there are two ways to extrapolate the data relative to the models.
1) maybe social distancing worked. this suggests that they modeled it correctly, and our good behaviour changed the outcome.
2) or, maybe they overestimated the severity of the disease, including both the need to seek medical help for mild cases and ultimately the mortality rate. this would suggest that the modeling was exaggerated because it was based on exaggerated data, and the reality isn't getting close to it because it was exaggerated in the first place.
i would consider social distancing to be untested science, and am deeply skeptical of the clams being made. yet, my projection was actually that there would be many fewer deaths, because it was based on a mortality rate of 0.1%. if 70% of the population gets infected, a 0.1% mortality rate would suggest it should burn out around 5-6,000 deaths in the city, which is looking like it will be much closer to reality than the projections.
they should be pushing a 40% infection rate in the city, right now.