Patterns of human motion were found to mirror cognitive processes in both psychological studies and applied research in Human-Computer Interaction (HCI). Notably, the behavioral entropy of movement trajectories of users was identified as a reflection of workload and fatigue across different settings, including Virtual Reality (VR). In the context of VR particularly, this metric is predominantly derived from the movements of VR controllers, denoted as the Entropy of Controller Movements (ECM). Despite its promising sensitivity and unobtrusive nature as a metric for human workload, ECM's application and proven efficacy in practical and authentic VR-based applications, such as industrial teleoperation platforms, has not been validated yet. Additionally, current literature predominantly features younger experimental samples, leaving unresolved the potential impact of age-related alterations in motor performance on using ECM as a workload metric. This study explored these dimensions by examining the relationship between workload and ECM among 15 young and 15 senior participants who manually operated an industrial robot within a VR environment. Participants were instructed to navigate the robot through a pick-and-place task by using their physical movements in VR. Our research identified unexpected variations in ECM values, particularly in older users, revealing an inverse relationship between movement entropy and task complexity in our scenario. High levels of behavioral entropy were also observed in younger participants. These findings unveil some criticalities in using ECM as a measure of workload in our VR-based industrial contexts, posing new questions regarding its applicability and effectiveness.