Identifying measures which accurately quantify reactive balance adaptation during walking is essential to understand how emerging perturbation-based gait paradigms impact stability over the course of an intervention. These perturbation paradigms have shown promise in reducing falls for numerous clinical populations, however tracking progress in objective terms throughout an intervention remains challenging. Whole body angular momentum (H) may be particularly suited to detect subtle adaptations in the reactive balance response and is applicable within numerous perturbation environments. We assessed the ability of young healthy adults to adapt to varying intensities of discrete, unexpected, treadmill-based perturbations directed mediolaterally, anteriorly, and posteriorly during a single session while ambulating at their comfortable walking speed. We assessed corrective step length and width, trunk deviation and flexion, peak H over a stride, peak-to-peak differences in whole-body angular momentum over a stride (H_R), and the participants ability to maintain their H trajectory within two standard deviations of their normal (PNT). Measures derived from H, particularly H_R and PNT, demonstrated significant changes with increasing intensity and repetition. Corrective step length and width, trunk deviation and flexion, and peak H also demonstrated significant, but weaker, differences with increasing intensity and repetition. Derivatives of H are sensitive to changes in intensity and repetition, particularly when assessed as peak-to-peak differences and ability to maintain a normal trajectory over a stride. These measures may be utilized to detect changes in reactive balance during perturbation-based gait paradigms.