0.9 during the new lockdown, and extrapolating forwards to the end of lockdown (which gives deaths on Christmas Eve), and with the same colour scheme:
Quite depressing. That's a total of 36,000 deaths between now and Christmas.
I really hope this is an overly pessimistic extrapolation.
Are we staying within hospital capacity? Otherwise that IFR will go up significantly.
I think so. The peak is very close to the March/April peak, and we stayed (just?) within hospital capacity then. It does imply that they acted not a moment too soon. Another week (if - and it's a big if - the virus spread was continuing at that rate) and we'd have gone over.
Will be interesting to see how reality compares to this prediction. Can we expect updates?
Sure. I can do it daily, but the death figures are usually fairly laggy. Might make more sense to update it on Thursdays and Mondays. And I'll be hoping the model is overpredicting.
Friday (when the ONS survey is updated) and Tuesday (when the highest number of backdated deaths are normally reported) would seem the obvious days if possible.
Good point. I'll update it every Thursday after the ONS survey is updated. This will cause the translucent grey bars to alter; in order to have the ability to measure against the earlier versions, I'll actually add a separate series for "following new ONS survey data," when it comes in.
How does it feel to be the new owner of the "example scenario"?
Yay! Mrs RP reports that two of her teaching colleagues now have Covid. Does make me laugh how people are flapping about "lockdown" when its nothing of the sort
I still remember the Daly Mail/Daily express comments from July/August about how safe it would be in schools when the pupils went back. I remember them saying that students don't transmit to each other so it's all right. When challenged about infecting teachers they said "the jury is still out on that one". Outrageous. How many teachers, parents and grannies have died due to this.
It would be refreshing if the "lockdown sceptics" ever admitted error.
- There are no excess deaths from covid in March/April - The excess deaths aren't actually that high; they were sort of "owed" already - All the models were wrong, we'd have been fine if we hadn't locked down - Only oldies get it - Loads of us have already had it; we've got close to herd immunity already - Children can't get it or transmit it on - There won't be a second wave - All the positives are false positives
We're going around the cycle again, with no reference to the fact they were totally wrong last time around. As I count it, we've repeated the first five already.
I'm sure I will get no credit for saying a lockdown was inevitable and calling for it weeks ago, instead my 2019 prediction will be brought up instead.
With all the talk about models this morning, I did a quick'n'dirty simplistic model to try to work out how many deaths are already baked in.
First graph: Using ONS stats for infections per day in England, take an IFR of 0.65%, take date of death as infection day + 19, multiply by 1.2 for all of UK (matches existing stats pretty well), and extrapolate forward. - Red bars are deaths already recorded - Translucent orange bars are against the existing ONS stats (baked in) - Translucent grey bars are extrapolating the ONS rate of increase between the last two surveys (questionable, but the best prior we can get, really. If rate of increase has fallen, will over-predict. If spread has pushed further into vulnerable demographics, will under-predict. I expect they'll cancel out to some extent).
That's another 5,600 deaths until the lockdown kicks in. Peaking at over 900 deaths per day.
Piling assumption on assumption ("all models are wrong, but some are useful"), assuming an R of 0.9 during the new lockdown, and extrapolating forwards to the end of lockdown (which gives deaths on Christmas Eve), and with the same colour scheme:
Quite depressing. That's a total of 36,000 deaths between now and Christmas.
I really hope this is an overly pessimistic extrapolation.
Are we staying within hospital capacity? Otherwise that IFR will go up significantly.
I think so. The peak is very close to the March/April peak, and we stayed (just?) within hospital capacity then. It does imply that they acted not a moment too soon. Another week (if - and it's a big if - the virus spread was continuing at that rate) and we'd have gone over.
Will be interesting to see how reality compares to this prediction. Can we expect updates?
Sure. I can do it daily, but the death figures are usually fairly laggy. Might make more sense to update it on Thursdays and Mondays. And I'll be hoping the model is overpredicting.
Friday (when the ONS survey is updated) and Tuesday (when the highest number of backdated deaths are normally reported) would seem the obvious days if possible.
Good point. I'll update it every Thursday after the ONS survey is updated. This will cause the translucent grey bars to alter; in order to have the ability to measure against the earlier versions, I'll actually add a separate series for "following new ONS survey data," when it comes in.
Thanks Andy – can you link the images rather than embedding them though?
With all the talk about models this morning, I did a quick'n'dirty simplistic model to try to work out how many deaths are already baked in.
First graph: Using ONS stats for infections per day in England, take an IFR of 0.65%, take date of death as infection day + 19, multiply by 1.2 for all of UK (matches existing stats pretty well), and extrapolate forward. - Red bars are deaths already recorded - Translucent orange bars are against the existing ONS stats (baked in) - Translucent grey bars are extrapolating the ONS rate of increase between the last two surveys (questionable, but the best prior we can get, really. If rate of increase has fallen, will over-predict. If spread has pushed further into vulnerable demographics, will under-predict. I expect they'll cancel out to some extent).
That's another 5,600 deaths until the lockdown kicks in. Peaking at over 900 deaths per day.
Piling assumption on assumption ("all models are wrong, but some are useful"), assuming an R of 0.9 during the new lockdown, and extrapolating forwards to the end of lockdown (which gives deaths on Christmas Eve), and with the same colour scheme:
Quite depressing. That's a total of 36,000 deaths between now and Christmas.
I really hope this is an overly pessimistic extrapolation.
Are we staying within hospital capacity? Otherwise that IFR will go up significantly.
