The common cutworm, Spodoptera litura Fabricius (Lepidoptera: Noctuidae) is a major pest of soybean. Pheromone traps are used to monitor male adults, but the catch peaks do not always predict leaf damage in soybean fields. Thus, there is no accurate means of forecasting soybean damage, and insecticide is applied on the basis of farmers’ observations of actual damage in fields. To understand the occurrence of soybean leaf damage, the dispersal of S. litura in a soybean field in southwestern Japan was preliminarily investigated using a searchlight trap in comparison to a pheromone trap at one location, from August to mid-October in 2016–2018. To determine the relationship between pest arrival and crop damage, trap catch numbers and the number of soybean leaves damaged by 1st-instar larvae were examined by separately comparing raw numbers and cumulative numbers. The raw catch numbers of the two trap types in August and September 2016 and 2018 preceded subsequent damage peaks by an average of 5.3 days. This temporal difference coincided with the estimated duration of the egg stage plus an assumed mating period. Furthermore, the cumulative catch numbers of the two traps in August and September were linearly associated with cumulative damaged leaves in the same period in each year and in the three-year period. The coefficient of determination (R2) of linear regression between the cumulative catch numbers of the searchlight trap and the cumulative damaged leaves for the three-year period was much higher than that between the cumulative catch of the pheromone trap and cumulative damage. This suggests that soybean leaf damage is closely linked to the number of S. litura arrivals at the survey site. Thus, the searchlight trap captured S. litura arrivals better than the pheromone trap. As the linear regression function of the cumulative catch of the searchlight trap for the three-year period was tentatively correlated with a prefectural economic injury level for soybean fields, it might be feasible to predict S. litura-induced soybean damage using searchlight traps. The cumulative female catch number of the searchlight trap was also linearly associated with damaged leaves, but the coefficient of determination was generally lower than that with the cumulative total catch. The female ratio of searchlight trap catches in September was <0.5 in contrast to S. litura migrating overseas (>0.5). The advantages and disadvantages of the two trapping methods, as well as necessary further studies are discussed. Our findings provide a foundation for S. litura monitoring with searchlight traps.