Implicit in your analysis is the quantum of coronavirus cases in the States equaling the number of influenza patients. You're also assuming a fatality rate of 7% for those so diagnosed: as of today, per the Johns Hopkins map, there are 20,704 confirmed cases and 427 so deceased, or a 2.1% fatality rate.
There are a lot of factors that go into contracting influenza vs coronavirus. For example, there are almost NO precautions we take in the US when it comes to influenza aside from extra hand washing during the flu season. That's why, in part, so many people get the flu.
If Americans adopt precautionary measures for coronavirus like not going to work, then the likelihood comes way down.
But let's have some fun with numbers: if we accept the PRC numbers (a BIG assumption) then we have 20k+ confirmed cases out of about 35MM quarantined people in cities, or a 0.06% chance of contracting coronavirus.
If we apply that 6 basis point chance to 21MM seeking treatment, you get 12,600 people contracting coronavirus. If we apply a 2.1% fatality rate, we get 265 dead people in America.
That'd be bad, but it's not 375k.
You’re also assuming a fatality rate of 7% for those so diagnosed: as of today, per the Johns Hopkins map, there are 20,704 confirmed cases and 427 so deceased, or a 2.1% fatality rate.
The 2.1% fatality rate you cite is just plain bad math. The most you can say there is that out of the 20,704 cases, only 427 are deceased *already*. That is not a diseases’ fatality rate.