Trends in pairs figure skating have shown that increasingly difficult jumps have become an essential aspect of high-level performance, especially in the latter part of a competitive program. We compared a repeated jump power index in a 60 s repeated jump test to determine the relationship of repeated jump test to competitive rank and to measure 2D hip, knee, and ankle angles and angular velocities at 0, 20, 40, and 60 s. Eighteen National Team Pairs Figure Skaters performed a 60 s repeated jump test on a large switch-mat with timing of flight and ground durations and digital video recording. Each 60-s period was divided into 6, 10-s intervals, with power indexes (W/kg) calculated for each 10-s interval. Power index by 10-s interval repeated measures ANOVAs (RMANOVA) showed that males exceeded females at all intervals, and the highest power index interval was during 10 to 20 s for both sexes. RMANOVAs of angles and angular velocities showed main effects for time only. Power index and jumping techniques among figure skaters showed rapid and steady declines over the test duration. Power index can predict approximately 50% of competitive rank variance, and sex differences in jumping technique were rare.

• The repeated jumps test can account for about 50% of the variance in pairs ranks.

• Changes in technique are largely due to fatigue, but the athletes were able to maintain a maximum flexion knee angle very close to the desired 90 degrees. Changes in angular velocity and jump heights occurred as expected, again probably due to fatigue.

• As expected from metabolic information, the athletes' power indexes peak around 20s and decline thereafter. Coaches should be aware of this time as a boundary beyond which fatigue becomes more manifest, and use careful choreographic choices to provide rest periods that are disguised as less demanding skating elements to afford recovery.

• The repeated jumps test may be a helpful off-ice test of power-endurance for figure skaters.