Research has indicated that teams who cover greater distance during matches and complete more basic tasks such as passes, tackles and shots are more successful. Identifying means of increasing these physical and technical outputs is therefore a significant opportunity for performance enhancement. There has been limited research performed on hockey, especially at the elite level. An issue that is even more relevant given that in the past 15 years the sport has undergone some significant rule changes including the introduction of unlimited substitutions. With sixteen players able to be used per match and eleven players on the field at any one time the coach can make substitutions as frequently as desired to try and maximise the overall performance of the team. The objectives of this thesis were to use methods of performance analysis to measure the physical and technical outputs of players during elite hockey and to specifically measure the impact of differing substitution strategies on the physical and technical outputs of strikers during match play. Three striker conditions were assessed; three strikers with no substitutions, four strikers with a moderate amount of substitutions; and, five strikers with a large amount of substitutions. Five matches between the New Zealand men’s hockey team and Tasmania state representative team were played over eight days. Physical outputs of players were measured using portable GPS units and heart rate monitors and technical aspects of match play were measured using team performance statistics and a set of technical criteria which awarded points to strikers for each contribution they made to the game based upon a scale of effectiveness. Average total distance covered during 70 minutes by a position was 8160 ± 428m of which 479 ± 108m (6.1%) was performed at speeds greater than 19km.h-1. Within this high intensity distance were 34 ± 12 sprints per player with an average duration of 3.3s. Average match HR was 85.3 ± 2.9% HRmax and average peak HR was 96.3 ± 2.7% HRmax. Distance covered decreased by 6.2% between the 1st and 2nd halves and there was a trend of decreasing distance in both halves when total distance was broken into five-minute time periods. When assessing the impact of substitutions on the performance of strikers it was found that there were no significant differences in physical outputs between conditions with total distance (S5 = 8414 ± 125m, S4 = 8422 + 34m; S3 = 8282m) and distance covered at speeds greater than 19km.h-1(S5 = 701 ± 46m, S4 = 685 ± 28m, S3 = 723m) being similar. Substantial differences were found in technical outputs between the substitution conditions with more strikers and greater substitutions offering a better total output than less strikers and fewer substitutions (S5 = 241 ± 35, S4 = 207 ± 38, S3 = 173) but statistical significance between conditions was also not found. In conclusion, the results suggest that although substitutions are not a means to increase the physical work of strikers they do appear to be a way to enhance the contributions that strikers are making to the game.