Friday assorted links

1. Paul Kedrosky assorted comments.

2. Thread on collider bias.

3. The Florida strategy.  Too early to say in my view, but still this piece is of interest.

4. “Magnus the platform” lets it rip.

5. America’s top spelling searches: in Virginia, they don’t know how to spell “Virginia.”  Recommended.

6. Volunteer history booming during the lockdown.

7. Mainstreaming Karl Marx.

8. Plexiglass vs. Plexiglas(TM).

9. The polity that is Dutch: grandmother ordered to delete Facebook photos.  At least they didn’t kill her.

10. A real smokescreen.  Really.

11. The new seroprevalence studies show relatively low rates of infection.

12. Fast Grants active at UC Berkeley.  Good coverage.

13. TreatEarly, a new biomedical initiative (possibly influenced by Fast Grants?), looks interesting note I have no direct knowledge of their work.


# 11 the new prevalence study is not that new. Look at Stockholm, it was completed at the end of April, and it takes 2 to 3 weeks to show antibodies. This means it was the story around Easter. It’s much higher now.
There was another sample that showed 10% in that time frame. I posted it a few days ago.

I know commenting will be slow today, because everyone’s out there drinking beer and eating hotdogs with their friends. But the responses to 3 & 10 should be amusing.

#10 Actually it wasn't that effective because smokescreens of that era were designed to be 'persistent' enough that, based on weather, the body of the cloud mass would remain intact. The masking effect of a smokescreen (of that era) is provided by it's depth, not breadth. This is predominantly why smoke on the water or land's surface produced by destroyers, massed rocket fires, or large groups of Soviet tanks (pumping diesel into their exhaust) had more effectiveness. A curtain isn't necessary, having it broader but maybe not 30 feet high is.

Modern smoke screens in military applications don't have this problem because the main use of smoke (other than a location market) is to provide other masking effects such as lasers or thermal. This is accomplished with additives that work regardless of the thickness of the screen.

I didn't know that about modern smokescreens, good to know.

The airplane-launched smokescreen depicted in the film was relevant for only a short time, if ever: post WW I (the film shows a ship with the characteristic pair of basket masts of WW I era American battleships), when ships could carry planes capable of performing that feat, and pre-radar and more importantly pre-WW II era naval attack planes.

By the time WW II was fought, if you had airplanes you were using them to first scout and then batter the other side's ships, as happened to the Bismarck, the Prince of Wales and Repulse, Pearl Harbor, etc. The interwar use of planes as spotters and smokescreen layers became distinctly secondary (I don't know if airplanes were ever actually used during the war to lay smokescreens; as EverExtruder says destroyers were used to do the job).

There were a number of technologies that were developed after WW I but were either obsolete by WW II or were never effective in the first place. Airplanes laying smokescreens. Tanks with two turrets. Submarines that could carry airplanes. Pocket battleships.

So this film comes from that narrow slice of inter-war history; the technology was unavailable prior, and obsolete soon after (or maybe it was never a practical idea, as EverExtruder explains).

Re: Florida. Maybe the covid virus is weather-affected? Indochina has also largely escaped a serious epidemic despite bordering on China and not doing a whole lot to prevent the disease from spreading. And as far as beaches go, I was skeptical that outdoor transmission was ever going to be a major danger. Back in mid March I had Facebook friends who were just shocked that I would go out biking-- they seemed to think there were viruses lurking behind every tree and bush just waiting to pounce. The real risk has always been crowded indoor spaces.

Vietnam was extremely organized. They were the first country to clear SARS too. They were already monitoring in early January and canceled Wuhan flights on Jan 24.
They had a 15 day quarantine, but contact , traced and quarantined aggressively. They wore masks pretty much ubiquitously.

>canceled Wuhan flights on Jan 24.

Nonsense. In Jan. 2020, everyone knew that it was racist to cancel fights from China.

All the best people are telling me that. So: evidence.

A non--racist Trump would have stopped all incoming flights from Asia, Europe, and banned all outgoing flights to anywhere and closed all land borders, stopping all spreading of SARS out of the US.

Today, Trump is intentionally spreading disease to nations with poor health care systems, especially south of the US, seeking out long term US residents who caught SARS as a critical worker in a critical US production facility and deporting them while still carrying the US acquired infection.

Note, Trump has blocked US citizens from returning to the US even after 30 plus days in quarantine and no positive test for SARS-Cov2, so Trump forces other nations to bear costs and risks he says the US should not bear.

