Can you cite some references on "there was really going to be a curve"? I haven't seen any models that expect it to die out on its own without infecting a significant proportion of the population first - all of them assume exponential spread, just different rates of it (which is exactly what social distancing lowers).
Your question is part of my complaint. There have to be models that this didn't spread exponentially in the severe cases. Where are they, and what were their odds?
But we haven't observed much, yet. This is akin to the xkcd about how married you will be for days after your wedding. We are way too early to know what the later curve looks like.
