According to the nursery size hypothesis, flatfish recruitment is constrained by nursery area. Thus, if resource selection models can be shown to accurately predict the location and geographic extent of flatfish nursery areas, they will become important tools in the management and study of flatfish population dynamics. We demonstrate that some resource selection models derived previously to predict the presence and absence of juvenile flatfishes near shore were applicable to the broader continental shelf. For other age-species groups, derivation of new models for the continental shelf was necessary. Our study was conducted in the western Gulf of Alaska (GoA) during October 2011 on four groups of age-0 juvenile flatfishes: Pacific halibut (Hippoglossus stenolepis), arrowtooth flounder (Atheresthes stomias), northern rock sole (Lepidopsetta polyxystra), and flathead sole (Hippoglossoides elassodon); and three groups of age-1 juvenile flatfishes: northern rock sole, flathead sole, and yellowfin sole (Limanda aspera). Sampling occurred at 33 sites across the continental shelf. Fish were collected using a 3-m beam trawl, and a midwater trawl. Environmental data were collected on sediment composition and water temperature and depth. Many of the age-species groups co-occurred in the Shumagin and Barnabus sea valleys; however, age-0 arrowtooth flounder occurred at more locations than other juveniles, perhaps due to a relatively broad tolerance of environmental conditions and to the utilization of midwater habitat. Thus, the large nursery area of arrowtooth flounder may be one reason why they are currently the most abundant GoA flatfish. In fact, among all species, mean recruitment at age 3 increased with the percent occurrence of age-0 juveniles at the 33 sites, a proxy for relative nursery area, in accordance with the nursery size hypothesis, suggesting that mean recruitment among GoA flatfishes is structured by nursery size.