OBJECTIVE: To establish a standardized stance position for balance testing based on average preferred foot placement, and to compare this to existing standards. DESIGN: Cross-sectional study. BACKGROUND: It has been shown that the orientation of the feet can have a marked influence on the results obtained during balance testing, prompting the need for standardized foot positioning. Unfortunately, current recommendations do not appear to address concerns about the potential effects of 'uncomfortable' or 'unnatural' foot positions on the control of stabilizing reactions. METHODS: The present study identifies the central tendency and variance of the preferred stance width and foot angle, measured from foot tracings in 262 subjects (89 male, 173 female) ranging in age from 19 to 97 years. RESULTS: Results revealed a great degree of variability in preferred stance width and angle across subjects, although mean differences due to gender or age (</= 50 versus > 60) were small. The average preferred foot position was 0.17 m between heel centres, with an angle of 14 degrees between the long axes of the feet. Existing standards for stance position lie well outside the range of preferred foot placement. CONCLUSIONS: The wide range of preferred foot placements clearly highlights the need for standardization during balance testing. A standard based on average preference would reduce potential effects of 'uncomfortable' or 'unnatural' foot positions, in comparison to existing standards. RELEVANCE: The development of a standardized stance position for balance testing is necessary since foot placement can influence stabilizing reactions. The present results provide a standardization based on average preferred foot placement, which will minimize between-subject variability and reduce the abnormal constraints that are placed on subjects by existing standards.