Oh my God... in just these two above comments, we have a Dem saying that no one called Trump racist for banning flights from China, and another Dem calling Trump racist for banning flights from China.

We truly live in a Golden Era.

Yes, but Cambodia never banned flights from China and kept all its tourist attractions open, and is doing just as well as Vietnam.

In fact, Cambodia not only kept its border open with China, but Cambodia's leader downplayed the virus, told people not to wear masks, and welcomed cruise ships that were turned away by other countries in the region:

Still, Cambodia today has 0 active cases and 0 deaths.

I am a little skeptical of Cambodia. They had 122 cases overall and no cases since April 19. They have tested at less than 1/3 of the Vietnam rate per M.
It's also endemic for Dengue fever which has very similar symptoms ( flu-like) and you can have the Covid-19 virus and test positive for Dengue fever. This happened a few times in Singapore. I think Indonesia also may also have this Dengue fever/Covid-19 confounding. Vietnam health system is more sophisticated than Cambodia's.

The dengue and poor health systems issue would exist in Vietnam too.

Cambodia's monitoring has actually been pretty good. For example, today they reported a case from a traveler from the Philippines right away and then quarantined everybody on the plane:

Moreover, even if their numbers are understating a bit, I have not heard any evidence or even a rumor that there is any significant outbreak or cluster of cases in Cambodia.

Cambodia no doubt benefited from Vietnam’s attention to the problem. since they share a large border with them. Cambodia also monitored China flights carefully, banned cruise ships and travelers from Italy, France, Germany, Spain, US and Iran.
Vietnam has a very good public health system which was demonstrated against SARS. They isolated and cultured the virus by Feb.7. The first Covid-19 genome sequence ( WUH01) was only entered in GenBank only on Feb.11

Doesn't that go both ways though? You could also say Vietnam benefited from few cases in Cambodia.

Cambodia did have a temporary travel ban on those countries (which were more justified than a ban on the non-Wuhan parts of China, given that those countries all had significantly more cases per capita than the non-Wuhan parts of China). However, Cambodia has already lifted those restrictions: They are going for a strategy of requiring international travelers to be tested, monitoring them, and quarantining the sick and those who have been in contact with the sick, rather than banning such travelers. This seems to be a sensible strategy to me, which will be equally effective as a ban on travelers while being much less disruptive. I predict it will work just fine and Cambodia will not suffer a resurgence of the virus (or if it does, it will be because of a new mutation or something that also causes all of Cambodia's neighbors to be affected).

Also, wasn't COVID-19 sequenced and published by Chinese scientists already in early January? Are you referring to something else?

I look at it as Vietnam has about 3 X the number of international visitors as Cambodia and has a border with China. It could have been a big source of cross border infection for Cambodia as Vietnam is Cambodia's largest trade partner ( with China, then lower Thailand)
It didn't happen because Vietnam was extremely vigilant of the virus.
Yes the date is Jan 11 not Feb 11.

#3 " DeSantis and his team have followed the science closely from the beginning, which is why they forged a nuanced approach, but one that focused like a laser on the most vulnerable population, those in nursing homes"
That's the strategy ! contrast with New York

Or contrast with Finland -
Florida deaths per million - 102
Finland deaths per million - 55
Or Greece - 16, Austria - 71, Poland - 26.

OK, now do the UK, France, Spain, and Italy.

It's been amusing all this time to watch a commentariat that is allegedly so rational and even-minded resort to this sort of behavior just to confirm their priors. Cargo cult.

This is also from Worldometer. Ethiopia 0.04 deaths per M . Finland 55.
By your cherry picked comparison methods , Finland is a spectacular failure by a factor of 1375.

Cases per million / deaths -
Florida - 2,302 / 102
Finland - 1,180 / 55

Finland was picked because it mirrors Florida, pretty much in similar fashion that Germany does at 2,142 / 99. Some places have done better (with the reasonable caveat of till now) than others, using a variety of strategies. Countries with similar ratios include Poland 545 / 26 and Austria 1,826 / 71. Greece is a bit of outlier at 276 / 16, but the 16 is the lowest in Europe in the same fashion that Belgium is the highest in Europe, at 4,878 / 795. Clearly, something went very, very wrong in Belgium (just as something seems to have gone very, very right in Greece), yet it does not seem to be discussed much when talking about how to best handle this pandemic in its initial stages.

But you forgot to point out how much better Florida is doing than Sweden - 3,251 / 389. Possibly because Sweden did not effectively focus like Florida on nursing homes.

