The aim of this study was to evaluate and compare postero-anterior (PA) cephalograms and cone-beam computed tomography (CBCT) images for the detection of mandibular asymmetry. Six asymmetric anonymous dry human skulls with visible chin deviation were available for this study. Metallic markers were glued on the anatomical landmarks to avoid identification error. PA cephalograms and CBCT scans were made by means of a standardized set-up. Each scan and cephalogram was measured three times by a single observer and the means used for analysis. Asymmetry was defined by the subtraction of the left side and right side measurements. CBCT was reliable [intraclass correlation coefficient (ICC) > 0.957] and very accurate (within 0.5 mm) in detection of all asymmetry. PA cephalograms were not accurate in detection of asymmetry of the mandibular ramus length, the mandibular body length, and the total mandibular length. PA cephalograms were the least reliable in determining the mandibular body length asymmetry (ICC = 0.686). The use of CBCT to detect mandibular asymmetry was validated with this study. CBCT images are very reliable and accurate for the detection of asymmetry and should be considered over conventional PA cephalometry when a chin deviation is present.