The paper shows that, in some important respects, the differences between the Nash equilibrium and competing concepts such as the quantalresponse equilibrium are smaller than they appear. I start from the observation that, in many experiments, parameter shifts that leave the Nash equilibrium unchanged affect behavior. I explain the direction of change with a heuristic structural approach, relying on properties such as strategic complementarities and increasing differences. I justify the approach using existing comparative statics results for the Nash equilibrium and new comparative statics results for the quantal response equilibrium. Further, I show that the experimental observations can also be rationalized by a model of adjustment to change that does not rely on any equilibrium concept. Finally, I relate the structural approach to equilibrium selection concepts.