Belgium has a very aggressive way of counting deaths. Any death in a nursing home is counted as a Covid death. From their own health minister recommendation divide their death count in 2 and you are counting like other countries.
Sweden could have been more effective with their nursing homes, but their death rate achieved a lot of immunity. Stockholm has achieved herd immunity now because most children are effectively immune and add to the tally so they’re effectively done with an R of 1.5 or less. Their 7 day moving average is 1/2 from the peak on April 24 If you look at a different statistic death per M per % infected, they have done pretty well.
The worst strategy is lockdown everyone ( economically very costly ) but infect nursing homes. That’s New York and Minnesota ( 93% of Minnesota deaths are in nursing homes)

'Stockholm has achieved herd immunity now'

This is false, and that according to Sweden's own health authorities -
"Just 7.3% of Stockholm’s inhabitants had developed Covid-19 antibodies by the end of April, according to a study, raising concerns that the country’s light-touch approach to the coronavirus may not be helping it build up broad immunity. ... Sweden’s state epidemiologist, Anders Tegnell, said the antibodies figure was “a bit lower than we’d thought”, but added that it reflected the situation some weeks ago and he believed that by now “a little more than 20%” of the capital’s population had probably contracted the virus.

However, the public health agency had previously said it expected about 25% to have been infected by 1 May and Tom Britton, a maths professor who helped develop its forecasting model, said the figure from the study was surprising. ... In April, officials estimated one third of Stockholm residents would have contracted Covid-19 by early May, subsequently suggesting that the capital could reach herd immunity of between 40% and 60% by the middle of June."

They’re just being careful if you have antibodies on April 30, then you must have been infected by Easter. W’re now May 22 !
Stockholm cases peaked on April 24. Why is that if there’s no immunity and no lockdown ? They’re declining because the virus finds less and less susceptible to infect.
They’re also not including the naturally effectively immune but the evidence is piling up that children hardly spread the virus at all and don’t register positive much

> Belgium has a very aggressive way of counting deaths.

As does the US. Cough and fever and meth overdose is marked down as a covid death.

Nobody is counting covid deaths as aggressively as the US.

Also, United States as a whole is at 274 deaths/million.

And states doing better than Florida include Texas 1,849 / 51, North Carolina 2,077 / 73, Oregon 905 / 34 and Tennessee 2,776 / 46. It seems that Florida is in the broad middle, which does not say too much except when trying to make a point - even by making a contrast to the worst or the best.

Some places are hit harder, some places less (with the reasonable caveat till now).

Cases per million is an idiotic metric since it will be entirely based on testing.

There’s one statistic: deaths.

With the caveat that the death data from Germany is complete bullshit since they don’t record deaths in nursing homes as Covid unless they test and they don’t test nursing home patients if they’re symptomatic

Two of the states you mentioned are well along the path to being back to normal. One of the others had well-publicized protests a couple of weeks ago that the Very Smart And Serious People ™ here assured us would lead to a spike in cases and deaths. What is your point again?

Even if Florida is in the "broad middle" normal people will take the middle and a functioning economy over being shut down until a leftist gets back in the White House—and as New York and other communist hellholes in the Northeast show, that doesn't even mean you won't be at the top of the list for cases/deaths.

This fake pandemic is doing nothing but solidifying the case that leftists need to be evicted from the country ASAP. The most charitable reading of this debacle is that they're willing to take advantage of people's fears to scratch their itch for authoritarianism. The worst is that they're willing to put millions of Americans out of work and kill tens of thousands of nursing home patients for the same. This is sabotage and treason and should be punished as such.

12. Any private company could move much faster than Fast Grants. What that company cannot take advantage of is a tax break, in contrast to those funding Fast Grants.

Paying workers to work is a tax break.

You confuse profit with government corporate welfare which is generally conditioned on paying workers to work, like the paycheck protection loans that turn into welfare if workers are paid, for working or not. Workers paid with welfare are not tax dodge eligible just as profit is not a tax dodge.

9: It's difficult to tell from the article on this GDPR ruling, but is there a distinction like what US libel law makes between people of public interest and those who are not? Otherwise there's some definite concerns about censorship.

It involves posting pictures of children on Facebook against the direct wishes of the children's parents. Parents have rights concerning their children, and the person not following the request of the parents needed to be taught that reality by a court. Whose penalties do not seem especially stiff, but mark a clear signal. Though the common sense perspective at the end of the article should have been enough. "I think the ruling will surprise a lot of people who probably don't think too much before they tweet or post photos," said Neil Brown, a technology lawyer at Decoded Legal.

