PLAYPEN: Covid-19 Everything Political

[William Walker]

MEMO:
TO: Lula's Four Horsemen Of The Apocalypse

New talking points please:
Start questioning the accuracy of the USA Covid-19 death toll.
"Obamagate" - a good slogan, now we need to attach it to a good "scandal".

Tasks Completed: "Evolution" - check
                           "Climate Change" - Check
                           "Russiagate" - Check
                           "Ukrainegate" - Check
                           "Science in General"
From: CULT H.Q.

[Slobodan Blagojevic]

... infections will be rising...

Yes, that's the hope.

At least until November.

[Slobodan Blagojevic]

... unsubstantiated conspiracy theories...

Chris, it would be more helpful if you would provide a rebuttal. As I said earlier, I am late to the party, so would like to be informed from a reputable source (you). 

[elliot_n]

Good listen, former NYTs medical reporter on the models and on our current response.  The title suggest he is being punitive and scornful, but he is not.  He is just explaining why the models have been so wrong, and therefore why the policy has been so wrong. 

The Failure of Expert Predictions and Models | The Coronavirus and Public Policy

So I will admit that I have been a little too tough on Ferguson.  Essentially he was guessing in his initial estimate since data from China was not trustworthy.  Shortly after his initial crazy estimates came out, he revised his estimates down drastically to 20K deaths in Britain.  Not to mention he did say most would be the elderly or sick and would have died within a the year anyway. 

Unfortunately the press did not bother to give coverage to this change.

There was no change. This guy is misrepresenting Ferguson.

Ferguson's prediction of 20,000 deaths, if the UK followed a suppression strategy (i.e. a lockdown), is included in the original March 16 report (see the tables on page 13).

If we look at the current UK figures, it is clear that Ferguson massively underestimated the severity of the virus.

The current official UK death toll is 32,000, but most experts believe the real figure is closer to 50,000. Our daily deaths and cases are still high (though thankfully falling), so by the end of the summer we will likely have a death toll of around 100,000 — that's five times Ferguson's original estimate.

[elliot_n]

His critique of the U. of Washington model is more judgmental.  Essentially, U of W was updating its model but not making sure it the model accurately predicted the conditions on the ground of the day for each update. 

He recorded the video two weeks ago and commends the the University of Washington model for making a good prediction for the number of deaths. At the time of the recording, their model was predicting 66,000 deaths and the actual deaths were 60,000. Since then, the modellers have revised the projected death toll up to 147,000 (by Aug 4), with the current death toll at 87,000.

This guy is trying to get famous by arguing that the modellers have overestimated the dangers of the virus, when in fact the opposite is the case.

[Slobodan Blagojevic]

... the modellers have overestimated the dangers of the virus, when in fact the opposite is the case.

The original overestimate was 2.2 million in the US. That scared the sh&t out of everyone. If they used the numbers they are using now, there would have been no lockdowns.

[elliot_n]

The original overestimate was 2.2 million in the US. That scared the sh&t out of everyone. If they used the numbers they are using now, there would have been no lockdowns.

2.2 million is simple maths. It's only wrong if the IFR (Infection Fatality Ratio) is less the 0.9%, and if less than 80% of the population contract the virus.

Hopefully the IFR will turn out to be less than 0.9% — but so far it seems like a reasonable estimate.

If social distancing measures are adhered to, and if people wear masks, then it is likely that a vaccine will be available before 80% contract the virus.

So less than 2.2 million deaths.

[Robert Roaldi]

MEMO:
TO: Lula's Four Horsemen Of The Apocalypse

New talking points please:
Start questioning the accuracy of the USA Covid-19 death toll.
"Obamagate" - a good slogan, now we need to attach it to a good "scandal".

Tasks Completed: "Evolution" - check
                           "Climate Change" - Check
                           "Russiagate" - Check
                           "Ukrainegate" - Check
                           "Science in General"
From: CULT H.Q.

That's very funny. Thanks.

[Jim Pascoe]

The original overestimate was 2.2 million in the US. That scared the sh&t out of everyone. If they used the numbers they are using now, there would have been no lockdowns.

That makes no sense surely?  The numbers are lower BECAUSE of the lockdown are they not?

[Slobodan Blagojevic]

That makes no sense surely?  The numbers are lower BECAUSE of the lockdown are they not?

They are not.

[JoeKitchen]

2.2 million is simple maths. It's only wrong if the IFR (Infection Fatality Ratio) is less the 0.9%, and if less than 80% of the population contract the virus.

Hopefully the IFR will turn out to be less than 0.9% — but so far it seems like a reasonable estimate.

If social distancing measures are adhered to, and if people wear masks, then it is likely that a vaccine will be available before 80% contract the virus.

So less than 2.2 million deaths.

