Coronavirus_Models

I find it interesting that when I go to the IHME web site and look at the graphs, the heading states: "COVID-19 projections assuming full social distancing through May 2020"

http://covid19.healthdata.org/united-states-of-america

"Full social distancing" is not what is happening in Chicago.

https://chicago.cbslocal.com/2020/04/08/chicago-mayor-lightfoot-curfew-on-liquor-sales-starts-thursday/

Another article on modeling:
https://www.theatlantic.com/technology/archive/2020/04/coronavirus-models-arent-supposed-be-right/609271/

All these models have a range of outcomes based on what people do and react (for example, response to the guideline of social distancing) that's both hard to predict and may be very different between what going on in the inner city of Chicago versus, lets say, Naperville.

How "local" does a model have to be - depending on population behavior - to be useful?