You can't control for mortality here, that doesn't make sense, because mortality is the variable under investigation. He does control for the most obvious conf

You can't control for mortality here, that doesn't make sense, because mortality is the variable under investigation. He does control for the most obvious confounder (age). Some people - like the UK HSA and ONS - have tried to argue there must be other confounders like vaccines being given preferentially to unhealthy people, but they're self-contradicting on this point. I've seen them argue for both a healthy vaccinee effect and also an unhealthy vaccinee effect, both times with no direct evidence. They're shooting in the dark to try and avoid the obvious inference. The bandage analogy is illogical. Injuries cause bandages so of course there's a correlation. Vaccines are not meant to cause death, so there should be no correlation. They're meant to prevent death, so the ratio should run in the opposite direction - the unvaccinated should die more often then the vaccinated.