The importance of specific testing has been shown in ice hockey performance and the understanding and connection between them may lead to adapting training and follow up in elite athletes. The aim of the present study was to compare two off-ice power tests with performances on ice with a complete professional hockey team. The broad jump test and the vertical jump test were used for the off-ice evaluations, while the SAS-40 and the repeated skate sprint modify (RSSm) tests were used as the performance variables. 31 players (3 goalkeepers, 10 defence players and 18 forward players) from the National Hockey League (NHL) performed during a summer camp. Study includes four evaluation sessions with a 48-hour time interval in each test. All dependent variables were expressed by their means and standard deviations form (Mean SD). Correlation analyses were conducted for each group to establish the relation between off-ice tests and the on-ice performance test. The two jumps were correlated (r2= 0.45 p<0.01). There was a significant correlation between vertical jump and best trial during SAS-40 for the defence player but not for forward players (r2= 0.76 p=0.02; r2= 0.21 p= 0.16 respectively). There was a significant correlation between broad jump and best trial during SAS-40 for forward players but not for defensive players (r2= 0.75 p<0.01; r2= 0.19 p= 0.28 respectively).Results of this study seem to show that the positions have specific ability needs in assessment and that needs to be taken into consideration for the choice of tests, mostly during the off-ice tests.