Based on the relevant data of the top 8 trading partners of China's soybean import trade from 1980 to 2016, the stochastic frontier gravitational model was used to study the influencing factors of China ’s soybean import trade while the regression analysis was performed after the variables were divided into basic variables and trade inefficiency terms. The results show that six variables, including the GDP of the two trading parties, the soybean output of the two trading parties, the Chinese population, and the bilateral exchange rate, are positively related to China's soybean import trade. Four variables, such as the population of the exporting country, the absolute distance from China, a landlocked country or not, adopting the GM safety management policy, have negative correlation with China's soybean import trade. Based on the empirical results, China's soybean import trade potential and trade efficiency from major trading partner countries are calculated, and relevant policy recommendations are proposed to strive for a more favorable trade environment of China's soybean import and to ensure China's soybean trade security.