Introduction: The purpose of this study was to investigate fracture resistance and the fracture modes of endodontically treated teeth restored with cast posts and cores in the presence of various configurations of incomplete ferrules. Materials and Methods: Fifty maxillary anterior teeth were endodontically treated and divided into five groups (n = 10) according to ferrule design: group complete ferrule (CF) had a 2-mm circumferential ferrule; group buccal-mesial-palatal (BMP) had a 2-mm ferrule on the buccal, mesial, and palatal sides; group P had a 2-mm ferrule only on the palatal side; group B had a 2-mm ferrule only on the buccal side of the tooth; and group no ferrule (NF) had no ferrule. Each tooth was restored with a cast post, core, and a Ni–Cr crown. All specimens were lingually loaded at 135° to their long axis in a universal testing machine until fractured. Fracture patterns were recorded. Data were analyzed using one-way analysis of variance and Tukey’s honestly significant difference (Tukeyi?½s HSD) test (α = 0.05). Results: The highest load to fracture was CF (534.33 ± 100.30 N), followed by BMP (467.71 ± 54.54 N), P (462.71 ± 54.92 N), B (330.48 ± 54.86 N), and NF (275.93 ± 28.35 N), respectively. There was no statistically significant difference in load to fracture among CF, BMP, and P and between B and NF (P > 0.05). Conclusion: A tooth with incomplete ferrule had lower fracture resistance than one with complete ferrule, but it was still higher than one with no ferrule. The presence of a palatal ferrule was more effective than a buccal ferrule in providing fracture resistance to palatal occlusal loads.