In this paper I question whether the randomized controlled trial (RCT) and the evidence-based medicine model are well suited for psychotherapy research. First I discuss general characteristics of the randomized controlled trial and argue that, given a number of incompatibilities with the discipline of psychotherapy, these trials tend to provide us with data that have limited validity. The methodological objections I raise are important as the treatment protocols produced within the context of randomized controlled trials tend to be forced upon clinical practice in uncritical ways. On the other hand, I argue that systematic research is relevant for psychotherapy. If research is designed to address concerns arising from clinical practice - in particular, the far-reaching complexity of the therapeutic encounter - it can be a powerful tool for further developing our discipline.