Not really the death rate will depend on how the spread is occurring. You are making an assumption of equal spread over the whole population but if the older and less healthy are doing more social distancing then this may not be the case.
Yes, tue that is a possibility; but in any case the exact death rate does not change the result of the calculation by a great deal. Changing the death rate from 1% to 0.5% for example only doubles the calculated figure for the number infected.
For a rough calculation like this, I think you can only hope to get order of magnitude accuracy, no more than that.
Equally death rate is a lagging indicator (I would have thought) so it doesn't take into account recent mitigations that have come into place.
Yes, that's right, if new mitigations are applied, which changes the spread dynamic, you would have to wait a week or so for the effects of the mitigations to manifest before you obtain the new exponential spread parameters (ie, the doubling time for the number infected).
But I don't think that affects the calculation I did today, because in the UK major mitigations were only put in place over the last few days, so over the previous 17 days (which is the time period the calculation runs) it would not affect things much.
But for future use of this formula, we would have to wait a week or two for the effect of the mitigations to manifest, and then get the updated exponential parameters.
Even a small change in the parameter of 2 to 2.05 changes the estimate by 1million. So if you are really going to believe in this formula you really need very highly accurate figures for the 'death number doubling time in days'. Rounding to an int will produce huge changes.
True, but again I think we can only hope for order of magnitude accuracy in this calculation.
And note that the sensitivity to small changes is greater when the doubling time is very small, like 2 days in this case. It's not so sensitive when the doubling time is say 6 days.
In fact I can't really believe that the number of infected people is doubling every 2 days, as the 2-day doubling of UK death figures suggest.
I don't really understand why these deaths are doubling so fast; maybe the doubling of deaths is happening faster than the doubling of the number infected, for some reason that I can't put my finger on at the moment.
I certainly hope that the calculated figure of 8.4 million is larger than actuality, because if it is really true that 10% to 20% of those infected require hospitalization, then as those 8.4 million start to become sick, it's going really overload the hospitals and the poor doctors and nurses. Not the mention the patients who will die for want of a ventilator.