> Generously assuming people live to 90, saving a 30yo who would have died of covid saves 60 life-years, while saving a 70yo saves 20 life-years, i.e. 3x more.

This makes the fallacious assumption that if life expectance at age X is n years, life expectancy at age X+k is n-k years.

This is very much not true. See, e.g., https://www.ssa.gov/oact/STATS/table4c6.html

Using actual life expectancy by age makes the number of life years saved at age 70 much closer to that at age 30.

Overall good point. Much closer but still significantly different. The life expectancy is 14 years vs 48

> The life expectancy is 14 years vs 48

...which is still a factor of ~3x, which means my napkin math wasn't too shabby, which means this tangent was completely irrelevant, because my main point is unchanged.

But you're giving equal weight to a year lived in the prime of someone's life vs. a year lived in a care home, which might not be a reasonable way to calculate it.

fair enough. "closer" but still significantly different is a support for your position.