It is very simple actually. You build a model. You make up whatever range of numbers you want about infection rates, rates of hospitalization, and rates of people who will die. You show that the number of people needing to be hospitalized will far outstrip the current bed capacity. You write some formulas to produce curves and charts that grow really fast and produce really big numbers. You run a range of scenarios, but be sure to only publicize the one that has the most deaths. Be sure not to use any actual evidence and facts to support your assumptions. And be sure that no one does a large scale, randomized test of the population to see what actual infection and antibody rates might be. Now if you have built that model right, you should easily show that at least 2 or 3 million Americans are going to die of coronavirus disease.
Now you can use those big numbers to completely shutdown the economy and throw tens of millions of people out of work. And when the real number of deaths turns out to be much lower, and it will, you can claim to be a hero who saved millions of lives. It is a no-lose situation for you. Not so much for the general population.
Anyone doubt that this is what has happened in most of our states? It definitely has in Minnesota.