Reliable determination of elastic parameters, such as Young’s modulus and Poisson’s ratio, is critical for the design of underground excavations. Conventional estimation methods, including the Secant 50 and Tangent 50 methods, frequently incorporate non-linear segments of the stress–strain curve, resulting in high variability and unreliable characterization of the elastic response in hard brittle rocks, where the linear-elastic regime is mobilized earlier than in weaker rock types. This study introduces the Tangent 30 method, a reliable approach grounded in the crack damage criterion, which isolates the linear-elastic region of the stress–strain curve by calculating the elastic modulus over a ±5 % strain range centered at 30 % of peak stress, well below the crack initiation threshold. Laboratory testing on Adelaide black granite under uniaxial and triaxial conditions demonstrates that the Tangent 30 method yields significantly more consistent estimates of elastic modulus and Poisson’s ratio, with coefficients of variation reduced to 4 %, compared to 10–20 % for conventional methods. The method also better captures the increase in stiffness with confining pressure, particularly when volumetric strain is considered. Its broader applicability to other hard, brittle rock types highlights the Tangent 30 method as a practical and potentially more reliable approach for elastic parameter estimation in geotechnical design.