Background: While systemic chemotherapy for gastric cancer has evolved significantly in the last decade, the role of surgery for liver metastasis remained controversial. The aims of this study were to examine the trend in outcomes of patients with gastric cancer liver metastasis (GCLM) undergoing resection at our center, and to define the optimal selection criteria for resection. Methods: One hundred and twenty-two consecutive patients who underwent liver resection for GCLMs with curative intent at our institution from 1993 through 2021 were included. The patients were grouped into 2 eras: era I, 1993 to 2010; era II, 2011 to 2021. Our criteria for liver resection included absence of extrahepatic disease, and the limited number of liver metastasis (often 3 or less). Median follow-up was 34 months for all and 71 months for the survivors. Results: There were 63 patients in era I and 59 in era II. The patients in era II received chemotherapy prior to liver resection more often (68% vs 41%, p = 0.004), and their primary tumor was less advanced (T3/4; 66% vs 85%, p = 0.021) with less regional lymph node metastasis (median No. of positive nodes; 4 vs 1, p = 0.034) compared to those in ear I. Median and 5-year disease-specific survival (DSS) were better in era II (not reached, 55%) than in era I (35 months, 35%, p = 0.025). In the multivariate analysis for the entire cohort, elevated CEA (> 50 ng/ml), primary T4, and primary at the proximal stomach emerged as independent factors associated with shorter DSS. Number of these risk factors are highly predictive for long-term survival; among the patients without or with a single factor alone, 5-year DSS was 50%, whereas none of those with multiple risk factors survived beyond 5 years (p< 0.0001). Conclusions: Long-term outcomes for patients with oligo-GCLM has improved in the modern chemotherapy era. Liver resection should be considered unless they carry multiple risk factors.