To facilitate the correct diagnosis and treatment planning, a range of diagnostic orthodontic indices have been developed. Pont’s index, as one of them, was documented to estimate the ideal maxillary dental arch widths from the sum of crown widths of maxillary incisors. This retrospective study aimed to evaluate the applicability of Pont’s index to Iraqi individuals. Fifty stone dental models for 25 females and 25 males were retrieved from the archives of the Department of Orthodontics in Baghdad College of Dentistry. The dental arch widths at the level of 1st premolar and molar areas in addition to the mesiodistal crown widths of all teeth were measured using an electronic vernier. The estimated arch widths were compared to the actual ones using paired sample t-test. The results showed highly significant differences between the actual and estimated maxillary inter-premolar and intermolar widths with over-estimated values in both genders. New formulas were developed to make Pont's index more reliable, and the newly estimated widths were non-significantly different from the actual one. The original Pont’s index is an unreliable and inapplicable method for arch width prediction. A specific modification of Pont’s index is developed and can be applied for more reliable results.