Only those with symptoms are being tested, and way less than all of those. Furthermore the significant differences in defintions of a Wuhan Virus fatality between nationalities mean we can only trust American statistics for virus fatalities, because every American state uses the same defintion.
At best we have American-only figures for an unknown proportion of infected with symptoms to known dead who had the virus. This makes even American-only estimates of the Wuhan Virus' lethality suspect.
I'd be thrilled if we could put some firm numbers on lethality rate. But we still don't know.
“We don’t have firm figures on the proportion of asymptomatic infected to symptomatic infected.”
Disagree. First, the DP provided an adequate “poll” of infected, margin of error +/- a small percent, and second (per the doctor I know involved in the containment/mitigation effort) these numbers seem to be reliable and consistent with full chain contact tracing.
We also have the widespread testing and mitigation in South Korea producing numbers consistent with the DP. Nearly 10,000 cases with very close to the same percentage of dead.
What we don’t know is where they are cutting off the testing in places like NYC and NO and LA, as in what degree of mild is too mild for a test. Sadly, the standards differ from state to state and perhaps even more granularly. So while we can reasonably assume the proportions of asymptomatic, etc., we can’t know how many of the confirmed are in which category (with large numbers of unconfirmed cases apparently left wandering around in the wilderness). We can hope that the US efforts will be as successful as South Korea and the DP, but sadly, I doubt that certain areas will do as well even if the majority of the nation does.
All in all I think we have other things to worry about than the precision of the numbers beyond perhaps 2 significant figures. When somebody goes back and writes an authoritative paper with well researched numbers people are going to trash it anyway given the chaos in the counting currently taking place. It will never be accurate, so why bother with unnecessary precision at this point?
Tack on everybody’s pet conspiracy theories about “already had it 7 times since last summer” and “antibodies that disappear two weeks after you recover” and “the Chinese have a secret vaccine they gave to their Deep State minions” and so on, and it becomes ridiculous to even bother past what we know right now about the distribution of case severities.
I’ll stick with my estimate that the total cases number no more than 5 times the confirmed cases plus 2 or 3 cohorts of incubating cases of the same size as the high end figure for infected in the US. As I type, that’s about 3.2 million cases from critical to incubating. Approximately 1% of the population.
Tomorrow it will be more than 1% by that calculation.
However, we know it is not just the 20% of serious and critical cases in the confirmed stat. Many mild and even asymptomatic are included as a result of contact tracing. I just use that as an upper limit. I’ll be comfortable saying 1% for a few more days. I have no intention of getting dragged off into some swamp of quibbling over exactly how sick is sick and who counts as infected.