Red light cameras are installed in many jurisdictions around the world to improve road safety by discouraging red light running behaviour at signalized intersections. In many jurisdictions, a smaller number of cameras are rotated among a larger number of locations to increase the effectiveness of the red light camera programs. This study outlines some arguments for deciding the best deployment strategy that can be used to rotate these cameras, focusing on the deterrent effects of enforcement and punishment on driver behaviour. A simple on-road experiment is then conducted to determine the effectiveness of three different rotation schemes. Our results suggest that not moving the cameras and keeping enforcement predictable yields a lower average violation rate than moving the cameras around and making enforcement uncertain.