"Irrespective of the legal position, would it be reasonable for the people who've posted those photos to think, 'Well, he or she doesn't want them out there anymore'?"

"Actually, the reasonable thing - the human thing to do - is to go and take them down."

Also same question about expectation of privacy. If the pictures were taken in a public place would this still be the same story? What about people in the background of me taking pics of my kids at the zoo?

Depends on jurisdiction. Basically all of Europe follows the idea that identifiable published pictures of non-public people are not allowed without explicit permission of the people in the picture. In contrast, in the U.S. pictures taken in public are allowed, with a few caveats - making money using such pictures outside of them being completely incidental or newsworthy is not allowed.

Images of children have always had a different perspective than images of adults, particularly when the parents request that the children's images not be posted.

This was just as true before Facebook or the GDPR as after.

The troll is of course just being a troll, but the position that there's no censorship because the law says there's not is a level of statism that US law would have more difficulty with. And certainly that level of privacy over pictures taken in public could easily be used to shut down news coverage of mass political gatherings.

So, you consider copyright to be censorship?

Ie, grandma can't post her grandkids ebooks on facebook. Remember, copyright now is granted anytime you compose anything. (Penalties require filing government paperwork to register the already granted automatic copyright.)

I was an "author for hire", eg, software engineer, who was updated regularly on Congress expansion of censorship in opposition to anti censors like Richard Stallman.

The difference between the US and Europe is the US is trying to protect corporate monopoly profits/rents on information while Europe tries to protect individual monopoly of personal information.

#3 Maybe we should demand more from our leaders than we have been demanding. Brazil, whose coronavirus epidemics started after America's, has already almost beaten the virus. Rio de Janeiro City, Brazil's New York City, is already planning the full reopening. Flamengo, Brazil's top soccer team has already resumed training. Brazil's Army has already produced millions of chloroquine pills, enough to last the whole crisis.

Just when you thought this blog had achieved complete immunity to Thiago....

Isn't Brazil kind of a mess on COVID? 5-day moving average of new daily deaths stands at about 900 and is increasing sharply. It was at ~400 on May 1. Where do you see "has almost beaten the virus"? Are the data on Worldometer wrong? Or was this all sarcastic?

"Isn't Brazil kind of a mess on COVID?"

If it were, would it be ready for reopening? Would soccer teams be ready to get back to saddle? Evidently, not everything is perfect, but, all things considered, the situation is basically under control.

Do the Straussian reading here....

Yeah, just like Trump has totally defeated "islamic" terrorism....

Just like Trump won the trade wars bringing prosperity to American businesses and workers...

Just like Trump has returned jobs to America making PPEs...

Trump or Bolsonaro saying something doesn't make it true.

Of course, on Brazil, Bolsonaro is saying there is no epidemic in Brazil while Trump is saying Brazil is so disease ridden with epidemic he's imposing a travel ban.

It is not just Captain President Bolsonaro. Rio de Janeiro's Mayor Bishop Crivella is planning a full-scale reopening. Brazil's most popular soccer team, Flamengo, has resumed training. The end of the epidemics is at hand.

So by beaten the virus you mean the daily incidence of cases and deaths continues to rise and the 7-day rolling average of cases per million is the highest in the world (deaths/million comes in a number 2)... sure if that is what you mean by beaten the virus, then yes... Brazil has beaten the virus.

Many commenters on this blog maintain two things -

1. Covid cases are no longer rising anywhere in the world
2. A warm humid country with a city next to the equator is unlikely to be affected by a corona virus.

Brazil is the sort of horrible result that one would expect after seeing what happened to Italy or Spain, but do not expect that reality to be commonly mentioned here.

Isn't it widely expected that the Southern Hemisphere will start to experience increasing impact, about now.

“Many people are saying”. “Many commenters maintain”

Trumpist, thy name is prior

This is a weird one in particular as unlike some other places online, I don't think there are actually *any* strong Covid19 temperature hawks on here.

I've only seen people endorse the proposition that temperature helps a bit at reducing R, which is fairly mainstream and almost settled. (Just as being relatively young probably helps Brazil with fatality rates, etc.)

I get the impression prior has a particular antipathy to this reasonably sensible, mainstream proposition because he thinks it perhaps distracts in some sense -and it's not totally clear how? - from his ongoing and daily bitter crusade in the comments here against the United States in general, the Republican Party more particularly, and the GMU most specifically of all.

