This study aims to explore the reliability and validity of essential performance metrics derived from a modified 505 Change of Direction (CoD) test. Twenty-seven physically active male participants (age: 18.77 ± 1.73 years) who engaged in three to five training sessions per week were assessed using both standard and modified sprint and CoD protocols. The evaluation focused on sprint time, CoD time, total time, and the Limb Symmetry Index (LSI). The modified protocols demonstrated good reliability for sprint and CoD times (ICC > 0.8, CV < 3%) and high validity, with strong correlations between standard and modified tests for sprint (r = 0.59–0.80) and CoD times (r = 0.58–0.80). The total time metric showed excellent reliability (ICC > 0.9), supporting its utility as a comprehensive measure. Reaction-time inclusion increased variability (CV > 7%) but enhanced ecological validity by reflecting real-world conditions. However, the LSI exhibited lower reliability (ICC < 0.6), indicating the need for further refinement. Combining multiple performance measures, the modified protocols streamlined assessments, reducing fatigue and redundancy without compromising precision. These findings highlight the potential of integrating multidimensional agility metrics in athletic evaluations, bridging the gap between controlled testing and competitive demands. Future research should explore broader populations and sport-specific stimuli to enhance applicability.

Short sprints are predominantly assessed using timing gates and analyzed through parameters of the mono-exponential equation, including estimated maximal sprinting speed ($\mathrm{M}\mathrm{S}\mathrm{S}$) and relative acceleration ($\mathrm{T}\mathrm{A}\mathrm{U}$), derived maximum acceleration (MAC), and relative propulsive maximal power ($\mathrm{P}\mathrm{M}\mathrm{A}\mathrm{X}$), further referred to as the No Correction model. However, the frequently recommended flying start technique introduces a bias during parameter estimation. To correct this, two additional models (Estimated TC and Estimated FD) were proposed. To estimate model precision and sensitivity to detect the change, 31 basketball players executed multiple 30 m sprints. Athlete performance was simultaneously measured by a laser gun and timing gates positioned at 5, 10, 20, and 30 m. Short sprint parameters were estimated using a laser gun, representing the criterion measure, and five different timing gate models, representing the practical measures. Only the MSS parameter demonstrated a high agreement between the laser gun and timing gate models, using the percent mean absolute difference ($\%\mathrm{M}\mathrm{A}\mathrm{D}$) estimator ($\%\mathrm{M}\mathrm{A}\mathrm{D}$ < 10%). The MSS parameter also showed the highest sensitivity, using the minimum detectable change estimator ($\%\mathrm{M}\mathrm{D}{\mathrm{C}}_{95}$), with an estimated $\%\mathrm{M}\mathrm{D}{\mathrm{C}}_{95}$ < 17%. Interestingly, sensitivity was the highest for the No Correction model ($\%\mathrm{M}\mathrm{D}{\mathrm{C}}_{95}$ < 7%). All other parameters and models demonstrated an unsatisfying level of sensitivity. Thus, sports practitioners should be cautious when using timing gates to estimate maximum acceleration indices and changes in their respective levels.

Informacione tehnologije (IT) u sportu uključene su u sve segmente i to pr- venstveno u rad sportista i trenera, a onda i zvaničnika i gledalaca