I have a pet theory, based not on hard data but on insights from postmortem chitchat. My theory says that novices and experts, when facing evenly matched opponents, make roughly the same number of screw-ups in a game. The difference is that the novice’s oversights will be far more significant.