There doesn't seem to be any temperature hawks but plenty of temperature chickens—that is, people who keep cherry-picking cases (or trying to do so, cf. New Orleans in winter) that seem to disprove the notion.

The reasonable position to take would be that the Deep State Hail Mary Fever ought to be seasonal in nature, as similar viruses are, but the data is so far inconclusive. Unfortunately that doesn't reinforce the notion of this virus as an unstoppable pox on humanity, so that position goes out the window, ironically by people who otherwise want us to Trust Science™.

I think part of the difficulty is that there are actually 4 linked variables that move together in the ( likely) direction of less virus.
1- temperature
2- humidity increases with temperature , reduces droplet suspension time , increases mucosal secretion thickness the easier to trap the virus
3- UV increases with inclination of the sun across the season , correlated with temperature and generally detrimental to viruses
4- Vitamin D: human production increases with insolation which correlates with temperature.
None of it is conclusively proven but the evidence has been accumulating.

Laser de Requin, if SARS-CoV-2 spreads similar to influenza and cold causing coronaviruses, then it will only be moderately affected by the change in seasons. If a flu or cold virus is sufficiently infectious it will spread regardless of the season. We see this in Australia which can have multiple year round outbreaks of flu thanks to international travel bringing infectious strains from the northern hemisphere in the southern hemisphere. The current Coronavirus has shown itself to be pretty infectious as these things go and so won't stop due to a change of seasons. Better weather can help contain it, but by itself, it's not enough to stop something that is sufficiently infectious.

Suffices to say, Brazil has had a lot fewer COVID deaths than the United States, the richest country in the history.

4. Some interesting and quite strange goings on in the chess business lately. Is there another example where the top player (Magnus) owns the league? This seems quite bizarre to me. It is still not clear how chess should exist as a market sector online, but the game is attracting huge attention among gamers lately due to Hikaru collaborating with a top Twitch streamer.

3. DeSantis is now joined at the hip with Trump. But he has traveled a very different path. A Yale and Harvard Law School graduate, he is an unlikely acolyte to Trump and his know-nothing followers He is only 41, and came along in politics at the same time as Trump, unlike Little Marco who didn't have Trump's Guiding Light. DeSantis is believed to be a future president, and his actions in re-opening Florida now is a bold stroke. I've mentioned in a prior comment that a large part of my family is gathering this weekend in Florida, largely because of DeSantis and their confidence in him; thus, I have a personal stake in DeSantis being right. I will mention that, except for the panhandle, Florida was adhered to the stay at home order. I traveled in Florida in April and there was hardly any traffic, evidence that the success in Florida was due to the order. I worry that success breeds complacency.

If the conditions in Florida don't warrant reopening, I think you should specify what conditions would, in your view.

I'm guessing a Biden victory in November would do it for him.

Trump went to Wharton, thats very comparable to Yale/Harvard - its an adjoining Ivy League.

#3 Imperial released state-by-state analysis of the US - web pages and pdf off By their measure "In
Florida, Rt reduced noticeably before the stay-at-home order, suggesting that behaviour change started before the stayat-
home order. However, increasing in mobility appears to have driven transmission up recently (1.2 [0.8-1.6])."

I've see suggestions that Rt is taken less seriously in the US than in the UK, where it is the centre of every official presentation. Rt > 1 implies exponentially increasing spread.

"2. Thread on collider bias."

+1, I found this informative.

#5: Mississippi's most common "how do you spell..." search is "bidet". I don't think so. I'll buy all the quarantine searches. Is this one of those things that you get when you throw most of the data out because people in most states search for the same thing?

Maybe they ran out of toilet paper and looked for Mrs. Bidet, which is a brand?

Yeah, I don't know whether to blame google or the people who are publicizing (and analyzing?) Google's data, but I've come to believe these maps that are allegedly based on google data are half BS. Maybe it's faulty data processing procedures at JFA surmises, maybe it's sample bias, maybe it's something else. But a lot of these maps are just words or numbers that somebody sprayed on a page.

3 - I'm totally prepared to believe that DeSantis has a good, science-driven strategy. Part of politics and leadership is explaining your strategy to citizens so that they can get onboard, reduce fear, and do their part. Mike DeWine--also a Republican--was getting pretty good coverage at the same moment. The broadly negative media coverage is a sign that someone was doing a lousy job, whether it was DeSantis or his PR people or his health department, or whatever.

Yeah, that's the reason. The negative coverage is due to his PR people.