I think so. The peak is very close to the March/April peak, and we stayed (just?) within hospital capacity then. It does imply that they acted not a moment too soon. Another week (if - and it's a big if - the virus spread was continuing at that rate) and we'd have gone over.
Will be interesting to see how reality compares to this prediction. Can we expect updates?
Sure. I can do it daily, but the death figures are usually fairly laggy. Might make more sense to update it on Thursdays and Mondays. And I'll be hoping the model is overpredicting.
Friday (when the ONS survey is updated) and Tuesday (when the highest number of backdated deaths are normally reported) would seem the obvious days if possible.
Good point. I'll update it every Thursday after the ONS survey is updated. This will cause the translucent grey bars to alter; in order to have the ability to measure against the earlier versions, I'll actually add a separate series for "following new ONS survey data," when it comes in.
Thanks Andy – can you link the images rather than embedding them though?
I think images are fine, it's just the tweets which seems to be slowing things down on the iPhone.
Mr. Max, I posted on the preceding thread that it seemed reminiscent of bread knives and medical students.
In Philip Matyszak's Classical Compendium there's an anecdote of a trial in Athens. The accused woman is disrobed by her lawyer, who remarks that the court cannot possibly find someone so clearly a handmaiden of Aphrodite to be guilty.
She was found innocent.
The audio that's in the public domain does make the verdict somewhat surprising to me, though I haven't been following the ins and outs.
That Early Voting blog looks like very bad news for Biden.
Am I missing something?
I think you are overreacting a bit.
I think it points to a Biden narrowish win mainly because in some Battleground states there are too few OTDV still to come and overall as long as most of the 31m" not yet counted" are counted that will boost Bidens EV lead further
FL for example is close but with a current 995k lead in EV and if the rest of EV come in that could go up by a further 200k
So Trump has to claw back 1.2m on election day. So if we assume he gets a 75% to 25% advantage he would need 2.4m voters tomorrow and then he would need to rely on the no party affiliation breaking evenly within the EV
I think Biden is very slight favourite in Florida
Not that there is a lot in it though.
I personally think overall i wouldnt bet on Biden to be President at anything must shorter than the currently available 1.5
With all the talk about models this morning, I did a quick'n'dirty simplistic model to try to work out how many deaths are already baked in.
First graph: Using ONS stats for infections per day in England, take an IFR of 0.65%, take date of death as infection day + 19, multiply by 1.2 for all of UK (matches existing stats pretty well), and extrapolate forward. - Red bars are deaths already recorded - Translucent orange bars are against the existing ONS stats (baked in) - Translucent grey bars are extrapolating the ONS rate of increase between the last two surveys (questionable, but the best prior we can get, really. If rate of increase has fallen, will over-predict. If spread has pushed further into vulnerable demographics, will under-predict. I expect they'll cancel out to some extent).
That's another 5,600 deaths until the lockdown kicks in. Peaking at over 900 deaths per day.
Piling assumption on assumption ("all models are wrong, but some are useful"), assuming an R of 0.9 during the new lockdown, and extrapolating forwards to the end of lockdown (which gives deaths on Christmas Eve), and with the same colour scheme:
Quite depressing. That's a total of 36,000 deaths between now and Christmas.
I really hope this is an overly pessimistic extrapolation.
Are we staying within hospital capacity? Otherwise that IFR will go up significantly.
I think so. The peak is very close to the March/April peak, and we stayed (just?) within hospital capacity then. It does imply that they acted not a moment too soon. Another week (if - and it's a big if - the virus spread was continuing at that rate) and we'd have gone over.
Will be interesting to see how reality compares to this prediction. Can we expect updates?
Sure. I can do it daily, but the death figures are usually fairly laggy. Might make more sense to update it on Thursdays and Mondays. And I'll be hoping the model is overpredicting.
Friday (when the ONS survey is updated) and Tuesday (when the highest number of backdated deaths are normally reported) would seem the obvious days if possible.
Good point. I'll update it every Thursday after the ONS survey is updated. This will cause the translucent grey bars to alter; in order to have the ability to measure against the earlier versions, I'll actually add a separate series for "following new ONS survey data," when it comes in.
Good work! I have my fingers crossed that the school half-term holiday may have helped to reduce R somewhat already, so hopefully the effect of this will become evident before the effect of the lockdown itself.
Farage has judged this wrong in terms of mass appeal among those that don't hate his guts. Oldies are scared and want to be locked away. It is mostly the youngsters, who believe he is a massive racist, that are willing to entertain the idea of just getting on it, as lockdowns are destroying their jobs and prospects.
Hmmm, not sure about this, I guess it depends on how you define "old". The sixty and seventy somethings I know aren't too keen on lockdown.
You need to stay in more like most old people right now.
The final result is probably the most telling. Among people who dislike both candidates, Trump had a clear advantage over Clinton in 2016, but now it's flipped to Biden.
Quick shout out to @HYUFD . (1) Yes we’ve seen the Trafalgar Michigan Poll, please don’t embed a tweet in a post as it slows the site from loading, (2) no, no one with a brain is saying Trump can’t win and (3) we take Nate Silver seriously because he deals in odds and probabilities and this is a gambling site he’s not a pollster - even he takes Trafalgar into account. Hope that clears things up for you.
Thanks. The site is absolutely unusable on the iPhone when people embed tweets.
Surely that is the kind of enhanced user experience which prompts iFans to pay £lots for handsets and £more for special expensive tariffs.
