Conservation of skeletal structure and unique body ratios in fishes facilitated the development of truss analysis as a taxonomic tool to separate physically-similar species. The methodology is predicated on the measurement of across-body distances from a sequential series of connected polygons. Changes in body shape or condition among members of the same species can be quantified with the same technique, and we conducted a feeding experiment using yellow perch (Perca flavescens) to examine the utility of this approach. Ration size was used as a surrogate for fish condition, with fish receiving either a high (3.0% body wt/d) or a low ration (0.5%). Sequentially over our 11-week experiment, replicate ration groups of fish were removed and photographed while control fish were repeatedly weighed and measured. Standard indices of condition (total lipids, weight-length ratios, Fulton's condition) were compared to truss measurements determined from digitized pictures of fish. Condition indices showed similarity between rations while truss measures from the caudal region were important for quantifying changing body shape. These findings identify truss analysis as having use beyond traditional applications. It can potentially be used as a cheap, accurate, and precise descriptor of fish condition in the lab as shown here, and we hypothesize that it would be applicable in field studies.