Objective To evaluate the test-retest reliability of the Ballistic Push-Up (BPU) test and to establish predictive models for upper-limb strength and power in young males. Methods 71 male college students performed assessments of upper-limb bench press maximal strength (1 Repetition Maximum, 1RM), bench press power, and two BPU tests with a 48-hour interval. BPU test data were recorded using a 3D force platform and motion capture system to calculate concentric metrics such as peak force (PF) and mean velocity (MV). Test-retest reliability of the BPU was evaluated using the intraclass correlation coefficient (ICC), and Pearson's correlation coefficient was used to assess the relationship between BPU metrics and upper-limb strength and power. Stepwise regression analysis was conducted to establish predictive equations for upper limb strength and power. Results BPU metrics showed good test-retest reliability (ICC = 0.764 - 0.935). PF and MV, along with body weight, were effective predictors of bench press 1RM in young males: bench press 1RM = 0.109 × PF – 16.772 (R2 = 0.790, Standard Error of the Estimate [SEE] = 8.17 kg); bench press 1RM = 1.511 × body weight + 87.15 × MV – 110.136 (R2 = 0.767, SEE = 8.60 kg). PF and body weight were also predictors of bench press power: bench press power = 2.755 × body weight + 0.287 × PF – 17.351 (R2 = 0.620, SEE = 46.1 W). Conclusion The BPU test demonstrates good test-retest reliability, and PF and MV from the BPU test can be used to predict upper limb strength and power in young males.