The aim of this study was to compare the effect of self-control, frequent and reduced feedback of KR on children's learning skill of throw targeting in Aleshtar city. For this purpose, 60 subjects were selected and assigned through step-by-step random method to three groups of self-control, frequent, and reduced. They performed the task of targeting through throwing in two steps of acquisition and retention with a delayed interval of 24 hours. The self-control group was divided into two groups of high (more than 30%) and low frequency (lower than 18%) after the acquisition phase. In both phases of acquisition and retention, inferential and descriptive statistics were used. In descriptive statistics, central indexes and dispersion of variables, in inferential statistics in acquisition phase, analysis of variance (series of practical efforts) 6 × 4 (how to present feedback) and in retention phase, one-variable analysis of variance and Tukey post hoc test as well as SPSS 11.5 software were used (P?0.05). The results indicated no significant difference among these groups in acquisition phase. In retention phase, frequent group had better performance than other groups. The comparison between coach-oriented and subject-oriented practice methods shows that frequent coach- oriented practice method leads to better results for children.