Ah, finally figured out the problem. The reason why sacreenshot tests keep failing between either header "O" or "R" is because this uses a new timeout based system every 750 ms or so: Two scrolls that happen to start in between might be perceived as a lesser swipe to conserve future scrolls for the velocity. Will be looking to see if I can add a system timer to start actions when the time % 750 == 0 to ensure every screenshot test starts at the same time.