There is a growing body of evidence that a significant portion of the population has some level of T-cell immunity to COVID-19 due to exposure to previous viruses. The result would be that we don't need to get anywhere near the original 60-80% infection rate number that has been thrown around before things would die out.
- This fits with the anecdotal evidence both that only about 15-25% of a population becomes infected before the rate of spread plummets (Diamond Princess, USS Teddy Roosevelt, NYC, Sweden)
- It also fits with how COVID has been especially brutal on the very elderly, because T Cell function declines with age.
We will see.
There have been some occurrences that say "hold on" regarding your first point. I mentioned a couple in the big thread. One was a fishing vessel that went out and almost everyone (like 100 out of 110 or something like that) got infected. The other is prisons where very high percentages get infected. If there was really a "magic herd immunity" that kicked in once about 25% of people got infected, that wouldn't be happening.
More likely, to me, is that hard hit places like NYC have about 25% of the population already infected, removing them from the transmission chain. Then there is some unknown percentage of people who have some natural level of immunity. No one really knows what that is, but lets say 25% (which seems high given the cases I mentioned). That would bring NYC to 50%. Then add in the effect of the social distancing, mask wearing, etc which is hard to quantify but let's say it's equivalent to another 25%. That would total up to 75% which is generally what most estimates say is needed for true herd immunity in a community. This would explain the low rates of occurrence currently in places like NY, NJ.
The big "but" though is if you relax social distancing and mask wearing, suddenly you're at 50% and that could start transmission chains going. Not as effectively as back in Feb/March, but still possibly enough to fill hospitals again.
As I said...we will see!