Global Positioning System (GPS) units have recently become popular for monitoring and assessing the workloads of football players. Currently there is a lack of studies examining the validity and reliability of these systems for that purpose, so the aim of the current study was to determine if 15 Hz units (SPI HPU, GPSports, Canberra, Australia) could accurately be used to describe the physical demands of football. A series of cohort studies were carried out with Canadian university football players (n=28). In order to assess the accuracy of the units’ ability to measure high velocity sprinting, 12 players performed multiple electronically timed 36.6 m sprints while wearing theunits. To assess the inter-unit reliability, five players wore two units each during a training session. An analysis of the units’ validity for measuring collisions was carried out by comparing the correct number of tackles and blocks notated on video by an expert rater in two games with the number of collisions recorded by the units. The units were accurate for measuring high sprinting velocities (CV=0.9%) and had good inter-unit reliability for recording distances at velocities between walking and sprinting CV (1.4-7.8%). The collision algorithm filter the accompanying software uses was found to have its best balance between precision and recall using a cut-off of2.65 g for linemen and 2.9 g for non-linemen. The devices used are effective at providing acceptably valid and reliable information to describe the physical demands of football. Position specific locomotor zones are recommended when using GPS units with football players.