As long as users are better than 50% accurate, it shouldn't matter if they're experts or not. That being said, it's difficult to measure user accuracy in this c

As long as users are better than 50% accurate, it shouldn't matter if they're experts or not. That being said, it's difficult to measure user accuracy in this case without running into circular reasoning.