3 ms·

0.15% turned out to be a lot closer to the real number (something like 0.3% or 0.4%) than the IHME models used to justify lockdowns (which were putting out numb

0.15% turned out to be a lot closer to the real number (something like 0.3% or 0.4%) than the IHME models used to justify lockdowns (which were putting out numbers like 3.4% around the same time), but somehow no one ever brings up their wildly off-base forecasts as a reason to dismiss everything they have to say in perpetuity.

watwut 5y ago
Because that was "worst case scenario" model, in order to anger "worst case scenario estimation" question.
jsnell 5y ago
0.3% clearly can't be the real IFR, given there are places with significantly higher PFRs. E.g. 0.6% of the entire population of Peru died. Please show a source for that claim about a 3.4% IFR in the IHME. As far as I know, nobody serious ever made such a claim. The serious IFR estimates from very early on were roughly 1% plus minus a little (e.g. the Imperial models were assuming a 0.9% IFR), and it looks like those initial estimates were spot on. The exception to that is Ioannidis, who early on was claiming that the IFR would be two orders of magnitude lower than that. He predicted there would be only 10k dead in the US when there were 120 deaths. Three weeks later, there were 20k. He staked his reputation on absurd claims, and has been doubling down on borderling fraudulent research since then to support those initial mistakes.
- jrochkind1 5y ago
  The actually experienced IFR/CFR depends on availability and quality of medical care, right? I don't have a cite, but I feel like I remember people suggesting (whether based on something other than their imagination I don't recall) up to 5% IFR if no medical care was received, perhaps because medical system had collapsed under demand? But in general of course what you say seems right, and I appreciate your well-informed discussion contribution, thanks.

3 ms·

0.15% turned out to be a lot closer to the real number (something like 0.3% or 0.4%) than the IHME models used to justify lockdowns (which were putting out numb

by AndrewBissell 5y ago

watwut 5y ago
Because that was "worst case scenario" model, in order to anger "worst case scenario estimation" question.
jsnell 5y ago
0.3% clearly can't be the real IFR, given there are places with significantly higher PFRs. E.g. 0.6% of the entire population of Peru died. Please show a source for that claim about a 3.4% IFR in the IHME. As far as I know, nobody serious ever made such a claim. The serious IFR estimates from very early on were roughly 1% plus minus a little (e.g. the Imperial models were assuming a 0.9% IFR), and it looks like those initial estimates were spot on. The exception to that is Ioannidis, who early on was claiming that the IFR would be two orders of magnitude lower than that. He predicted there would be only 10k dead in the US when there were 120 deaths. Three weeks later, there were 20k. He staked his reputation on absurd claims, and has been doubling down on borderling fraudulent research since then to support those initial mistakes.
- jrochkind1 5y ago
  The actually experienced IFR/CFR depends on availability and quality of medical care, right? I don't have a cite, but I feel like I remember people suggesting (whether based on something other than their imagination I don't recall) up to 5% IFR if no medical care was received, perhaps because medical system had collapsed under demand? But in general of course what you say seems right, and I appreciate your well-informed discussion contribution, thanks.