We compare Limit-Equilibrium (LE) and Discrete Element Method (DEM) analyses, in both 2D and 3D, for evaluating rock slope stability and runout. The comparison highlights the critical role of kinematics and progressive deformation in controlling rock slope failure. While 3D LE methods provide an estimate of the factor of safety, the failure surface and mode must be assumed a priori, and the rock mass is idealized as vertical slices (2D) or columns (3D). As a result, the kinematic response is artificially constrained and the influence of natural structure (e.g., joints, shears, and fractures) is largely neglected. Consequently, the computed factor of safety may not reflect the governing failure mechanism. In contrast, DEM does not require predefinition of a failure surface and can explicitly incorporate mapped discontinuities and fracture networks, enabling a more realistic representation of rock mass behavior. Using DEM, we show that increasing joint density (i.e., decreasing block size), without changing intrinsic material properties, reduces slope stability. This indicates that more highly fractured rock is less kinematically constrained and requires greater mechanical strength to remain stable. Furthermore, during slide initiation and propagation, internal fracturing and disintegration of the sliding mass can progressively reduce kinematic confinement, leading to accelerating displacements and progressive failure. A large-scale open-pit mine case study is presented to demonstrate how rock mass structure governs failure initiation and post-failure behavior, and to illustrate the importance of accounting for both strength and displacement-dependent degradation when evaluating fractured rock slopes.