Proportion of tests producing inconsistent results on identical code.
Tests with differing outcomes across reruns of the same commit ÷ total tests × 100
Under 1%