BACKGROUND: We aimed to evaluate a comparative analysis of the prognostic value of the metastatic lymph node ratio (LNR) and pN (TNM) in stage III gastric cancer. PATIENTS AND METHODS: A total of 159 stage III gastric cancer patients with curative gastrectomy were retrospectively analyzed. Cutoff values for LNR were designated according to 25%, 50% and 75% percentiles, 0.07, 0.20 and 0.44 respectively. The LNR was divided into four groups as 0 > LNR1 ≤ 0.07; 0.07 > LNR2 ≤0.20; 0.20 > LNR3 ≤0.44; 0.44 > LNR4 ≤1. RESULTS: The mean age of the patients was 61.1 ± 11.3 years. Male predominance was apparent (73.6%). The 1-year overall survival and recurrence rates were 73.6% and 33.6%, respectively. The univariate cox regression analysis demonstrated age and LNR were the main variables that affected overall survival (OS) (p < 0.05). Harvested lymph nodes less than 16 did not affect OS (p = 0.255). The results of the multivariate cox regression analysis revealed that only LNR was an independent prognostic factor (P < 0.001), while pN was not (p > 0.05). Similar results, as with overall survival, could not be revealed clearly for disease free survival (DFS). CONCLUSIONS: LNR was an independent significant prognostic factor and superior to pN staging in predicting OS but not for DFS in stage III gastric cancer patients. The high LNR levels in our research were found to be associated with poor survival rates. The percentile system we used to determine cutoff values may be considered as a reliable method. Similarly, LNR also provides a reliable prognostic parameter in future staging systems to help guide treatment algorithm plans.