With all the talk about models this morning, I did a quick'n'dirty simplistic model to try to work out how many deaths are already baked in.
First graph: Using ONS stats for infections per day in England, take an IFR of 0.65%, take date of death as infection day + 19, multiply by 1.2 for all of UK (matches existing stats pretty well), and extrapolate forward. - Red bars are deaths already recorded - Translucent orange bars are against the existing ONS stats (baked in) - Translucent grey bars are extrapolating the ONS rate of increase between the last two surveys (questionable, but the best prior we can get, really. If rate of increase has fallen, will over-predict. If spread has pushed further into vulnerable demographics, will under-predict. I expect they'll cancel out to some extent).
That's another 5,600 deaths until the lockdown kicks in. Peaking at over 900 deaths per day.
Piling assumption on assumption ("all models are wrong, but some are useful"), assuming an R of 0.9 during the new lockdown, and extrapolating forwards to the end of lockdown (which gives deaths on Christmas Eve), and with the same colour scheme:
Quite depressing. That's a total of 36,000 deaths between now and Christmas.
I really hope this is an overly pessimistic extrapolation.
Are we staying within hospital capacity? Otherwise that IFR will go up significantly.
I think so. The peak is very close to the March/April peak, and we stayed (just?) within hospital capacity then. It does imply that they acted not a moment too soon. Another week (if - and it's a big if - the virus spread was continuing at that rate) and we'd have gone over.
Will be interesting to see how reality compares to this prediction. Can we expect updates?
Sure. I can do it daily, but the death figures are usually fairly laggy. Might make more sense to update it on Thursdays and Mondays. And I'll be hoping the model is overpredicting.
Friday (when the ONS survey is updated) and Tuesday (when the highest number of backdated deaths are normally reported) would seem the obvious days if possible.
Good point. I'll update it every Thursday after the ONS survey is updated. This will cause the translucent grey bars to alter; in order to have the ability to measure against the earlier versions, I'll actually add a separate series for "following new ONS survey data," when it comes in.
How does it feel to be the new owner of the "example scenario"?
Weird. Especially as I'm keen for it to be wrong!
Imperial now saying IFR for high income countries is 1.15% (95% prediction interval 0.78-1.79).
Labour seem to want to focus on Rishi as he may succeed Boris sooner than later
Also notice Starmer did not reference his 8% proposed corporation tax increase to the CBI
It's a possibility, though it seems like a decent attack line anyway.
Rishi is about the only senior Conservative with a good public image right now. Partly, he does seem genuinely good for a young'un, partly because his shiny image is still shiny (though I can see him becoming more than faintly absurd if he keeps the Insta-vibe going much longer). But mostly because, he hosed money around like a drunken sailor. The right thing to do, but also easy popularity.
But was that the real Rishi? He is, after all, a rich man who worked for a hedge fund. The government has been considerably less generous during the time of tiers. I don't think he's denied that he doesn't like the cost of Lockdown 2. If the opposition can convert Dishy Rishi into Stingy Sunak in the public's mind, that seems worth their while.
All political parties have members who are good people and members who are not. Rishi stands out from so many of current front line Tories because he is a decent human being. Yes he's a Conservative and is having to filter events through that world view. But he has thrown caution to the wind and thrown oceans of cash at people in ways that were almost inconceivable for a Tory chancellor to do.
He - and the Treasury - have also handled things competently - a skill sorely missing on the government benches.
I've just been for a Covid test myself. I don't have any of the typical symptoms, but I admitted to the Covid-19 Symptom Tracker app that I'd had an unusually long-lasting headache over the weekend and it sent me for a test. Presumably this is intended to determine whether headaches could be a possible indicator of Covid!
What is problematic for the Dems is that less doorstepping stuff hasn't been done, reportedly, because of Covid whilst the GOP has been at it for some time. As a result, there is a scramble on the Dem side for last minute canvassing in key states.
Here's a reality check again for the nervous.
There is
a) that transition period which should see out any contesting. b) the way the electoral college works if its close enough might end up being to Biden's advantage due to the faithless electors c) Biden already looks to have 2 of the 3 mid west states he needs in the bag and his options after that are several to get the 1 (largely seen to be PA) or 2 states/distributions from split states he needs . It is going to be some failure in his campaign and the pollsters d) If Biden's popular vote lead is anywhere near the numbers suggested, not only will Trump not win via the Electoral College, he wouldn't carry his own party very far in disputing the results. There is a point where they will get off the train. e) The situation in Congress may well have changed to a Democratic hold on both Houses, they will just cripple Trump even if he did somehow hold on
I have made the claim that perhaps we should ignore Florida, we should ignore Texas. Instead Mid West & South West is where its all at. I will admit that I think Trump will do better than the polls but not by enough. Enough of those who gave him the benefit of the doubt in 2016 are fed up with the man to have him put out.
Yay! Mrs RP reports that two of her teaching colleagues now have Covid. Does make me laugh how people are flapping about "lockdown" when its nothing of the sort
I still remember the Daly Mail/Daily express comments from July/August about how safe it would be in schools when the pupils went back. I remember them saying that students don't transmit to each other so it's all right. When challenged about infecting teachers they said "the jury is still out on that one". Outrageous. How many teachers, parents and grannies have died due to this.
Just to add to the fun my eldest starts his work experience tomorrow in a primary school...
I wish him well for his career. I'm viewing this from the other end of my career....
With all the talk about models this morning, I did a quick'n'dirty simplistic model to try to work out how many deaths are already baked in.
