Ummm... Can someone explain to me how the following is logical?: "Your model predicts too many deaths, because it is based on bad or incomplete data. My model is correct--although it also relies on bad or incomplete data, because there is no other kind of data available right now." ALL of the models pile assumption on assumption--and they are the best we can do.
Many of the models we are seeing now are going to be wrong. And all the models predict future events that are highly dependent on actions by third parties the model makers can't control (the timing of lockdown orders in 50 states plus DC and PR, for instance). I think one major problem is that we really can never know for certain if Fauci, et al.'s recommendations were too conservative; but we will get very stark evidence if we remove lockdown orders too soon.
There is no getting around that we are making very important decisions under conditions of radical uncertainty. You can pick your favorite predictions and I can pick mine--but I do hope both will come from folks who actually have relevant training and qualifications. (Yeah, the experts may all be wrong. But when you break your leg, you don't call a plumber--not even a really good, really smart one.)