Significant advances have been made towards the design, fabrication, and actuation of dynamic DNA nanorobots including the development of DNA origami mechanisms. These DNA origami mechanisms integrate relatively stiff links made of bundles of double-stranded DNA and relatively flexible joints made of single-stranded DNA to mimic the design of macroscopic machines and robots. Despite reproducing the complex configurations of macroscopic machines, these DNA origami mechanisms exhibit significant deviations from their intended motion behavior since nanoscale mechanisms are subject to significant thermal fluctuations that lead to variations in the geometry of the underlying DNA origami components. Understanding these fluctuations is critical to assess and improve the performance of DNA origami mechanisms and to enable precise nanoscale robotic functions. Here, we report a hybrid computational framework combining coarse-grained modeling with kinematic variance analysis to predict uncertainties in the motion pathway of a multi-component DNA origami mechanism. Coarse-grained modeling was used to evaluate the variation in geometry of individual components due to thermal fluctuations. This variation was incorporated in kinematic analyses to predict the motion pathway uncertainty of the entire mechanism, which agreed well with experimental characterization of motion. We further demonstrated the ability to predict the probability density of DNA origami mechanism conformations based on analysis of mechanical properties of individual joints. This integration of computational analysis, modeling tools, and experimental methods establish the foundation to predict and manage motion uncertainties of general DNA origami mechanisms to guide the design of DNA-based nanoscale machines and robots.