Mistake in performance is when we simply fail to do what we intend to do. According to Anscombe, in cases of error in action, the notion of mistake in performance allows for the possibility of there being nothing wrong with the agent’s judgment(s), the mistake is solely to be found in his performance. One may choose to refuse this thought and argue that mistake in performance presuppose some erroneous judgment, or, on the other hand, one can claim that mistake in performance is a unique phenomenon. I support the latter strategy, which again opens up for two new options: given that there is some such phenomenon, are these cases to be regarded as actions, or not? I think that they are. How then shall cases of mistake in performance be captured as actions? There are two strategies to choose between here, either to argue that they are actions on theory-independent grounds or to provide theory-dependent reasons for this claim. I consider both alternatives. I offer different reasons for why cases of mistake in performance should be regarded as actions independent from any theory of action, and I discuss the latter option through Peabody’s claim that Davidson’s account cannot capture mistakes in performance as actions, while that of Jennifer Hornsby can. I support the former, but not the latter. It is my aim to show that Peabody has not successfully demonstrated that Hornsby can capture cases of mistake in performance as actions. I conclude by arguing that Anscombe’s original account is best suited for that purpose.