I think some of our disagreements come from us not all being on the same sheet of understanding the definitions of data/number presented, JMHO, no offense to anyone.

Some thoughts:
Example is "John" who has been tested 4 times, every 3 months for the passed year(JAN, APR, JUL AND OCT. Lets say John was negative in Jan and July, but postive in March, got well, then tested positive again in October.

I'd say the metrics are:
Total tested: 4
Total postive tests:2 (so yea, it was one person, but his 2 illnesses are 2 seperated CASES)
Total negative tests :2

So lets say it qwas only those 4 total tests in our data pool.
-Postivity rate would stand at 50%. But since John has 10 meighbors not tested, this indicates we ought be testing more to really know how spread the virius is because if those tested 50% of the time are sick, then there probably more untested unidentified sick people out there, and we need to know how many are sick so we can figure out how much medicine, medical people, and hospital beds to plan for. (CDC says: Positive test rate is the number of positive PCR tests over the last 7 days, divided by the total number of PCR tests over the last 7 days.)

And the CDC discloses clearly: Some states track positive tests while others track people who test positive. So someone who takes three tests and eventually tests positive may be counted in noticeably different ways depending on whether the state tracks tests, people, or encounters.

So it appears to this "sober for the minute" dumbarse old Army Tanker, that in some cases while we have different opinions of what the numbers say and how they are collected, and while we beat each other to death arguing, IN FACT WE ARE BOTH RIGHT BUT FOR DIFFERENT REASONS.