Abstract This study evaluates a hypothesis for the role of vertical wind shear in deep convection initiation (DCI) that was introduced in Part I by examining behavior of a series of numerical simulations. The hypothesis states, “Initial moist updrafts that exceed a width and shear threshold will ‘root’ within a progressively deeper steering current with time, increase their low-level cloud-relative flow and inflow, widen, and subsequently reduce their susceptibility to entrainment-driven dilution, evolving toward a quasi-steady self-sustaining state.” A theoretical model that embodied key elements of the hypothesis was developed in Part I, and the behavior of this model was explored within a multidimensional environmental parameter space. Remarkably similar behavior is evident in the simulations studied here to that of the theoretical model, both in terms of the temporal evolution of DCI and in the sensitivity of DCI to environmental parameters. Notably, both the simulations and theoretical model experience a bifurcation in outcomes, whereby nascent clouds that are narrower than a given initial radius R0 threshold quickly decay and those above the R0 threshold undergo DCI. An important assumption in the theoretical model, which states that the cloud-relative flow of the background environment VCR determines cloud radius R, is scrutinized in the simulations. It is shown that storm-induced inflow is small relative to VCR beyond a few kilometers from the updraft edge, and VCR therefore plays a predominant role in transporting conditionally unstable air to the updraft. Thus, the critical role of VCR in determining R is validated.