First graph: Using ONS stats for infections per day in England, take an IFR of 0.65%, take date of death as infection day + 19, multiply by 1.2 for all of UK (matches existing stats pretty well), and extrapolate forward. - Red bars are deaths already recorded - Translucent orange bars are against the existing ONS stats (baked in) - Translucent grey bars are extrapolating the ONS rate of increase between the last two surveys (questionable, but the best prior we can get, really. If rate of increase has fallen, will over-predict. If spread has pushed further into vulnerable demographics, will under-predict. I expect they'll cancel out to some extent).
That's another 5,600 deaths until the lockdown kicks in. Peaking at over 900 deaths per day.
Piling assumption on assumption ("all models are wrong, but some are useful"), assuming an R of 0.9 during the new lockdown, and extrapolating forwards to the end of lockdown (which gives deaths on Christmas Eve), and with the same colour scheme:
Quite depressing. That's a total of 36,000 deaths between now and Christmas.
I really hope this is an overly pessimistic extrapolation.
Are we staying within hospital capacity? Otherwise that IFR will go up significantly.
I think so. The peak is very close to the March/April peak, and we stayed (just?) within hospital capacity then. It does imply that they acted not a moment too soon. Another week (if - and it's a big if - the virus spread was continuing at that rate) and we'd have gone over.
Will be interesting to see how reality compares to this prediction. Can we expect updates?
Sure. I can do it daily, but the death figures are usually fairly laggy. Might make more sense to update it on Thursdays and Mondays. And I'll be hoping the model is overpredicting.
Friday (when the ONS survey is updated) and Tuesday (when the highest number of backdated deaths are normally reported) would seem the obvious days if possible.
Good point. I'll update it every Thursday after the ONS survey is updated. This will cause the translucent grey bars to alter; in order to have the ability to measure against the earlier versions, I'll actually add a separate series for "following new ONS survey data," when it comes in.
Good work! I have my fingers crossed that the school half-term holiday may have helped to reduce R somewhat already, so hopefully the effect of this will become evident before the effect of the lockdown itself.
The Halloween parties in the woods on Saturday followed by a return to Sixth Form today doesn't fill me with buckets of confidence in that regard.
The final result is probably the most telling. Among people who dislike both candidates, Trump had a clear advantage over Clinton in 2016, but now it's flipped to Biden.
What about those who like/ liked both candid... Oh. I see my mistake.
Two households will be able to form a bubble in Wales after the firebreak lockdown ends on 9 November, First Minister Mark Drakeford said.
There will be no travel restrictions in Wales, but people will not be able to leave the country without a reasonable excuse during the English lockdown. Groups of 15 will be allowed to meet indoors, and groups of 30 outdoors.
15 people...15.....indoors....that seems more unwise than banning Tesco's from selling oven gloves. And 30 outdoors, that just prime for gangs of youth to congregate.
I will admit that I think Trump will do better than the polls but not by enough. Enough of those who gave him the benefit of the doubt in 2016 are fed up with the man to have him put out.
I think too much attention is given to the marginal Trump voter in 2016 and not enough to the marginal non-voter. Did they spend the last 4 years thinking: "Trump is a disaster; I should have voted," or: "The guy won fair and square and the establishment is still fighting him tooth and nail"?
Labour seem to want to focus on Rishi as he may succeed Boris sooner than later
Also notice Starmer did not reference his 8% proposed corporation tax increase to the CBI
It's a possibility, though it seems like a decent attack line anyway.
Rishi is about the only senior Conservative with a good public image right now. Partly, he does seem genuinely good for a young'un, partly because his shiny image is still shiny (though I can see him becoming more than faintly absurd if he keeps the Insta-vibe going much longer). But mostly because, he hosed money around like a drunken sailor. The right thing to do, but also easy popularity.
But was that the real Rishi? He is, after all, a rich man who worked for a hedge fund. The government has been considerably less generous during the time of tiers. I don't think he's denied that he doesn't like the cost of Lockdown 2. If the opposition can convert Dishy Rishi into Stingy Sunak in the public's mind, that seems worth their while.
All political parties have members who are good people and members who are not. Rishi stands out from so many of current front line Tories because he is a decent human being. Yes he's a Conservative and is having to filter events through that world view. But he has thrown caution to the wind and thrown oceans of cash at people in ways that were almost inconceivable for a Tory chancellor to do.
Absolutely. Rishi is an impressive politician, decent human, firmly in a part of the Conservative tradition. Furlough was a huge achievement.
But since than, there have been definite missteps by the government, mostly quibbling over tiny amounts of cash- FSM and Tier 3 funding for example. The government does seem to have overdone "back to normal" in a way that has Treasury fingerprints all over it. If Rishi did try to avoid or delay Lockdown 2, that might be costly in lives and money.
Rishi is good politician and a good person. But not above criticism.
CBS has Trump winning 279 to 259 if there is a surge of Trump voters on the day, Biden wins 375 to 163 if early voters dominate and otherwise it is Biden 279 and Trump 163 and 96 toss ups
Two households will be able to form a bubble in Wales after the firebreak lockdown ends on 9 November, First Minister Mark Drakeford said.
There will be no travel restrictions in Wales, but people will not be able to leave the country without a reasonable excuse during the English lockdown. Groups of 15 will be allowed to meet indoors, and groups of 30 outdoors.
15 people...15.....indoors....that seems more unwise than banning Tesco's from selling oven gloves.
