- take a baseline run - break two separate things in the test - new run shows both changes - fix one of them - new run still shows both changes instead of just the remaining one
