This was more what I was getting at. If two students face very different opponents, how do you know if one student is doing better than the other. Maybe the opponent was just harder for someone of that level to read. You might be able to mitigate that with your own read, though that's subjective. It's a common problem in MA testing: how do we remove as much of the variability and subjectivity as possible? Its certainly not unique to the kind of test you're proposing - I just always like to hear how folks consider the problem.If I was going to grade a student it will be on things such as ability to use techniques, read the situation, read the appoint, and exploit the opponent, and defend against the opponent. If I see that the student is constantly getting surprised by the opponent then I wouldn't be happy. This stuff would be unacceptable for any student who I would consider being on an intermediate level.