US Coronavirus deaths per day...

[Bernardo de La Paz]

5. That is not to deny the OPost figure, is it. There are two death rates.

One is per detected cases. That what is usually meant when death rate is discussed.

The other is your figure which is per actual cases. It is the more desirable figure to know but at this point it is inscrutable. The ratio of true cases to known cases varies from time to time and from place to place. It is only estimable after the fact and even then will have bounds of uncertainty around it.

The problem is that the per-actual adds another layer of uncertainty onto the detected figure (which has uncertainty) which is based on an already clouded official death count. A triple layer or triple cloud of uncertainty makes the final figure effectively meaningless.