The variations in deaths between years is so large (over 30k deaths difference, sometimes up sometimes down) that year on year comparison may not be meaningful.

The variations in deaths between years is so large (over 30k deaths difference, sometimes up sometimes down) that year on year comparison may not be meaningful. You could just twist the data to prove any point, even that covid-19 reduced deaths.