100% accuracy is neither feasible or required. Only a statistically valid random testing is necessary.
I agree, and realize I should have phrased that better. That said, it’s not yet clear what the optimal requirements for a statistically valid random test should be, given the unknowns of COVID-19, but at this stage any new data improves upon what we’ve got. And what we’ve got can be compared to a mountain island, representing the known/tested infected, with the snowy peak the deceased, and the larger submerged base the unknown infected.
