Because of the lack of an accurate sequence-based algorithm for estimating the evolutionary distance between distant homologous protein sequences, the phylogenetic tree construction problem remains unsolved. Although incorporating structural information can enhance phylogenetic analysis, the limited availability of high-resolution structure data has hindered its practical application. In this study, we propose a sequence-based evolutionary distance algorithm termed sequence distance (SD), which innovatively incorporates site-to-site correlation within protein sequences into the distance calculation. SD is correlated with the similarity of the protein structure, which is more conserved than the sequence during evolution. In protein superfamilies, SD can effectively distinguish evolutionary relationships both within and between protein families, producing phylogenetic trees that closely align with those based on structural information. Overall, the development of SD will significantly advance phylogenetics, providing researchers with a more accurate and reliable tool for exploring evolutionary relationships.