Utilitarians are attracted to the idea that an act is morally right iff it leads to the best outcome. But critics have pointed out that in many cases we cannot determine which of our alternatives in fact would lead to the best outcome. So we can’t use the classic principle to determine what we should do. It’s not “practical”; it’s not “action-guiding”. Some take this to be a serious objection to utilitarianism, since they think a moral theory ought to be practical and action-guiding. In response, some utilitarians propose to modify utilitarianism by replacing talk of actual utility with talk of expected utility. Others propose to leave the original utilitarian principle in place, but to combine it with a decision procedure involving expected utility. What all these philosophers have in common is this: they move toward expected utility in order to defend utilitarianism against the impracticality objection. My aim in this paper is to cast doubt on this way of replying to the objection. My central claim is that if utilitarians are worried about the impracticality objection, they should not turn to expected utility utilitarianism. That theory does not provide the basis for a cogent reply to the objection. Originally presented at the 2004 Bellingham Summer Philosophy Conference, Bellingham Washington, August 2, 2004, with comments by Gustaf Arrhenius and Elizabeth Harman. I am grateful to Arrhenius and Harman for their challenging criticism, as well as to Chris Heathwood, Michael Zimmerman, Owen McLeod, Elinor Mason, Eric Moore and other participants at the Bellingham Conference for comments and suggestions. In October, 2004 I discovered Mark Strasser’s ‘Actual Versus Probable Utilitarianism’. I see that in that paper Strasser anticipates a number of the points I make here.