We investigated the variability of electrode positions for a multi-channel, custom electrode cap placed onto participants' heads without taking scalp measurements. The electrode positions were digitized in a three-dimensional space for 10 young adult participants on three separate occasions. Positional variability was determined for 15 selected electrodes within the three-dimensional preauricular-nasion (PAN) coordinate system and from this system, angular coordinate variability was also determined. The standard deviations of the 15 selected electrodes ranged from 3.0 to 12.7 mm in the PAN system. These data resulted in a variability of 2.0 degrees to 10.4 degrees among the angular coordinates. The measurements indicated slightly greater variability of electrode positions compared to studies when electrodes were placed using scalp measurements. The implication of this study is that the use of electrode caps may not be appropriate when electroencephalographic (EEG) or evoked potential (EP) techniques depend on accurate electrode placement. Additionally, if a longitudinal study is performed, electrode locations should be checked to ensure that they conform with previous sessions.