So you imply that you have final results, but again, "what/who is your "codebreaker in the back"? Any machine/computer has fixed limitations on the possiblities it will consider based upon criteria and algorithms that were programmed in. It would not be impossible to devise a code with such obscure formulas and advanced interrelationships between the values that a "codebreaker" couldn't solve it without first being given at least a portion of the algorithm. This would especially be true if the code writer were the developer of the "codebreaker" and knew it's limitations.
The flip side of that is that we've noticed many patterns that seem to develop but either contradict themselves or never fully develop, so is there any possibility that the data we are considering may have gotten corrupted some how?
So I finally looked at the spoiler because I've been beating my head against my desk trying to figure out which label was on which of the final two cups. Then I realized I simply can't read - I guess I missed the "total value" and remaining two labels bit. Arrgh!! Great little puzzle - thanks.
I don't get to crack at it a lot but I'm actually working on C4 - you already pointed out the relationship between B4 and C4 in the 020# series. I'm sure you notice the deviation that occurs at 0208 - since the quotient of (B2+B4),10 is 1 - B4 jumps by 2 instead of 1 like from 0206 to 0207.
I'm working on the actual algorithm for this relationship as well as how B1, B2 & B3 play into it - several theories but none fleshed out well enough to pass on. I got close once but couldn't quite make it work for all cases.