The idea behind the reputational measure for assessing power of political actors is that actors involved in a decision-making process have the best view of their fellows' power. There has been, however, no systematic examination of why actors consider other actors as powerful. Consequently, it is unclear whether reputational power measures what it ought to. The paper analyzes the determinants of power attribution and distinguishes intended from unintended determinants in a data-set of power assessment covering 10 political decision-making processes in Switzerland. Results are overall reassuring, but nevertheless point toward self-promotion or misperception biases, as informants systematically attribute more power to actors with whom they collaborate.