I was thinking the same thing, but a more conservative estimate using people who have had at least 1 dose of the vaccine or those who have recovered to calculate herd immunity.
It depends on the r value of the virus. A more transmissible virus requires a higher level of population immunity. Most common estimate I've seen for covid is 70-80%
Having no background whatsoever in this area of science, I would imagine that the percentages change over time as some individuals are still going out more or less living their lives and some are holed up extremely risk averse and won't be infected unless their 1 trip to the grocery store per week is compromised.
And it'd be similar for who is readily willing to take the vaccine vs. will refuse to ever do so.
https://github.com/Arkwin/US-COVID-Herd-Immunity-Tracker/
I update the data from the CDC website every day.