Your thought being that maybe the press would not have called Florida's reopening "human sacrifice" if the PR outreach had been different?

Sure. PR and press maintenance is a thing. Some people do it better. Other people do it worse. You can't move the narrative beyond a certain point--people on Fox were never going to be super-supportive of the Obama administration--but you can move the needle. There is a big difference between "skeptical" coverage (think responses to Sweden's coronavirus response) and "dismissive" coverage. And that difference is apparently hurting DeSantis; for all that the article celebrates his response, in comparison to Cuomo's, Cuomo has seen a much bigger polling bump.

You can whine about media coverage, and you might even be right, but at the end of the day it is what it is. If DeSantis had a great strategy, he should have been carefully explaining it at press conferences, or getting Fox News segments, or giving lengthy interviews to all the major Florida papers, or publishing an op-ed in the NYT / WaPo, who lap that stuff up for their editorial sidebars. Where was this National Review piece a month ago? I'm sure he was busy. THAT'S WHY YOU HAVE COMMUNICATIONS PEOPLE. It's not rocket science. This media environment didn't suddenly emerge in March.

Politics is a job. The argument of this article seems to be that he did it well but explained it badly. But explaining is PART OF THE JOB. It is literally a job where one of the things you do is get up in front of microphones and say stuff that goes on TV. As Melania would say, "Be best!" He wasn't, or his staff wasn't, and his polling numbers are suffering as a result.

I wonder if any amount of "good communication" would have gotten him good coverage in the establishment media. Perhaps he thought, "Screw it. I'm going to plant my flag, get pummeled by the media for it, and then when it turns out I'm right, I can run on 'I followed the science and went against the tide and didn't care if it made me unpopular with the media. My loyalty is to the people.'"

It can help to look bad for a while. At the beginning of Ronald Reagan's presidency, the economy was bad and his policies were being portrayed as a failure in the media. He kept saying they would work and "stay the course" became a slogan. Things turned around. In the 1984 election, he ran on "it's morning in America" and won a very large victory.

"Morning in America" probably had more to do with Paul Volcker cutting rates from their high of 22% in 1981 to single digits by 1984.

Yes, both the recession and the recovery probably had more to do with Volcker than Reagan (though the decontrol of oil probably helped move things toward an Arnold Kling "pattern of sustainable specialization and trade").

Gov. DeSantis does not have on his PR staff the major news media outlets in Florida spinning for him like Andrew Gillum, whose personal friends are apparently drawn from the most colorful spectrum of the rainbow.

1, Kedrosky: "Epidemiological models continue to acquit themselves poorly, in part because such models are difficult, but also and importantly, to abstract away from this pandemic, because most interesting systems involve humans, and humans adapt and change in ways that work to make models’ predictions fail."

I think the issue here is a misunderstanding. An epidemiological model should not be taken as a 'prediction', it should be taken as a warning. And that is exactly for the reason cited here. The model may well be perfectly 'accurate' when made and based on conditions in place. Now warned, persons change behavior. This is EXACTLY the purpose of those models.

#3: I did some research into mortality rates by age group for COVID19. Turns out COVID19 is really only fatal for age groups above 60. It's near zero for people under 30. The Florida data just confirm that we shut down our entire economy in a big experiment that wasn't even data driven at the national level.

It takes around two weeks to develop antibodies, and adding on the time to conduct and release the tests means the seroprevalance numbers are about four weeks old. There is also an important distinction between how many people have been infected and how many have been exposed, as the T-Cell study shows that many people may be resistant rather than immune -- in the same way that children are resistant. When determining how deadly the disease is, the important factor is how many people are either resistant or immune. This is also important for herd immunity, if, as seems likely, those who are resistant are less likely to transmit the disease.

New interview with the Prof behind the Oxford study of a few weeks ago, Sunetra Gupta, Professor of Theoretical Epidemiology at the University of Oxford. She estimates the IFR at less than one in a thousand.

Well that would be <0.1%, which would be <seasonal flu. That can’t be true. Dr. Fauci told me so.

I think it's looking increasingly likely that this "resistance" thing will turn out to be the factor that was the most confusing all along. It's almost obvious in retrospect: children are almost completely immune, but not because of antibodies. If this is true of some adults too, then the "herd immunity" factor is in fact some combination of antibody and resistance immunity. It also explains why certain kind of events (churches, choirs, conferences) lead to superspreading, but not others (schools, sports): if age is linked to resistance then the combination of age > 50 with indoor speaking event is crucial.

