Haptic shared control is a promising approach to increase the effectiveness of remote handling operations. While in haptic shared control the operator is continuously guided with assistive forces, the operator’s response to forces is not fully understood. This study describes the development of a computational model of a human operator controlling a teleoperation system based on feedforward control. In a simulation, the operator’s response to repulsive forces in free-space motions was modeled for two degrees of freedom, for two operator endpoint admittances (estimated by means of closed-loop identification techniques). The simulation results show that similar repulsive forces lead to substantial discrepancies in response when admittance settings mismatch; wrongly estimated operator admittances can lead to assistive forces that are either not perceived, or deflect the combined system of human operator and telemanipulator. It is concluded that assistive forces should be tailored to the arm configuration and the type of task performed. In order to utilize haptic shared control to its full potential, it is required to study, measure and quantitatively model operator behavior for teleoperated tasks in more detail.