Full-scale roller rigs are recognized as useful test stands to investigate wheel-rail contact/damage issues and for developing new solutions to extend the life and improve the behaviour of railway systems. The replacement of the real track by a pair of rollers on the roller rig causes, however, inherent differences between wheel-rail and wheel-roller contact. In order to ensure efficient utilization of the roller rigs and correct interpretation of the test results with respect to the field wheel-rail scenarios, the differences and the corresponding causes must be understood a priori. The aim of this paper is to derive the differences between these two contact cases from a mathematical point of view and to find the influence factors of the differences with the final aim of better translating the results of tests performed on a roller rig to the field case.