Variants of the cranial closing wedge ostectomy (CCWO) may be used to manage cranial cruciate ligament deficient stifles by reduction of the tibial plateau angle (TPA). Concerns have been raised about predictability of achieving target TPA, limb shortening, axis shift, and distalization of the tibial tuberosity. Landmarks derived from tibial radiographs (n = 50) by 5 observers, along with outlines of the cranial and caudal proximal tibial cortices, were used to simulate six CCWO variants. Tested variants were Slocum-type (n = 2), modified (n = 2), isosceles, and anatomical-mechanical axis (AMA)-based CCWO. Observer specific landmarks were used to model pre- and postoperative errors in TPA measurement. Mean landmarks were used to compare key outcome variables between the variants, as well as modelling errors in positioning the wedge apex. Variants differed markedly in TPA, axis shift, mechanical axis length changes, and tibial tuberosity distalization, with more proximally positioned wedges performing better than distally positioned wedges. Errors in TPA identification appeared partially self-correcting due to altered axis shifts. Craniocaudal shifts in wedge apex position had the most severe effects on outcome variables, especially in shorter tibiae with greater initial TPA values. Recent CCWO variants are likely to achieve a satisfactory postoperative TPA with limited axis shift. The AMA-based CCWO technique appears to influence MA length and tibial tuberosity position the least, and apart from the effect on TPA where no advantage could be discerned, it appeared more robust in the face of wedge apex positioning errors than the other techniques.