The decline of Pacific salmon (Oncorhynchus spp.) is well-documented, and the widespread degradation of freshwater juvenile rearing habitat is a primary contributor. In recent decades, river restoration has become a multi-million dollar a year industry, but salmon populations have not improved on a proportional scale. One of the most common forms of restoration involves the placement of large woody debris (LWD). These projects are typically implemented with the assumption that they will improve freshwater habitat, and an increase in salmon abundance will naturally follow. We synthesized data from 16 LWD placement projects throughout Washington State, dating back to 2004. These projects followed a multiple Before-After, Control-Impact study design, with paired "control" and "treatment" study reaches. Each study reach was monitored 1-4 years before restoration and 3-10 years after restoration, through physical habitat profiles and snorkel surveys. We used linear mixed effects models to evaluate the outcomes of these restoration projects, in terms of both habitat and fish response. As expected, habitat features responded positively, with increases in average residual pool depth, pool area, and habitat complexity. However, fish response was much more complicated. We looked for changes in both density and biomass of juvenile coho salmon (O. kisutch), Chinook salmon (O. tshawytscha) and steelhead (O. mykiss). Each species responded to LWD placement differently, with varying implications for restoration success. In most cases, however, density-dependent factors and variations in regional climate impacted fish response with equal or greater magnitude than did LWD placement. Our results suggest that LWD placement can be a useful tool for salmon restoration, but current levels of restoration will likely be insufficient to improve salmon populations.