As an essential contributor to thumb and hand function, it is necessary to establish metacarpal (MC) kinematics to track thumb health and determine the efficacy of treatments. In the literature, several hand dorsum coordinate systems (CSs) have been utilized for MC angle calculations. These systems were determined using varied motion capture marker setups. This work determined the differences in MC angle ranges across hand dorsum Cartesian coordinate systems, and identified the CSs with the least difference compared to reference. Motion capture with reflective markers on the wrist and metacarpals was used to determine MC angle ranges (flexion/extension, abduction/adduction, internal/external rotation) of six hand dorsum CSs during thumb circumduction. A custom-built reference system was used to identify the difference associated with each of the CSs. Statistics were used to identify differences between the six CSs for all angles. CS difference in flexion/extension, abduction/adduction, and total difference was minimized by a CS established from a rigid marker pod on the third metacarpal. A CS established using markers on the wrist and second metacarpophalangeal joint yielded the second least total difference. This study showed that CS choice matters, as clinically relevant differences in MC angle ranges were detected between CSs. If left unconsidered, changes in patient status may be missed or overexaggerated, potentially leading to less effective treatment. This work provided guidance for informed CS selection to ensure proper patient evaluation through robust data collection.