Presumably "Drakeford's Stasi" will be put on furlough.
Applying 160m from the Early Voting Blog and assuming 100m are EV The numbers are
Biden EV 66M OTD 16.2M Total 82.2M
Trump EV 32M OTD 41.4M Total 73.4M
If true and Biden doesnt win EC the system is completely fucked
Worth noting that Trump added only 2.1 million votes to Romney's 2012 total, so for him to add 10.4 million votes this time would be consistent with the signs that he has pleased a lot of people.
It will also show what the challenge will be for the Democrats in future elections.
Yay! Mrs RP reports that two of her teaching colleagues now have Covid. Does make me laugh how people are flapping about "lockdown" when its nothing of the sort
I still remember the Daly Mail/Daily express comments from July/August about how safe it would be in schools when the pupils went back. I remember them saying that students don't transmit to each other so it's all right. When challenged about infecting teachers they said "the jury is still out on that one". Outrageous. How many teachers, parents and grannies have died due to this.
It would be refreshing if the "lockdown sceptics" ever admitted error.
- There are no excess deaths from covid in March/April - The excess deaths aren't actually that high; they were sort of "owed" already - All the models were wrong, we'd have been fine if we hadn't locked down - Only oldies get it - Loads of us have already had it; we've got close to herd immunity already - Children can't get it or transmit it on - There won't be a second wave - All the positives are false positives
We're going around the cycle again, with no reference to the fact they were totally wrong last time around. As I count it, we've repeated the first five already.
The problem is that people see *their* set of problems, and want to fit the reality to *their* solution to *their* problems.
Saying "I don't know, exactly" is the truth, but it's not a truth that is easy to sell.
Quick shout out to @HYUFD . (1) Yes we’ve seen the Trafalgar Michigan Poll, please don’t embed a tweet in a post as it slows the site from loading, (2) no, no one with a brain is saying Trump can’t win and (3) we take Nate Silver seriously because he deals in odds and probabilities and this is a gambling site he’s not a pollster - even he takes Trafalgar into account. Hope that clears things up for you.
Thanks. The site is absolutely unusable on the iPhone when people embed tweets.
There are umpteen tweets on this site not just from me and if posting tweets stops the site from working for some on their phones that is a matter for the site owners and Vanilla not me
CBS has Trump winning 279 to 259 if there is a surge of Trump voters on the day, Biden wins 375 to 163 if early voters dominate and otherwise it is Biden 279 and Trump 163 and 96 toss ups
Farage has judged this wrong in terms of mass appeal among those that don't hate his guts. Oldies are scared and want to be locked away. It is mostly the youngsters, who believe he is a massive racist, that are willing to entertain the idea of just getting on it, as lockdowns are destroying their jobs and prospects.
Really the LibDems should be on this one. I don’t see they have anything to gain by joining Starmer in pushing the government to lockdown sooner and further
Arguing for more deaths and more economic damage might weigh a bit more heavily on their conscience than producing dodgy bar charts though. No, this is one for the far left and right.
They're in Opposition. They can just hammer the government over and over on its failure to isolate the infectious and say the lockdown is the price of that failure.
True. They can also argue over which measures are worth doing and which aren't, and over whether they should be nationwide or local, and of course how it's all paid for. But arguing against the principle of controlling the virus is one for the fringes.
Labour seem to want to focus on Rishi as he may succeed Boris sooner than later
Also notice Starmer did not reference his 8% proposed corporation tax increase to the CBI
It's a possibility, though it seems like a decent attack line anyway.
Rishi is about the only senior Conservative with a good public image right now. Partly, he does seem genuinely good for a young'un, partly because his shiny image is still shiny (though I can see him becoming more than faintly absurd if he keeps the Insta-vibe going much longer). But mostly because, he hosed money around like a drunken sailor. The right thing to do, but also easy popularity.
But was that the real Rishi? He is, after all, a rich man who worked for a hedge fund. The government has been considerably less generous during the time of tiers. I don't think he's denied that he doesn't like the cost of Lockdown 2. If the opposition can convert Dishy Rishi into Stingy Sunak in the public's mind, that seems worth their while.
All political parties have members who are good people and members who are not. Rishi stands out from so many of current front line Tories because he is a decent human being. Yes he's a Conservative and is having to filter events through that world view. But he has thrown caution to the wind and thrown oceans of cash at people in ways that were almost inconceivable for a Tory chancellor to do.
Absolutely. Rishi is an impressive politician, decent human, firmly in a part of the Conservative tradition. Furlough was a huge achievement.
But since than, there have been definite missteps by the government, mostly quibbling over tiny amounts of cash- FSM and Tier 3 funding for example. The government does seem to have overdone "back to normal" in a way that has Treasury fingerprints all over it. If Rishi did try to avoid or delay Lockdown 2, that might be costly in lives and money.
Rishi is good politician and a good person. But not above criticism.
Particularly the extension of furlough only at the very last minute. This has cost real people their jobs. If there was a likelihood of further restrictions, and there very much was, an extension till Christmas could have been announced weeks ago.
CBS has Trump winning 279 to 259 if there is a surge of Trump voters on the day, Biden wins 375 to 163 if early voters dominate and otherwise it is Biden 279 and Trump 163 and 96 toss ups
Yay! Mrs RP reports that two of her teaching colleagues now have Covid. Does make me laugh how people are flapping about "lockdown" when its nothing of the sort
I still remember the Daly Mail/Daily express comments from July/August about how safe it would be in schools when the pupils went back. I remember them saying that students don't transmit to each other so it's all right. When challenged about infecting teachers they said "the jury is still out on that one". Outrageous. How many teachers, parents and grannies have died due to this.
