<p>While robotic-assisted total knee arthroplasty (RA-TKA) demonstrates high precision in coronal plane alignment, its accuracy in the sagittal plane—particularly in restoring the planned posterior tibial slope (PTS)—remains suboptimal, and the influence of coronal plane alignment phenotypes on sagittal alignment accuracy is poorly understood. This dual-center, retrospective study analyzed 194 patients undergoing RA-TKA with a constrained kinematic alignment strategy. Sagittal accuracy was assessed using ΔPTS, defined as the deviation between the planned and postoperative PTS measurements, and patients were stratified into high-accuracy (ΔPTS ≤ ± 2°) and low-accuracy (ΔPTS &gt; ± 2°) cohorts. Factors associated with inaccurate PTS reconstruction were evaluated in univariate and multivariate logistic regression. Only 40.7% of cases achieved PTS reconstruction within ± 2° of the planned target, and 72.2% of cases exhibited a postoperative PTS lower than planned. Univariate analysis identified older age, use of Chunli prosthesis, and Coronal Plane Alignment of the Knee (CPAK) Type III phenotype as significant factors. Multivariate analysis confirmed CPAK Type III as an independent risk factor for inaccurate PTS reconstruction (OR = 2.7, 95% CI: 1.056–6.901, <i>P</i> = 0.038). The accuracy of PTS reconstruction in RA-TKA was suboptimal in this cohort, with a prevalent tendency toward under-correction. The CPAK Type III phenotype was identified as an independent risk factor for sagittal inaccuracy, warranting phenotype-specific consideration in preoperative planning and heightened intraoperative vigilance. Based on our clinical experience, compensatory strategies such as adjusting the planned PTS according to individual patient anatomy are warranted.</p>