Background: The use of the minimally invasive surgery robotic esophagectomy (RE) for esophageal cancer (EC) has been increasing over the past decade, as it reduces morbidity and mortality compared with other surgical approaches. However, although the technical feasibility and safety of RE for EC have been reported, few studies have evaluated oncological outcomes. This retrospective cohort study aimed to determine the long-term outcomes of RE for EC.  Methods: Twenty-four consecutive patients who underwent RE with total mediastinal lymph node dissection for EC between 2009 and 2013 were enrolled in this study. The short- and long- term outcomes, including the 5-year overall survival (5yOS) and 5-year recurrence-free survival (5yRFS) rates, were examined retrospectively. Results: With a median follow-up of 37 months, the 5yOS was 70.8% and the 5yRFS was 62.5%. Tumor recurrence was found in nine (37.5%) patients, and liver metastasis, the most common site of recurrence, was seen in five (20.8%) patients. Multivariate analysis demonstrated that pathological tumor stage (pT ≥ 2) and certain late complications of surgery as measured by the Clavien–Dindo (CD) classification (pneumonia CD grade II and stenosis CD grade III) were significantly associated with 5yOS, while pT ≥ 2 and pN ≥ 1 were significantly associated with 5yRFS. Anastomotic leakage was the most common complication, observed in seven (29.2%) patients. The median number of harvested lymph nodes was 41 and 13 patients (54.2%) had node-positive disease. Conclusions: We have shown that RE for EC is not only safe and feasible but also has encouraging oncological outcomes. We also demonstrated that late complications are significantly associated with long-term survival. Confirmation in a prospective study would assure the place of RE in the management of EC with curative intent.