The notion that IFR was 0.1% was refuted the moment covid-19 deaths in New York City surpassed 8,000, which they did a while ago.

#11. Isn't there a simple way of determining whether they are accurate or not? one need only divide confirmed/suspected Covid deaths in each of the regions for which the seroprevalance is done by the number of people (percentage found with antibodies) in the same city/region. If they all come out with the same, or close to the same, number, then don't we have a good idea that the IFR so determined is an accurate IFR for the virus? I'd do it myself, but I don't have the population sizes for each study to hand (actually, it seems like it will be a lot of work to dig these up).

" Similarly, people all over the world are realizing that “preppers” aren’t nuts (at least, in their prepping), that there is merit in thinking in terms of how much inventory of critical things—food, water, and yes, toilet paper—you have to avoid being whipsawed by low-frequency, high-consequence societal risks."

I don't understand people making this point. Surely, if anything, this is a datapoint in the opposite direction: we've had a global crisis that shuttered a lot of productivity worldwide, yet to my knowledge there hasn't been a shortage of food or basic supplies anywhere in the world, with the exception of toilet paper for about a week. Has there been a sustained critical shortage of anything, anywhere? The evidence so far seems to be that supply chains are much more robust than we might have imagined.

People converting to prepping because of this crisis are like people who sold all their stocks for cash in 2008.

>with the exception of toilet paper for about a week.

Well, this is what people believe when their Mommy does all their shopping for them.

Toilet paper shortage was at least 8 weeks, and I still know of one major grocery store that was still out of it last week, right here in the Democratic Peoples Republic of Massachusetts. Same goes for baby wipes, napkins, paper towels and kleenex.

There were severe meat shortages for about 4 weeks, and then intermittent ones for at least another three. You still can't count on getting exactly what you came for to this day (whereas you could in February).

I don't buy bottled water but that aisle appeared ravaged for at least a month.

And this was all for an above-average flu season. If we ever have a real pandemic, and not just a Dem Panic, we are right and truly screwed.

Prep away, children.

Most stores in Portland, OR only had toilet paper available first thing in the morning as of two weeks ago. I'm in Oklahoma City now and haven't noticed as much of a shortage here.

Somewhere in the middle of Stacey Abrams' second term in 2030 there will be another pandemic and then your side can be the ones freaking out

have noticed that zizek mostly talks in circles and touches his face a lot with his hands. the social sciences are teaching him
are we missing anything?

#11: The convergence on 0.7% is there, though it may be somewhat high once more factors are accounted for among the deceased (but likely not vastly high, not an order of magnitude).

Before we get locked into the world of "But the smart people *always* knew it was that...", let's go back and look at the "hammerbros" favourite by Pueyo -

>10 million dead (12.5 million, roughly), under a do-nothing assumption, projected on a case fatality rate of 4.0%, by day 160 of pandemic.

(And at the time as I recall some of them even claimed Pueyos 4.0% was a conservative death rate and Italy 9% was more like the real thing!)
In reality, on a conservative 0.7% IFR, and a conservative 63% assumption of herd immunity, you'd get to 1.3 million.

The >10 million would be off by about as much as the "flubros" IFR of 0.1% (which would predict a max of 185,000 deaths). Hammerbros as bad at prediction and as wrong as "flubros"? It seems so right now.

And before anyone objects "Oh, but that was in a do-nothing scenario" look at the absolute state of the hospitalization metric there.

Ludicrous assumptions about infection:hospitalization and efficacy of hospitalization on fatality...

Imperial College warned Sweden in early April that if they carried on with their craziness that 100,000 Swedes would be dead by early June. The good news, however, that if they had lockdowns across Sweden soon, then only 30,000 Swedish deaths by June 1.

There have been 3,800 deaths with a week still to go, though, and increasing at 1% a day.

It’s probably too late now, but it would be instructive to catalog all the expert predictions and compare them to actual results.

I've never heard of Tomas Pueyo but a quick search shows that he is not a doctor or infectious disease specialist.

One problem I see with the skeptical, right-leaning side of the argument is that, on the one hand, there is lots of anti-elitist denunciation of experts and technocrats but, on the other, few people on this side seem to be aware of what actual experts are or were saying and instead pick on random pundits or bloggers.

On the topic of IFR, any single number for IFR is inherently misleading since everyone now agrees that age is a major factor in how dangerous the virus is. South Korea is reporting a CFR of 0.2% for 40-49 year-olds but a CFR of 26.27% for 80+ year olds. It makes no sense under these conditions to debate whether IFR is 0.7% or 1% or 1.2%. It's none of the above, because almost everything depends on the demographic characteristics of your infected population, which will vary widely depending on the country and the policies followed by that country. A range would make much more sense.

