>When a positive outcome is almost certain, there's less need for fine measurements. Gonna have to strongly disagree, here. Declaring that a "positive" outcome

>When a positive outcome is almost certain, there's less need for fine measurements. Gonna have to strongly disagree, here. Declaring that a "positive" outcome is "almost certain" is an invitation for punting on specifically defining what that "positive" outcome is and then hand-waving any measurement of progress. This is especially dangerous when the "positive" outcome comes at non-trivial expense. Edit: a word