Background: Drinking alcohol during pregnancy can result in severe developmental disorders in the child. Symptoms of the fetal alcohol spectrum disorder (FASD) comprise growth deficiencies, abnormal facial phenotype and damage or dysfunction of the central nervous system. Numerous diagnostic methods for facial phenotyping in FASD exist, but diagnoses are still difficult. Our aim was to find additional and objective methods for the verification of FAS(D). Methods: Three-dimensional dental models of 60 children (30 FAS and 30 controls) were used to metrically determine maximum palatal depths at the median palatine raphe. Three-dimensional facial scans were taken, and vertical distances of the face were measured at five defined facial landmarks (FP1-FP5) for each child. Results: Mean palatal height, total facial length (FP1-FP5) as well as FP4-FP5 did not significantly differ between the FAS group and the control group. Comparing vertical facial subdivisions, however, resulted in significant differences for distances FP1 to FP2 (p = 0.042, FAS > controls), FP2 to FP3 (p < 0.001, FAS < controls), FP3 to FP4 (p < 0.001, FAS > controls) and FP3 to FP5 (p = 0.007, FAS > controls). Conclusions: Metric vertical measurements of the face can be used as additional objective criteria for FAS diagnoses. However, no significant differences were reported for palatal depth evaluation in the specific age range tested in the present study.