It is statistically vague and misleading to say that the IFR may be 0.9%.  For those under under 50, it is more like 7/10,000.  For those under 30, it is about 1/5,000.  It is the fact that the  IFR is very high for those over 60 that causes the IFR to spike higher for the overall population. 

We need to start to recognize this and shape policy that protects the elderly while allowing the economy to open.  For instance, 70% of all C-19 deaths in PA happened in nursing homes.  There is no reason to keep the state closed if that is the case.

FYI, our health secretary Dr. Rachel Levine also implemented the same oder NY did, requiring all those who tested positive to be admitted back into their nursing homes.  He apparently was so confident in this order that he removed his mother from her nursing home the same week it went into effect.  So keep that in mind when you question why I don't have confidence in my state's leadership. 

[Robert Roaldi]

That makes no sense surely?  The numbers are lower BECAUSE of the lockdown are they not?

That's the paradox of prevention. When prevention works, there will be always be some who insist that there was no problem to begin with.

[JoeKitchen]

Even for those under 65 in good health, the IFR seems to be really low. 

Coronavirus (COVID-19) Mortality Rate

Towards the end of the article they judge that the CFR (crude fatality rate) for those under 65 is 0.09%, or 9/10,000. 

This is just not as bad as people are making it out to be unless you are over 60 with a pre-existing condition. 

[JoeKitchen]

That's the paradox of prevention. When prevention works, there will be always be some who insist that there was no problem to begin with.

No it not.  This is just you using bad logic.  You need more proof then your goal occurring to prove that your cure actually worked.  Otherwise you could always just pick an impossibly high upper bound on the prediction to use to prove your procedure always works. 

For example, if you are testing a new drug and you gave it to someone who then got better, that does not mean the drug actually worked.  You need a control group that was given a placebo to compare to. 

Saying the lock downs worked because we never approached the high ends of the models is pretty much the same thing as saying your drug worked even though you did not have a control group. 

Fortunately Sweden bucked the trend, along with a few USA states, giving us a control group, and in a year we will be able to see if the locks downs actually did anything.  Personally I dont think they did given the group where the most deaths are concentrated. 

[Alan Klein]

My mother-in-law and all the other residents and staff at her assisted living community will be tested shortly, or just have.  It will be interesting to find out what the infection rate is.

[Alan Klein]

Donald Trump threatens to cut ties with China over coronavirus

https://www.youtube.com/watch?v=wRbaMNkBtaM

[elliot_n]

Iin a year we will be able to see if the locks downs actually did anything.  Personally I dont think they did given the group where the most deaths are concentrated. 

Bad logic.

[Alan Klein]

Bad logic.

Why?

[JoeKitchen]

Bad logic.

No it is not, or at least the first sentence is perfect logic.  You can not decide if an action like this worked without a control group, and we currently are not far enough into this giant experiment for the data to have fully come in.  So we have no way of knowing if the lock downs did anything. 

Great example of why you need a control group is with the Polio vaccine when they tested it in the 1960s.  The initial test preformed though out the USA with 2nd graders required school districts to set up a control group to compare the results to.  Most school districts broke up the students whom they received parental permission to perform the test on into two group, one who got the vaccine and one that got a placebo.  In this instance, it showed that the vaccine worked, and worked well. 

However, some school districts decided to give the vaccine to all students they received parental permission from and just decided to compare those results to the students whom they did not get permission to include in the test.  In this instance, it showed the vaccine had no effect what so ever.  Seems strange, and it took researchers 40 years to figure out why this discrepancy existed.  In short, it was due to an improper control group. 

Insofar as my second sentence, that is my personal opinion which I am not really using logic for.  But, if you look at the fact that 60% of fatalities occurred in nursing homes, it is kind of obvious the lock downs were overkill. 

[Alan Klein]

No it is not, or at least the first sentence is perfect logic.  You can not decide if an action like this worked without a control group, and we currently are not far enough into this giant experiment for the data to have fully come in.  So we have no way of knowing if the lock downs did anything. 

Great example of why you need a control group is with the Polio vaccine when they tested it in the 1960s.  The initial test preformed though out the USA with 2nd graders required school districts to set up a control group to compare the results to.  Most school districts broke up the students whom they received parental permission to perform the test on into two group, one who got the vaccine and one that got a placebo.  In this instance, it showed that the vaccine worked, and worked well. 

However, some school districts decided to give the vaccine to all students they received parental permission from and just decided to compare those results to the students whom they did not get permission to include in the test.  In this instance, it showed the vaccine had no effect what so ever.  Seems strange, and it took researchers 40 years to figure out why this discrepancy existed.  In short, it was due to an improper control group. 

Insofar as my second sentence, that is my personal opinion which I am not really using logic for.  But, if you look at the fact that 60% of fatalities occurred in nursing homes, it is kind of obvious the lock downs were overkill. 

Why was the control group incorrect?