Just to add to the fun my eldest starts his work experience tomorrow in a primary school...
I wish him well for his career. I'm viewing this from the other end of my career....
CBS has Trump winning 279 to 259 if there is a surge of Trump voters on the day, Biden wins 375 to 163 if early voters dominate and otherwise it is Biden 279 and Trump 163 and 96 toss ups
I will admit that I think Trump will do better than the polls but not by enough. Enough of those who gave him the benefit of the doubt in 2016 are fed up with the man to have him put out.
I think too much attention is given to the marginal Trump voter in 2016 and not enough to the marginal non-voter. Did they spend the last 4 years thinking: "Trump is a disaster; I should have voted," or: "The guy won fair and square and the establishment is still fighting him tooth and nail"?
I think there was a large number of marginal non-voters in 2016, on the abstain/Cinton border, who are thinking "Trump is a disaster; I should have voted" On one State (IIRC Wisconsin) Trump had less votes than Romney but the democrats lost even more between 2012 and 2016.
I do love that "Morning Consult". Not just the consistent (and correct) TrumpToast polling but the name is so appealing. Morning Consult. Lovely positive images and sensations are evoked. The sun popping up over the horizon of a deep blue sea. I can see clearly now. I can see for miles and miles. Exactly what polls are supposed to enable.
Compare and contrast to "Trafalgar". Not nice at all. An ugly feeling, claustrophobic and tense. You think grim pub, or an exposed and crowded square, or perhaps a battle, either way it's all noise and smoke and confusion.
Sturgeon now seriously considering following England into lockdown, because even though the current tiers have shown signs of working, she cannot politically afford case numbers getting worse than England for December and consequently having to impose stricter restrictions for Christmas.
I will admit that I think Trump will do better than the polls but not by enough. Enough of those who gave him the benefit of the doubt in 2016 are fed up with the man to have him put out.
I think too much attention is given to the marginal Trump voter in 2016 and not enough to the marginal non-voter. Did they spend the last 4 years thinking: "Trump is a disaster; I should have voted," or: "The guy won fair and square and the establishment is still fighting him tooth and nail"?
Its probably a bit of both given the suggestion on turnout this time but Trump has shed some who did give him the vote last time or maybe went somewhere else other than Democrat in 2016 but are going that way in 2020. So far I haven't seen anything to suggest he has restocked those loses somewhere else. That alone looks enough to shift the race.
Two households will be able to form a bubble in Wales after the firebreak lockdown ends on 9 November, First Minister Mark Drakeford said.
There will be no travel restrictions in Wales, but people will not be able to leave the country without a reasonable excuse during the English lockdown. Groups of 15 will be allowed to meet indoors, and groups of 30 outdoors.
15 people...15.....indoors....that seems more unwise than banning Tesco's from selling oven gloves. And 30 outdoors, that just prime for gangs of youth to congregate.
It is totally reckless on those numbers. And way too soon for that level of relaxation.
CBS has Trump winning 279 to 259 if there is a surge of Trump voters on the day, Biden wins 375 to 163 if early voters dominate and otherwise it is Biden 279 and Trump 163 and 96 toss ups
I like how you have buried the better Biden news behind a Trump victory.
I see from that link in the Trump surge he wins PA but not WI or MI to squeak home. This is the only way I see him winning,. Technically he has other paths but realistically he needs to hold AZ and PA (both unlikely but possible) and then all the other states from 2016 bar MI and WI. A tough ask but to me this is the only way he wins
I will admit that I think Trump will do better than the polls but not by enough. Enough of those who gave him the benefit of the doubt in 2016 are fed up with the man to have him put out.
I think too much attention is given to the marginal Trump voter in 2016 and not enough to the marginal non-voter. Did they spend the last 4 years thinking: "Trump is a disaster; I should have voted," or: "The guy won fair and square and the establishment is still fighting him tooth and nail"?
Its probably a bit of both given the suggestion on turnout this time but Trump has shed some who did give him the vote last time or maybe went somewhere else other than Democrat in 2016 but are going that way in 2020. So far I haven't seen anything to suggest he has restocked those loses somewhere else. That alone looks enough to shift the race.
There will be obviously be some churn as I can't see Trump outside the 44-46% range.
There will also be people like Ben Shapiro who didn't vote Trump in 2016 but are voting for him now as he has served the Republican establishment after all.
CBS has Trump winning 279 to 259 if there is a surge of Trump voters on the day, Biden wins 375 to 163 if early voters dominate and otherwise it is Biden 279 and Trump 163 and 96 toss ups
Yay! Mrs RP reports that two of her teaching colleagues now have Covid. Does make me laugh how people are flapping about "lockdown" when its nothing of the sort
I still remember the Daly Mail/Daily express comments from July/August about how safe it would be in schools when the pupils went back. I remember them saying that students don't transmit to each other so it's all right. When challenged about infecting teachers they said "the jury is still out on that one". Outrageous. How many teachers, parents and grannies have died due to this.
It would be refreshing if the "lockdown sceptics" ever admitted error.
- There are no excess deaths from covid in March/April - The excess deaths aren't actually that high; they were sort of "owed" already - All the models were wrong, we'd have been fine if we hadn't locked down - Only oldies get it - Loads of us have already had it; we've got close to herd immunity already - Children can't get it or transmit it on - There won't be a second wave - All the positives are false positives
We're going around the cycle again, with no reference to the fact they were totally wrong last time around. As I count it, we've repeated the first five already.
