Hemophilia A (HA) consists of a genetic X-linked blood disorder. It is caused by a diversity of F8 gene mutations, with missense type being the most prevalent. Amino acid substitutions may impact physicochemical properties of the protein, providing an abundant scenario for investigation. This work evaluates 71 substitutions contributing to distinct patients’ phenotypes (mild, moderate, and severe), in terms of physicochemical alterations (PA - electrostatic potential, hydrophobicity, surface solvent-accessible/excluded areas, disulfide disruptions, and substitutions indexes), through an in silico strategy. PA information extracted from models fuels a hierarchical clustering analysis (HCA - independently and in combination) in an attempt to connect mutations and patients’ phenotypes. The combined use of PA over the analysis of single features seems to better reflect the impact of substitutions in severity degree, apparently in a domain-dependent way. Besides, a principal component analysis (PCA) identified prominent properties impacting clustering results for each domain. Electrostatic potential has a greater contribution to A3 than C1 domain clustering, probably by A3 involvement in FVIII activation, for example. The conjugated use of HCA and PCA is a powerful tool to assess if and what kind of structural features are involved in FVIII protein functionality impact and HA disease severity.