https://www.arcgis.com/apps/opsdashboard/index.html#/8c4dcccd9e3845eb89f6401f919007f2
US has 941,000 licensed beds, of which 64,400 are ICU beds. Current 49% average hospital bed utilization rate, gives us an idea for what would be needed to all non-COVID19 related disease. Lets say 10 million people are infected at any one time, 5% of which require some form of hospitalization, with 10% of those hospitalized getting intensive care (so only 1 in 200 infected gets an ICU bed). That means 500,000 hospital beds, 50,000 ICU beds, and the increased personnel to manage them. If we could turn those cases around in a week (every week), and keep at 10 million new infections a week, we could get everybody through this in 33 weeks, and probably manage things from an available bed standpoint if infections are spread fairly evenly across the country.
Let's say we do a good job at that....mortality rate might be a little elevated if we can only provide intensive care to 10% of those hospitalized (one publication stated 26% of patients needed intensive care).....but lets say we can keep it at 1.5%. If that's the case, sometime in December we could be past this thing with 4.95 million deaths in the US attributed to COVID19.
I'm not saying you are right or wrong to believe that is the best course moving forward, I'm just saying that by choosing that path you do so knowing that millions will die of respiratory failure.
Also, we are fortunate that WW2 had an established end date so everyone knew ahead of time how long they would have to ration. Really made it easier for them to buy-in for the cause....
1 in 200 needing ICU seems high considering 80% of people exhibit few symptoms.
Or are you figuring 10 million of confirmed cases and assume the 80% asymptomic never get tested?