Maintaining an uninterrupted supply of electricity is a fundamental goal of power systems operators. However, due to critical outage events, customer demand or load is at times disconnected or shed temporarily. While deterministic optimization models have been devised to help operators expedite load shed recovery by harnessing the flexibility of the grid's topology (i.e., transmission switching), an important practical issue that remains unaddressed is how to cope with the uncertainty in demand encountered during the recovery process. This paper introduces two-stage stochastic programming models to deal with uncertain load parameters with known probability distribution, and one of these also incorporates conditional value-at-risk (CVaR) to measure the risk level of unrecovered load shed. The models are implemented using a scenario-based approach where the objective is to maximize load shed recovery in the bulk transmission network by switching transmission lines and performing other corrective actions (e.g., generator re-dispatch) after the topology is modified. The benefits of the proposed stochastic programming models are quantified via comparisons with a deterministic mean-value model, using the IEEE 118-bus test case. Experiments and discussion highlight how the proposed approach can serve as an offline contingency analysis tool.