The purposes of this study were to determine the (a) magnitude of variability associated with certain eccentric variables (eccentric peak velocity, displacement, and ground contact time) during unilateral countermovement jump performance (vertical [VCMJ], horizontal [HCMJ], and lateral [LCMJ]); (b) differences between limbs as well as between jumps; and (c) relationship between jump performance and the eccentric variables of interest. The jumping ability in 3 directions (VCMJ, HCMJ, and LCMJ) of 30 field sport athletes were assessed. The variability (coefficient of variation [CV]) of the eccentric variables was the lowest for the VCMJ (CV = 8.5-10.6%) and the highest for the HCMJ (CV = 11.7%-13.5%). No difference was found between limbs in the variables of interest. Significant statistical differences (p < 0.05) across the different jumps were found in the eccentric variables (9.1-29.4%). No significant correlations between the eccentric variables and jump performance were found for the VCMJ; however, significant correlations were found between jump length and eccentric displacement or eccentric peak velocity in both HCMJ (r = -0.60 and 0.57) and LCMJ (r = -0.54 and 0.37), respectively. It appears that in the absence of instruction and standardization of the countermovement, eccentric phase kinematics remains relatively stable over trials. In terms of the between jump analysis, it appears that the eccentric phase kinematics are relatively unique to each jump and directionally specific and therefore may need to be trained accordingly.