Not under the lockdown sceptics but similar going round in circles. - We'll have a vaccine by October - We'll have a vaccine by the end of the year - We'll have a vaccine by March.
And - "We can only tell how bad the virus is by looking at excess deaths". When the awful graph of excess deaths in March/April came out "But they are deaths of people who would have died this year anyway, we can only tell how bad the virus is by looking at excess deaths for the whole of 2020"
CBS has Trump winning 279 to 259 if there is a surge of Trump voters on the day, Biden wins 375 to 163 if early voters dominate and otherwise it is Biden 279 and Trump 163 and 96 toss ups
Shoudn't we be basing projections only on Trafalgar?
Trafalgar does mass polls aimed at identifying Trump voters others miss, especially in rural areas and who are otherwise politically disengaged
Real pollsters try to find out what people are thinking and look to find a representative sample.
Trafalgar (if they even poll anyone which seems implausible) claim to go looking for a specific result.
Trafalgar are not a pollster.
Aside from the strong doubts about their polling, Trafalgar might be taken a little more seriously if Cahaly didnt go on TV saying things like 'Dems will only win PA if there is voter fraud'. That just makes him sound even less like a pollster (even a biased one) than a Trump Acolyte
Quick shout out to @HYUFD . (1) Yes we’ve seen the Trafalgar Michigan Poll, please don’t embed a tweet in a post as it slows the site from loading, (2) no, no one with a brain is saying Trump can’t win and (3) we take Nate Silver seriously because he deals in odds and probabilities and this is a gambling site he’s not a pollster - even he takes Trafalgar into account. Hope that clears things up for you.
Thanks. The site is absolutely unusable on the iPhone when people embed tweets.
Further to our Javascript discussion last night, I discovered that the system I use for blocking PB's embedded tweets with Firefox on my PC now works on my Android phone. Given that Firefox is available on the iPhone I suspect it may work for you.
Install the Firefox browser then add the uBlock Origin adblocker add-on. Then go to uBlock Origin's settings and add a filter to block platform.twitter.com as below and you may find that it works for you. If it does, then this site loads fine, but the Twitter links don't automatically open (which is the problem)...
Biden +6 in FL and +9 in PA look very positive for him. Two quite different states that Trump must win well outside MoE. Admittedly Morning Consult is only rated B/C by 538 but encouraging...
Both those states you mention look too high. I think HYUFD posted some polls earlier (not just Trafalgars) from mid rated posters with much closer margins than that. But the trend does seem to suggest overall FL is very close and PA is fairly safe for Biden.
Stop looking at single polls and look at weighted average polls, such as in 538. You can argue about exactly how the weighting is done, but almost any sensible analysis will be more reliable than taking just one poll (that is until the benefit of hindsight becomes available to us) 538 has: FLA Biden up 2.1 PA Biden up 5.1
If Biden really is 9% ahead in PA then he will surely be leading at the end of counting on Tuesday night given Philadelphia and Pittsburgh are counting mail ballots on the night.
Comments
NC is interesting and does seem as if Trump has a better than even chance of holding on there, I did have that in his column but this post confirms my view. FL too close to call yes but expecting Trump to hold on there too.
AZ I think is flipping and the figures here support that somewhat. Still not convinced Iowa will flip so I'm keeping that in my red column too.
Am I missing something?
Rishi is about the only senior Conservative with a good public image right now. Partly, he does seem genuinely good for a young'un, partly because his shiny image is still shiny (though I can see him becoming more than faintly absurd if he keeps the Insta-vibe going much longer). But mostly because, he hosed money around like a drunken sailor. The right thing to do, but also easy popularity.
But was that the real Rishi? He is, after all, a rich man who worked for a hedge fund. The government has been considerably less generous during the time of tiers. I don't think he's denied that he doesn't like the cost of Lockdown 2. If the opposition can convert Dishy Rishi into Stingy Sunak in the public's mind, that seems worth their while.
How things are seen afterwards still depends.
Without one of those, his path to the white house is very much narrower.
https://apnews.com/article/7c0fd96f145930d1b487531c2feae60b
But, of course, Heard wasn't the one who brought a disastrous libel case. The focus is on Depp's conduct because the newspaper allegations were about him, and his conduct. The judge implies in the judgment that Heard probably was sleeping around, for example, but correctly points out that he doesn't need to make a determination on that because it wouldn't justify Depp's conduct even if she was.
That's the danger of libel cases. The simple rule is, if you have skeletons in the closet, even if you feel you've got right on your side, don't even think about risking it.
This will cause the translucent grey bars to alter; in order to have the ability to measure against the earlier versions, I'll actually add a separate series for "following new ONS survey data," when it comes in.
And the EC margin to be around 150.
The court wasn't deciding whether Heard or Depp was the worse person in the relationship... it was deciding whether the allegations in the Sun about Depp, to the effect HE was violent towards HER on several occasions, were substantively true. And, to the civil standard, they were.
What idiocy. A car is of course much safer Covid-wise.
