In fairness, the latest revision of the model is doing quite well so far. April 11 total deaths: predicted 20,334 actual 20,577. That’s an error of 243 deaths or a bit over 1%. Further the current revision is so far underestimating the death toll. Obviously that may change, but so far the latest model is doing well.
BTW, this is fairly typical in science. Initial models often are inaccurate. There is typically much initial uncertainty about important factors that go into the models. As more data becomes available it becomes easier to determine the important parameters more accurately, and an improved model is the result. Inaccuracy of a previous model does not imply inaccuracy of the current one.
Well, if a model is off by 2,140,000 deaths in its predictions, then it’s a crappy model. You don’t need an advanced degree to understand this.