Everything you have stated makes sense. That said, what I was alluding to is that any test scenario in WOFF cannot provide true consistency in that the scenario when flown, will have randomness within it such as number of generated enemy aircraft and "World" activities running which can have a varying degree of impact on performance. The only case I can think of that would provide true consistency is when you fly a QC mission where you can control how many aircraft will be active, cloud density and cover, etc, etc...
Does that clarify things for you?
Imagine that...It happens that the scenario I am referring to is actually a QC flight.
Again, if we fly the same flight many (many) times and the behavior is always
the same (meaning, "with the issue under test"), then it could be reasonably asserted that the only way it will change is if something has had the desired effect on said behavior. And, logically, just as with mathematics, the proof is the opposite operation: Un-do the changes, the issue should return - if we are actually affecting the issue.
I do understand that WOFF itself makes this kind of troubleshooting difficult or confusing for many. As I've explained, this is why it's essential to ensure you have a valid test case - that is, one that (without external influence) will always produce a consistent result. It's fairly easy to set up a QC that will consistently show the issue at hand.
Once again, I am not claiming to have "total control" - but, as long as I can reliably reproduce said issue (and, more importantly, if it can be shown to subside/recur by applying/removing a repeatable "fix") I'm pretty sure total control isn't necessary.