- There are no excess deaths from covid in March/April
- The excess deaths aren't actually that high; they were sort of "owed" already
- All the models were wrong, we'd have been fine if we hadn't locked down
- Only oldies get it
- Loads of us have already had it; we've got close to herd immunity already
- Children can't get it or transmit it on
- There won't be a second wave
- All the positives are false positives
We're going around the cycle again, with no reference to the fact they were totally wrong last time around. As I count it, we've repeated the first five already.
https://twitter.com/MorningConsult/status/1323239064284352514?s=20
In Philip Matyszak's Classical Compendium there's an anecdote of a trial in Athens. The accused woman is disrobed by her lawyer, who remarks that the court cannot possibly find someone so clearly a handmaiden of Aphrodite to be guilty.
She was found innocent.
The audio that's in the public domain does make the verdict somewhat surprising to me, though I haven't been following the ins and outs.
I think it points to a Biden narrowish win mainly because in some Battleground states there are too few OTDV still to come and overall as long as most of the 31m" not yet counted" are counted that will boost Bidens EV lead further
FL for example is close but with a current 995k lead in EV and if the rest of EV come in that could go up by a further 200k
So Trump has to claw back 1.2m on election day. So if we assume he gets a 75% to 25% advantage he would need 2.4m voters tomorrow and then he would need to rely on the no party affiliation breaking evenly within the EV
I think Biden is very slight favourite in Florida
Not that there is a lot in it though.
I personally think overall i wouldnt bet on Biden to be President at anything must shorter than the currently available 1.5
Good!
"Unprecedented levels of enthusiasm for Trump in PA" says the Democrat on the ground in PA
Hopefully its a GOTV comment but he looked concerned.
I am officially a Biden bed wetter.
Here's a reality check again for the nervous.
There is
a) that transition period which should see out any contesting.
b) the way the electoral college works if its close enough might end up being to Biden's advantage due to the faithless electors
c) Biden already looks to have 2 of the 3 mid west states he needs in the bag and his options after that are several to get the 1 (largely seen to be PA) or 2 states/distributions from split states he needs . It is going to be some failure in his campaign and the pollsters
d) If Biden's popular vote lead is anywhere near the numbers suggested, not only will Trump not win via the Electoral College, he wouldn't carry his own party very far in disputing the results. There is a point where they will get off the train.
e) The situation in Congress may well have changed to a Democratic hold on both Houses, they will just cripple Trump even if he did somehow hold on
I have made the claim that perhaps we should ignore Florida, we should ignore Texas. Instead Mid West & South West is where its all at. I will admit that I think Trump will do better than the polls but not by enough. Enough of those who gave him the benefit of the doubt in 2016 are fed up with the man to have him put out.
Just saying...
Oh. I see my mistake.
There will be no travel restrictions in Wales, but people will not be able to leave the country without a reasonable excuse during the English lockdown. Groups of 15 will be allowed to meet indoors, and groups of 30 outdoors.
15 people...15.....indoors....that seems more unwise than banning Tesco's from selling oven gloves. And 30 outdoors, that just prime for gangs of youth to congregate.
But since than, there have been definite missteps by the government, mostly quibbling over tiny amounts of cash- FSM and Tier 3 funding for example. The government does seem to have overdone "back to normal" in a way that has Treasury fingerprints all over it. If Rishi did try to avoid or delay Lockdown 2, that might be costly in lives and money.
Rishi is good politician and a good person. But not above criticism.
https://www.cbsnews.com/news/election-poll-biden-trump-vote-11-01-2020/
Trump 4.6m Clinton 4.5m = 9.1m
So if FL increases by the same proportion as the 135m to 160m nationally ie 18.5%
Would mean 10.8m in FL
Currently showing 8.7 M voted with 1m EV showing as not returned.
So if the majority of those 1m are counted only just over 1.1m votes to come OTD
Too big an ask for Trump if the numbers pan out like that
A big if BUT......
DYOR
It will also show what the challenge will be for the Democrats in future elections.
Saying "I don't know, exactly" is the truth, but it's not a truth that is easy to sell.
This has cost real people their jobs.
If there was a likelihood of further restrictions, and there very much was, an extension till Christmas could have been announced weeks ago.
From 09:00 on Tuesday 27 October 2020 until the start of 2020 US Presidential Election (scheduled for 22:00 on Tuesday 3 November 2020 (the top-price period), we have a ‘top-price guarantee’ on 2020 Presidential Election Winner market on Joe Biden.
https://promotions.williamhill.com/en-gb/offer/top-price-guarantee-on-joe-biden-2020-us-presidential-election-winner-market
On one State (IIRC Wisconsin) Trump had less votes than Romney but the democrats lost even more between 2012 and 2016.
Compare and contrast to "Trafalgar". Not nice at all. An ugly feeling, claustrophobic and tense. You think grim pub, or an exposed and crowded square, or perhaps a battle, either way it's all noise and smoke and confusion.
There will also be people like Ben Shapiro who didn't vote Trump in 2016 but are voting for him now as he has served the Republican establishment after all.
Trafalgar (if they even poll anyone which seems implausible) claim to go looking for a specific result.
Trafalgar are not a pollster.
- We'll have a vaccine by October
- We'll have a vaccine by the end of the year
- We'll have a vaccine by March.
And
- "We can only tell how bad the virus is by looking at excess deaths". When the awful graph of excess deaths in March/April came out "But they are deaths of people who would have died this year anyway, we can only tell how bad the virus is by looking at excess deaths for the whole of 2020"
Install the Firefox browser then add the uBlock Origin adblocker add-on. Then go to uBlock Origin's settings and add a filter to block platform.twitter.com as below and you may find that it works for you. If it does, then this site loads fine, but the Twitter links don't automatically open (which is the problem)...