According to actualism, an agent ought to φ just in case what would happen if she were to φ is better than what would happen if she were to ~φ. We argue that actualism makes a morally irrelevant distinction between certain counterfactuals, given that an agent sometimes has the same kind of control over their truth-value. We then offer a substantive revision to actualism that avoids this morally irrelevant distinction by focusing on a certain kind of control that is available (...) to an agent. Finally, we show how this revised view has two additional advantages over actualism. (shrink)