3 ms·

Sure. Nobody knows the real number and efforts to estimate it vary wildly in their outcomes. Please read my comment again, what you quoted and also what I didn

by titzer 6y ago

Sure. Nobody knows the real number and efforts to estimate it vary wildly in their outcomes. Please read my comment again, what you quoted and also what I didn't say. I am not asserting that confirmed == actual. Of course that would be silly to believe that. I am explicitly cautioning others from just fudging out numbers and then reasoning based on speculative numbers. I am not substituting a different number or different reasoning. I only quoted actual facts, and I am explicitly encouraging us all to stick closer to actual data, because we just wander into bad conclusions otherwise.

nradov 6y ago
Looking at confirmed case numbers leads to bad conclusions because those numbers are almost completely worthless. It's common knowledge now that the majority of cases are never counted in official statistics.
- barumi 6y ago
  > Looking at confirmed case numbers leads to bad conclusions because those numbers are almost completely worthless. They really aren't, and trying to portray official numbers as "worthless" is something that flies in the face of reason. In the very least they serve as a verified conservative estimate, or a reliable lower bound. Moreover, it's stupid and disingenuois to claim that "majority of cases are never counted in official statistics" because "official statistics" represent only test results taken at specific moments in time, and never were assumed to represent the total number of cases.
- Smithalicious 6y ago
  So instead look at this number I just pulled out of my fucking ass
ajsnigrutin 6y ago
Weren't there randomized antibody tests in multiple countries, dealing with just that (basically, calculating the approximate number of people who got it, and then comparing to number of people getting tested positive)?
- nradov 6y ago
  Yes but now that we are months into the pandemic those antibody seroprevalence studies also underestimate the true number of infections. Some patients, especially those with minor infections, never produce detectible levels of antibodies or they fade away quickly. In order to get really accurate data we would have to also check memory cells, but that requires specialized and expensive laboratory procedures.

by titzer 6y ago

nradov 6y ago

Looking at confirmed case numbers leads to bad conclusions because those numbers are almost completely worthless. It's common knowledge now that the majority of cases are never counted in official statistics.

barumi 6y ago
> Looking at confirmed case numbers leads to bad conclusions because those numbers are almost completely worthless. They really aren't, and trying to portray official numbers as "worthless" is something that flies in the face of reason. In the very least they serve as a verified conservative estimate, or a reliable lower bound. Moreover, it's stupid and disingenuois to claim that "majority of cases are never counted in official statistics" because "official statistics" represent only test results taken at specific moments in time, and never were assumed to represent the total number of cases.
Smithalicious 6y ago
So instead look at this number I just pulled out of my fucking ass

ajsnigrutin 6y ago

Weren't there randomized antibody tests in multiple countries, dealing with just that (basically, calculating the approximate number of people who got it, and then comparing to number of people getting tested positive)?

nradov 6y ago
Yes but now that we are months into the pandemic those antibody seroprevalence studies also underestimate the true number of infections. Some patients, especially those with minor infections, never produce detectible levels of antibodies or they fade away quickly. In order to get really accurate data we would have to also check memory cells, but that requires specialized and expensive laboratory procedures.