The unique challenges associated with sampling of macroinvertebrates in flashy urban streams create a methods gap. These streams form isolated pools for much of the year, interspersed with spates that scour and deposit excessive amounts of sediment. Commonly used stream grab sampling methods, such as nets and Hess and Surber fixed-area samplers, work well in wadable streams with perennial flow. Deployed samplers (Hester-Dendy, gravel tray) can be used in waters with or without flow. We evaluated three methods which don't require stream flow: modified Hester-Dendy (MHD), gravel tray, and bucket (a type of cylinder grab sample method), for their potential use in bioassessment of a project involving daylighting of a 180-m culvert on Congress Run, a flashy urban tributary to Mill Creek in Cincinnati, Ohio. Method efficacy was measured using three criteria: usability (level of effort and recoverability of samplers), variability, and community retrieval completeness. The bucket method required the lowest level of effort and had the highest sample recovery. The bucket sampler had the lowest variability for most metrics, including the critical metric of taxa richness, with a coefficient of variation (CV) of 20.9%. The MHD and tray samplers had taxa richness CVs of 42.9% and 53.9%, respectively. The bucket sampler also had the lowest CV (27.4%) for a multi-metric index. The bucket sampler performed best with respect to community retrieval completeness, with higher pooled and average taxa richness. The total number of taxa collected from all the replicate bucket grab samples (42) was greater than that collected by the HD and tray samplers combined (27). Multivariate analyses showed significant grouping with respect to methods and location. This study supports the bucket grab sampler method as a candidate for sampling of flashy urban streams.