The culture of yellow perch Perca flavescens has received considerable interest in the Midwest and elsewhere in the USA as a result of the decline of wild populations and their high market demand. In this study, the effects of size‐grading on the production efficiency and growth performance of food‐sized yellow perch were determined in six 0.1‐ha ponds by grading fingerlings into three size‐groups (mean initial size): small (5.3 g), large (18.3 g), and unsorted (11.3 g). The mean percentage of marketable production from large‐group ponds was significantly greater than that from the small‐ and ungraded‐group ponds. The final biomass of large size‐groups was 77% and 27% greater than that of small and ungraded groups, respectively, although no significant differences in total production were detected among the three size‐groups. Size‐sorting of fingerling yellow perch did not result in improved growth of the small‐fish group. No significant differences in growth rates among the three groups were detected in either the first half‐year (winter–spring) or the second half‐year (summer–fall) period. The growth performance of the three groups indicates that yellow perch populations under culture conditions experience a bimodal growth pattern, with two distinct size‐groups. The coefficient of variation (CV) in both weight and length for the small‐group fish increased about 80%, whereas the CV in the ungraded and large groups showed some reduction. The results indicate that grading by size should be advantageous to those wishing to culture yellow perch to food fish size in the second year. The significantly different results for the percentage of fish reaching market size among the large (65%), ungraded (53%), and small (35%) groups suggest that perch should be selectively graded by at least the end of one year of intensive culture to increase the number of individuals with the potential to reach market size.