Ice sledge hockey is a popular paralympic team sport where players rely entirely on their upper body to propel themselves rapidly across the ice surface. The isolated and repetitive poling movements provide a good model for examining upper-body sprint ability and the related movement and strength characteristics. Therefore, the purpose of the present study was to investigate the relationship between upper-body maximal strength, power, and sprint performance in ice sledge hockey. Thirteen male ice sledge hockey players from the Norwegian national team performed three 30-m maximal sprint tests recorded by fixed light sensors. The best 30-m time for each subject was used for further analyses, and the sprint was analyzed more in detail for the first and last 10-m split times and kinematics (cycle length and rate) using photocells and 2-dimensional video analysis. One repetition maximum (1RM) strength and peak power were assessed in the bench press, bench pull, and pull-down exercises using a barbell and a linear encoder. Both 1RM strength and peak power for all the 3 strength exercises correlated significantly with the total sprint time (-0.75 < r < - 0.86, all p < 0.005), the first (0.60 < r < 0.72, all p < 0.05), and the last (0.74 < r < 0.83, all p < 0.05) 10-m split times in the 30-m sprint test. There were no significant relationships between sprint kinematics and 1RM strength and peak power. Overall, these results demonstrate that there are close relationships between upper-body strength, power, and sprint performance in highly trained athletes and that the ability to produce propulsion and high frequency in combination is important for the sprint abilities in ice sledge hockey.