Quickie
Colonel
Of course in the real world there won't be a 'jump'. The model was made on simplifying assumptions. In the real world there is not even a sharp distinction between mild and severe cases. Everything will be on a continuum. But the phenomenon the model illustrates is real. On average it takes a shorter time for a mild case patient to recover than for a severe case patient to recover. On average mild case patients being discharged today would have been infected later than severe case patients being discharged today. If we're on the right hand half of the bell curve (which we are), this will cause a distortion, because the number of new infections keeps dropping over time.
I don't know why you think the decreasing daily crude fatality rate on the last four days would disprove the distortion I mentioned. It's clear that the number of daily resolved cases are by now too small to deduce a trend from just four data points. Also, in case you didn't notice, the five days average is higher for the second five days than for the first five days.
The proportion of mild cases patient will keep decreasing. This trend will continue and given the speed in which mild case patients are being discharged, a couple weeks from now almost all mild case patients will have recovered, and there will still be many patients in serious conditions. Take a look then at fatality rate calculated from your method and tell me it's not overestimating. Why don't we wait before revisiting this discussion?
The proportion of mild cases patient will keep decreasing. This trend will continue and given the speed in which mild case patients are being discharged, a couple weeks from now almost all mild case patients will have recovered, and there will still be many patients in serious conditions. Take a look then at fatality rate calculated from your method and tell me it's not overestimating. Why don't we wait before revisiting this discussion?
I've already explained why it won't happen that way.
Even if it does let's say because a big bunch of patients got infected on the same day, it won't affect the final mortality rate because the patients that are expected to recover will have already been counted for the calculation albeit at an earlier time. This is the same with the number of patients that are expected to die, albeit at a later time. So, at the end stage of this hypothetical epidemic with no new infections coming in, you may see more patients dying than recovering, but the mortality rate will still remain the same when the number of death is finally included in the calculation together with the number of the recovered patients that have already been counted earlier.
And what makes you think the progress of the infection follows a bell curve? It will arrive at a steady state when the conditions are stabilized and unchanging - conditions such as the availability of medicines, equipment, doctors, medics, etc. Just look at the mortality curve on this web site. The curve goes up and down at the earlier stage when things were in a mess before coming to the steady-state or near steady-state condition at the later stage when things finally settle down a bit.

Last edited: