Poaching of rhino and elephant has increased significantly over the last decade. We develop a model that views poaching and protection of a migratory species as a repeated, spatial game between strategic opponents. A poacher and an anti-poaching patrol make locational decisions each season on where to poach and where to patrol. We simulate different locational strategies to determine their affect on species abundance. Random locational strategies that account for the spatial distribution of the species are superior to non-random strategies for both the poacher and the antipoaching patrol. This mixed-locational strategy is shown to be a unique Nash equilibrium. If elephants avoid kill sites from the previous year, they may contribute to their own protection by reducing the number sites where they would be found.