|Data from HERE A modest amount of smoothing* was used to make the curves a bit less noisy.|
Each line represents a different age group. For instance, the blue line on top represents patients who are pre-school children.
The terra-cotta line (red if you are an eight crayon guy) represents school age children.
As you can see, in normal times those two groups have much higher rates of C-19 symptoms even before C-19 showed up. Closing schools crushed the rates for school aged kids in just a few days.
Surprisingly, the stay-home order eventually brought down the pre-school rates as well. Maybe mom and dad decided to economize and keep the kids home from daycare.
The remaining age groups show very similar characteristics. If you average those curves you get a chart that looks like this:
Last data from April 17.
* Smoothing algorithm was ((t-1) +2*t +(t+1))/4 where (t-1) was the observation from the previous day, t was the observation of the day plotted and t+1 was the observation on the day after the day plotted.
Splitting it out by regions, Region 3 was better than other regions. Data for regions 1, 5, 6, 7 and 8 looked very similar and were averaged (gray line). Region 2 was in the dog house and Region 2 South is plotted in dark blue and Region 2 North is plotted in brown.
Based on this data, Region 3 is at pre-Covid levels of symptoms and has been for a week. There is no reason to keep the stay-at-home order in-place in Region 3.
Regions 1, 5-8 are close to pre-Covid levels of symptoms and need a week to demonstrate it is not a statistical fluke.
Bear in mind that this analysis would be more meaningful if the "noisy" 0-18 year-old data was purged.
Region 2 South needs ten days.
Region 2 North is still in the ditch.