I found it difficult to monitor the various data offered for cases. At first there were few tests and lack of reports. As tests became available better data appeared, and many areas had no test kits (as here where I live). Now more places have tests, but reporting is not always current. i.e. here in California cases fall over the weekend and then go back up and exceed the previous weeks numbers. These lags, and now reports of decreased testing beg the question if all cases are being reported. Until a trend develops, consistent reporting and there is a leveling off and a fall of new cases with a corresponding increase of demise (for a short period) will the actual crest have happened WITH sheltering. If people lax off too soon on sheltering (I feel this is why the date in early May (1 or 3) was selected) the trend may run flat, or even increase for awhile. I've only monitored California and I am seeing an increasing trend over the past 4 days. More days will tell for sure. It will be equally interesting to see what happens in those locales with no shelter orders and also if there are Easter gatherings. Also age specific demographics of ill and deaths will be interesting, but doubtful if public would be given access. I'm also curious as to if testing of more people would be helpful in relaxing shelter orders if it shows many people with no or slight (un diagnosed) cv19 exist in the general population.