The aim of this study was to evaluate whether a mental training protocol could improve gymnastic jumping performance. Seventy-two rhythmic gymnasts were randomly divided into an experimental and control group. At baseline, experimental group completed the Movement Imagery Questionnaire Revised (MIQ-R) to assess the gymnast ability to generate movement imagery. A repeated measures design was used to compare two different types of training aimed at improving jumping performance: (a) video observation and PETTLEP mental training associated with physical practice, for the experimental group, and (b) physical practice alone for the control group. Before and after six weeks of training, their jumping performance was measured using the Hopping Test (HT), Drop Jump (DJ), and Counter Movement Jump (CMJ). Results revealed differences between jumping parameters F(1,71)=11.957; p<.01, and between groups F(1,71)=10.620; p<.01. In the experimental group there were significant correlations between imagery ability and the post-training Flight Time of the HT, r(34)=-.295, p<.05 and the DJ, r(34)=-.297, p<.05. The application of the protocol described herein was shown to improve jumping performance, thereby preserving the elite athlete's energy for other tasks.