The problem with the figures given for deaths due to Covid19 [C19] in both the USA and other countries is that numbers are subject to errors. It's a well known fact that deaths are routinely being attributed to C19 when this is not the case. It's one thing dying *with* C19 and quite another dying *of* C19. However, the two scenarios are – as far as I can tell – being lumped in together. This means that the numbers of people being reported as dying *of* C19 are vastly overestimated.
I annoys me no end when I see graphs in the media of estimates of any particular thing that don't show the error bars. What are 'error bars'? They're the uncertainty of the data.
Given the uncertainty of C19 deaths, the error bars could be absolutely enormous. They could be 50%, 75% or as little as 1%. But the fact is, we don't know because they never tell us the level of uncertainty. Currently, the 'official' number of deaths in the UK of C19 is 41,971. But this number could be much smaller.
The lack of an admission of error bars bedevils science. Uncertainty is passed off as truth. Doubt is waved away as an irrelevance. A graph can be a window into truth or a highly effective method of propaganda. Knowing the uncertainly – error bars – gives the data legitimacy. When the error bars are omitted, it's only reasonable to conclude that you're being hoodwinked.