Increased levels of randomness and variability are introducing new uncertainties into power systems that can impact system operability and reliability. Existing planning and operational methods for assessing operability and reliability are primarily deterministic, therefore, ill-suited to capture randomness and variability. This work proposes an approach to model, propagate, and measure the impact of uncertainties in power flow caused by stochastic grid resources. Using system sensitivities, statistical circuit analysis methods, and convex optimization, we demonstrate that we can accurately estimate the worst-case impact of stochastic resources on the health of a grid. We compare our method's performance to Monte Carlo analyses, and our results demonstrate an increase in efficiency of more than 2 to 3 orders of magnitude for the same probabilistic accuracy.