Positivity is used by many as a measure of success or failure wrt to Coronavirus. However, we know that when testing is low and we are testing only people with symptoms positivity is going to be high. We also know if we test enough the positivity rate will nearly match the prevalence of the disease in the population. The question becomes how to quantify this. I used the YYG model to estimate the prevalence in the population against time. The next step was to create a ratio of positivity vs prevalence. That ratio could then be compared the number of tests per day over time. The result was a curve fit using a power formula. It worked out to be tests to the -0.44 power. (I compared tests to (positivity/prevalence)) When I used this formula to create a weighted positive cases curves it was a very close match in shape to hospitalizations. Put more simply when you double the tests at a given prevalence you get 1.36 times the positive results.