The aim of this retrospective study was to describe the use and reliability of a numerical scoring system (modified Huddart/Bodenham) for the measurement of maxillary arch constriction in patients born with unilateral cleft lip and palate (UCLP), and to compare and contrast the new scoring system with the current methods of measuring treatment outcome, the Goslon and 5-year-old indices. Dental study models of 50 patients aged 5 years, and 50 patients aged 10 years, were scored using the 5-year-old and Goslon indices, respectively. Four examiners scored each set of models using the modified Huddart and Bodenham system, and repeated the scoring one month later. The intra- and inter-rater reliability of the numerical scoring system was assessed using the Kappa (κ) statistic. The scores using the new method were correlated with the 5-year-old and Goslon scores using Spearman's (ρ) and Kendall's (τ) rank correlation coefficients. There was a high level of intra-rater reliability for both the 5 (0.87: incisors, 0.91: canines, 0.88: molars) and 10 (0.9: incisors, 0.84: canines, 0.78: premolars/molars) year models. The weighted κ values measuring inter-rater reliability were above 0.85 and 0.74, respectively, for all examiners. There was a statistically significant correlation between the modified Huddart/Bodenham scores and both the 5-year-old and Goslon scores in all cases (P < 0.001). It is concluded that the modified Huddart/Bodenham system provides an objective and reliable assessment of maxillary arch constriction. It has a high degree of correlation with the recommended standards, but is more versatile and sensitive to inter-arch discrepancies.