There was a substack where someone took the deaths during Pfizer's trial (the same one where 95% effective came from) and compared it to the CDC's estimate of n

There was a substack where someone took the deaths during Pfizer's trial (the same one where 95% effective came from) and compared it to the CDC's estimate of number of injections to prevent a death, they came up with something insane like 5 vaccine-caused deaths to prevent 1 virus-caused death. Major caveats were included because all the numbers were small (IIRC it was 16 deaths in the vaccine group and 14 deaths in the control group, for example), but it otherwise checked out. That said, also keep in mind the much-touted 95% effective came from 8 and 170 cases out of around 40000 participants.