When it comes to equestrian disciplines, the horse-rider dyad is amongst the most discussed topics. Recently the emergence of equitation science has led to an increased interest in objectively quantifying the interaction between the rider and horse. In this paper a methodology is presented to evaluate how the mental state of police horses interacts with that of their riders in order to assess the performance of police horses. This paper demonstrates how Body Sensor Network technology can be applied for real-time monitoring of the horse-rider dyad. The results of the study demonstrate that the mental state interaction between rider and horse is significantly different between bad police horses and good police horses.