Maybe there is, but in my case I'm denouncing mainly left-leaning (but not exclusively) random pundits and bloggers and newspapers who were skeptical of experts ("explain the science!" as if they were peers), thought they understood it better than the experts, waved around misleading terms like "herd immunity strategy" and pressured governments to change tack to bad strategies, so...

Well, and Neil Ferguson.

(This whole episode shows, if there was any doubt, that the left is not particularly willing to be led by scientific experts, only to use claims of scientific management to accelerate policies they've developed preference for on mainly emotive reasons).

Neil Ferguson estimated the IFR from China to be 0.66%, lower than the estimate you provided above.

For the record (March 30, Lancet):

It's the initial models, in early March, found to be unreliable and buggy, that predicted enormous spikes absent lockdown which we did not see, that were then the problem with Ferguson's work.

Not his later endorsement at the end of March of that IFR, which probably maps over demographically across the West to estimates around 1% which are probably a little hit, but more realistic than the ones commonly endorsed among the hammer bloggers / armchair pundits (naively from looking at CFRs) and far removed from WHO 3.5%.

To be more specific, where Ferguson's Imperial model seems to have really gone wrong is on the hospital admission rate and critical care requirement.

We can test this now; based on current London data, where the infection has effectively ceased in terms of new hospitalizations, seralogical prevalence of 17% implies infection of 1.52 million, and a crude measure of admissions (taking London's share of hospitalizations in England at any time and multiplying by English new hospital admissions related to Covid19), yields about 22,000 likely hospitalizations.

That gives a true hospitalization rate of around 1.5%, about 1/3 of their 4.5%. This excessive hospitalization rate they estimate was then part what gives rise to their predictions of "overwhelmed" healthcare systems.

They then have gone further and badly modeled reproductive rates and impacts of interventions, leading to this recommendations that "lockdown" is the only way to sufficiently lower the R number, based on this flawed model of Ferguson's shoddy code.

These propositions combine in their dire predictions of healthcare resource exhaustion, without lockdown.

Again, while their model was wrong, I don't blame them too much for it. A lot of the further problem is these generally left leaning "hammer" proponent bloggers and pundits who then dragged out to even more questionable assumptions to hospitalization rates of 10-20% and other such nonsense that is well 10x reality. (Think on here of prior_approval repeating ad nauseum his claims that "At not 1-2% of Lombardy's population infected, they have overwhelmed healthcare systems! They are dragging out the dead!" when it is actually probably over 15x that.) Left leaners particularly as it allowed them to bang on with this nonsense about how "austerity" and lack of single-payer healthcare or whatever had left healthcare resources dangerously limited, and launch into a sales pitch for massively expanded and centralized healthcare.

This is what is immediately is then responsible for politicians obsessing over protecting "limited" healthcare resources for the use by young people who would lose many years of life and adopting these policies that sacrificed the old that we now know in hindsight are foolish, like Cuomo's care home policy (clearly meant to preserve hospital critical care resources for what he perceived to be young people's need).

9. Good news for individuals. Bad news for Facebook.

So long, Hertz.

7. Mainstreaming Karl Marx.
Then we discovered the partial defaults so we wouldn't have to go all central planning.

Fresno city auditorium, full of 400 unused beds. Oddly that is what I recommended, the numbers seemed proportional. We are a city of a half million, much less than 1% infected, so far. I would still keep the 400 unused beds until some clearer resolution comes, herd immunity is very painful if you talk to survivors. This is the placed to model Susceptible, Infected, and Third party response. We can color this beach ball.

#3 When we have herd immunity and we look at total deaths and economic costs and allow for voluntary and local action. TC must know of the argument that "policy" has made little difference to behavior.

Some of these article links are about wealthy individuals stepping in to fund potential medical solutions to the COVID crisis. Interestingly, most of these funds are going to University research teams. Universities are huge organizations with billion dollar budgets. Can't they find a couple of hundred thousand dollars to fund promising efforts themselves? Are they prevented from doing so by some kind of law? Is it just not part of the culture? Is it bureaucratic inertia?

Where can you buy such PR for a tax deductible donation?

The cynicism is priceless.

#3 yes protect the nursing homes, but there are TONS of people 70+ outside of nursing homes, and the one relative of mine who died from Covid-19 fell into that category.

Comments for this post are closed