This paper investigated if a pursuit tracking task could be used to quantify the loss of motor skills due to Parkinson's disease (PD) by using system identification methods. A human-in-the-loop experiment consisting of PD patients and a healthy age-gender matched control group was conducted at the Erasmus University Medical Center. A pursuit display combined with a quasi-random multisinus target signal and single integrator dynamics was used. Such a tracking task makes it possible to model human controller behavior to estimate human parameters, such as the control gain and time delays. The performance of patients was found to be worse compared to the control group. The control gain was found to be significantly higher for the control group compared to patients. Patients had a significantly higher neuromuscular damping compared to the control group. Therefore, a tracking task could be used as a tool to quantify the loss of motor skill due to PD. This might improve the early detection and the treatment of PD.