The aim of this article is to present and critically discuss a game-theory-based argument in favour of the view that sports organizations ought to ban the use of performance-enhancing drugs in sport. After presenting the argument in detail, I try to show that the argument is not convincing. First, the argument cannot be used to argue in favour of WADA’s (World Anti-Doping Agency) current ban on doping, at least if it rests on the assumption, that doping use is always harmful. (...) However, that in itself may not be a problem for adherents of the argument, and they can and should modify the harm assumption to cover only harmful use of doping. Second, even with this modification, it is argued that the harm assumption is flawed, for example, because it is not obvious why we should accept certain harms in sport but not harm to athletes caused by doping. Third, the argument is also flawed because it entails the non-competitive assumption: if all athletes dope, then no competitive advantages are gained by any athletes assumptions. The non-competitive assumption is challenged in view of the observations that doping can have some non-competitive advantages and is, so to speak, not only a positional good and because doping, due to unequal responsiveness, can give some highly responsive athletes a competitive advantage over less responsive athletes. (shrink)