Soccer referees are required to keep up with play at all times to ensure optimal positioning in making key decisions. While the physiological aspects of soccer refereeing have been extensively reviewed, other key areas of preparation and performance have yet to be considered in detail. We present a contemporary examination of methodological considerations for the interpretation of referees' match activities, the validation of fitness testing and training protocols, match and training injury profiles, and the understanding and development of perceptual-cognitive expertise. A referee covers approximately 11 km during a match, with ∼900 m of high-speed running and, consequently, the demands of match play represent a significant physical challenge. The analyses of within-match activity profiles have attempted to assess the possible occurrence of referee fatigue, with equivocal findings. However, researchers have demonstrated that referees' physical performances are interrelated with those of the players during the same match. Therefore, the evaluation of referees' match activity profiles should be made in the context of the players' performances. High match-to-match variability in key variables, namely, high-speed running and sprinting, along with age-related reductions in match running are other factors that require due consideration when interpreting physical performances. Fitness testing is used by national and international referee governing bodies as part of their match selection criteria. Therefore, the tests need to reflect the physical task of refereeing, yet for the recent fitness tests introduced by the Fédération Internationale de Football Association--a 20 × 150 m high-intensity and 6 × 40 m repeated-sprint test - only the repeated-sprint test possesses the appropriate construct validity for assessment of match-related running capacity. Also, the performance standards of the tests have not been validated. Consequently, the scientific rationale behind the tests and their associated standards is not clear. Soccer referees dedicate a large proportion of their overall training to the development of physical capacity and researchers have demonstrated that high-intensity (>85% maximal heart rate) training protocols are effective for improving fitness and match running performance. These high training loads combined with increasing age could, in part, explain an incidence of non-contact match injuries (18 injuries per 1000 match hours) similar to players, with lower leg muscle strains being the most common type of non-contact injuries in referees. The implementation of injury prevention programmes along with the careful monitoring of training and match loads may help minimize referee's injury incidence. The perceptual-cognitive demands of soccer refereeing are significant, yet there remains limited research examining the perceptual and cognitive processes informing referees' decisions. As such, a three-step approach for the study of expertise in soccer referees is proposed. First, objective and reliable markers of decision making should be established, with due consideration to the development of naturalistic test situations while maintaining experimental control. Second, process-tracing measures can be used to identify the perceptual and cognitive mechanisms involved in accurate decision making. Finally, research is required to help understand the acquisition of superior decision making and whether such expertise can be developed via training programmes.