The math doesn't work, because you are forgetting those that died already that if they had lived, would have been a coronavirus victim. It's survivorship bias.
For example, let's say the mean age was exactly 75 y.o., and that the coronavirus killed 100% of people 75 y.o. or older and 0% less than 75. By your logic, it would have no effect at all.
Another example, again with a 75 y.o. life expectancy, suppose it killed 100% of people 80 y.o. or over and none under. By your logic, being infected by coronavirus would add 5 years to your life expectancy.
Even in your extreme example all of the deaths would have to be extremely old people to work out like that. People that statistically have very very few years left.
And for your extreme example to be true, it would have to be the case that the disease doesn't discriminate healthy from weak. But still somehow only kills old people. Because if the disease does discriminate healthy from weak, all of the deaths would be those that would die soon anyway.
Then you must divide whatever estimate you get from that by one hundred. Assuming the highest estimates that it kills around 1% of the victims, which I think are bit high.
I don't see any possible way you could run the numbers and get more than 2 weeks lowered life expectancy. There just are so few young deaths, and the total death rate is very low.
> Even in your extreme example all of the deaths would have to be extremely old people to work out like that. People that statistically have very very few years left.
> And for your extreme example to be true, it would have to be the case that the disease doesn't discriminate healthy from weak. But still somehow only kills old people. Because if the disease does discriminate healthy from weak, all of the deaths would be those that would die soon anyway.
You do realize these are contrived examples to show errors in your math? They aren't suppositions about the actual true death rate of the coronavirus.
> Then you must divide whatever estimate you get from that by one hundred. Assuming the highest estimates that it kills around 1% of the victims, which I think are bit high.
> I don't see any possible way you could run the numbers and get more than 2 weeks lowered life expectancy. There just are so few young deaths, and the total death rate is very low.
I don't think you can "run the numbers", because the data you're using isn't enough to come to any conclusion. Unless you have data that breaks down deaths by age group, and data that breaks down deaths by age normally, you can't answer this question, as far as I know. If you think there is only 2 weeks lowered life expectancy or less, you haven't shown that.
I fully admit that it is a back of a napkin estimate. I've yet to see a better attempt at an estimation, or any attempt really. If you have a better way I'm happy to hear it. I have a small dataset that breaks down COVID and nonCOVID deaths for the same time period, and gives ages for each death.
For example, let's say the mean age was exactly 75 y.o., and that the coronavirus killed 100% of people 75 y.o. or older and 0% less than 75. By your logic, it would have no effect at all.
Another example, again with a 75 y.o. life expectancy, suppose it killed 100% of people 80 y.o. or over and none under. By your logic, being infected by coronavirus would add 5 years to your life expectancy.