It is known that the elderly population is increasing in relatively high numbers, generating concerns about the care they need. A form of seniors remain active is through the learning of new motor skills. In this sense, few studies involving the elderly population and the contextual interference effect have been found, particularly, studies that used sports skills as task. The objective of this study was to investigate if there is the contextual interference effect on sports skill learning in elderly. We selected 40 old persons (65-80 years old), physically active, which were divided randomly into two groups: random varied practice; and practice varied in blocks. The task was the throwing practiced at the bocce game, in which the ball should reach three targets at distances of 2, 4 and 6 meters. The practice consisted of 120 attempts divided into two practice sessions. Two Retention tests were conducted (10 min and 24 hours) and also two Transfer tests (24 hours), being performed with the preferential hand and also with the non-preferential hand of the participant, with the target at 5 meters. The performance measures used were the radial error, the standard deviation of radial error and frequency of gross errors. Kinematic measures were also used, such as amplitude, average speed and peak of acceleration in the displacement of the ball on the forwardswing; and also, speed of release of the ball. Comparisons between the practice groups, found no difference in the radial error or the standard deviation of the error (p > 0.05), but in the frequency of gross errors, there was a difference between the groups only at the acquisition phase (p < 0.05). In addition, the inferential analysis on retention and transfer tests, both the average speed as the speed of release, showed that the elderly of both groups have changed the speed (p < 0.05), probably trying to adjust the movement. Pearson correlations were performed between performance and kinematic measures and there was only a weak correlation between the radial error and amplitude on the transfer test with the non-preferential hand (p < 0.05). Overall, the results showed that the groups have similar performance on both practice structures (random and blocks) and, therefore, there was no effect of contextual interference in learning this sport skill by the elderly. However, it can be argued that the proper control of force needed in this task may require more practice time to be acquired.