Political science faculty have access to a wealth of innovative pedagogies thanks to a comprehensive literature on teaching and learning in the discipline and related fields. Yet, from among the hundreds of documented possibilities, how does one go about deciding which to incorporate into a given course? Few articles have much to say in response to this basic question, as most begin with a particular method, assignment, or assessment in mind and proceed from there. Drawing on the work of Fink (2003) and Wiggins and McTighe (2005), among others, I argue that, instead of taking the choice of activities as the starting point in course design, teachers--and, by extension, students--would be better served by first contemplating the desired results of the course and only thereafter devoting consideration to content and instructional methods. In making this claim, I reflect critically on my experience teaching a writing-intensive first-year seminar on comparative authoritarianism using "learning-centered" and "backward" course design. My experience speaks to both the potential pitfalls associated with the learning-centered model and the enormous promise that it represents.