In ergonomics, digital mock-ups of workstation design in virtual reality environments are often used to evaluate the occupational risks related to a working condition. However, the effectiveness of such technology, particularly in terms of human performance can be questionable. In the present study, we investigated the pattern of movement variability in a simulated assembly task and presented an index of task performance that can highlight the differences in terms of motor control strategy between a real assembly task and its virtual simulation. The movement variability was decomposed into goal-equivalent and non-goal equivalent components and the ratio of these components was introduced as an index of performance. The ratio was greater in the real than in the virtual environment. This study highlights the differences in terms of motor control strategies in real and virtual environments. Practitioner Summary: A simulated assembly task was implemented in a real and a virtual environment. A performance index in light of the motor control theory was introduced. This index could discern a lower performance in the virtual compared with the real environment. The developed approach can provide an objective index to investigate the system fidelity in virtual environments.