The reason it would be lower than that is demographics. Once the virus tears through the elderly, killing 5-10% of everyone in that population segment but an aggregate CFR of 1.5-2%, things would slow down. That's where the math gets murky, and this is what
@Dr_Zaiuswas talking about above (He'll never make a monkey out of me).
We can argue about true numbers of infected people across the entire population vs reported positive tests. But it's widely assumed the COVID death toll is underrepresented as well. There are errors on both sides of the equation.
My general premise is that to get to herd immunity without vaccine, a hell of a lot more people would have to get sick and a hell of a lot more people would die. Remember the initial projections of 2.5M? That's a far cry from the 4M that I was talking about, but 2.5M dead Americans is still a hell of a lot. We can use that number - call it 4x instead of 7x if that makes
@supermike80happy.
Further, as has been pointed out above, these numbers do not account for what happens when there is a run on PPE, oxygen, intensive care beds, etc. If we had stopped mitigation efforts in favor of just letting everyone get sick, the death toll would compound and be much, much higher - 10x? 20x?
If we just kept doing what we did, it might take 5 years to get to herd immunity. That's what a flattened curve would look like. 5 more years of wearing masks, restaurants being closed, no movies, empty stadiums, etc.