@apricotnuts
Basically it isn't actually possible to calculate the CFR until after an epidemic has finished and all cases are resolved one way or the other (recovered or died).
So there is no 100% accurate answer and there won't be until this is all done.
Some people quote the current CFR at 3.4%
As this is no. deaths/total cases x 100
However there is an obvious flaw with that as of the total cases (83k) there are a lot (43k) that have neither died nor recovered yet. 3.4% basically assumes that none of these 43k will die which is obviously not the case.
Some people quote the current CFR as 7%
This is no. deaths/cases that have reached a conclusion x 100
However this number tends to decrease over time as the beginning of a pandemic always has a higher death rate as we don't know what the new virus is and there's usually a bit of a clusterfuck situation. This 7% is alarming but has been steadily reducing.
Experts at modelling diseases estimate it will end up being around 1-3%, most say 1-2.5%. I tend to use this as they know better than me