There's one factor you left out: Time.
5800 breakthrough infections so far. How long have the 80 million been "vaccinated"? Give it a year and see how many more breakthroughs there will be, and how many will die--both from Covid, as well as the shot to prevent it.
What evidence do you have currently to suggest that there's even a remote chance that deaths from rare reactions to the vaccine, itself, will ever outpace (or even come close to) the number deaths from the disease itself?
As far as the rate of breakthrough infections, the rate of those will actually decline as more people get vaccinated...which is why establishing a certain measure of herd immunity is important.
From a clinical standpoint, the terms "efficacy" and "effectiveness" have different meanings.
But for the sake of keeping it simple, let's say that all of the calculations have been done to calculate for efficacy and effectiveness, and it's determined that a person, themselves, taking the vaccine will prevent them from catching it (upon being exposed) 9 times out of 10.
So if you went to a party with 11 people people (yourself included) where all 10 other people have it and you have close contact with all 10 of them, it wouldn't be terribly surprising if you still contracted it. However, the more people at at the party who are vaccinated, the less likelihood you have of encountering that "1/10 ten chance" of contracting it.
This clip (from 2017, pre-covid) visually explains herd immunity pretty well, as well as provides some estimates for the levels of vaccination required to establish community protection based on the basic reproduction number.
The reproduction number for Covid is around 3, so to really see the vaccines full potential, we'd need 2/3 of the population fully vaccinated.
So, as I noted before, the fact that it's already showing the kinds of protection numbers that it is, with only 20-30% of the population being fully vaccinated, is quite encouraging.