While designing a dynamic hand orthosis to assist during activities of daily living, the designer has to know whether a concept will have sufficient grasp performance to support these activities. This is often estimated by measuring the interaction force at the contact interface. However, this requires a prototyping step and limits the practicality of comparing several concepts in an early design stage. Alternatively, this study presents and compares basic static and dynamic models to numerically estimate grasp performance. This was applied on an exemplary concept for a hydraulically operated hand orthosis grasping a circular object. The models were validated with an experimental set-up that does not require sensors at the contact interface. Static and dynamic model results were almost identical, where the static model could be around 10 times faster and is generally more robust to a high contact stiffness. Both models were unable to make accurate quantitative predictions, which is believed to be due to differences in used contact stiffness. However, the models were able to make correct qualitative comparisons, making it a valid method to compare and choose concepts in an early design stage.