I think the problem with the statistics and models right now is that there is a HUGE margin of error in them - caused by incomplete/errorneous data going in. The problem is this margin of error is not communicated when the models and stats are presented to the public. Comparing stats from around the world is very inaccurate too, as many coutries are understating totals, while others are as accurate as they can be. Many people wind up not believing the models because, not surpringly, they can be inaccurate. Governments are forced to make life changes decisions against the competing medical and economic factors - not an easy decision, made even worse (in the US at least) by everything being highly politicized.
My "crystal ball"...
As more randon testing occurs we find that infection rate is much more widespread, so mortality rates are much lower (more in-line with flu).
But, unlike flu, the diseases spreads much more easily so many more people are impacted.
Wide disparity in mortality between younger/healthier people compared to older/co-morbidity people.
As more time passes with many people facing severe financial hardship, I think its inevitable that governments will be forced to open their economies as much as then can while trying to isolate and protect those with higher risks as much as possible. Once we hit this phase is where we will be forced to see if the hospitals can keep up with the volume.