The construction of impassable dams severely affected many Pacific salmon Oncorhynchus spp. populations, resulting in reintroduction efforts that are now focused on returning anadromous fish to areas located upstream of these dams. A primary strategy for moving adult salmon and steelhead O. mykiss around a dam or multiple dams involves trapping fish downstream and transporting them to upstream areas (“trap and haul”) for spawning. We conducted a 4‐year radiotelemetry study to evaluate behavior and movement patterns of hatchery‐ and natural‐origin adult spring Chinook Salmon O. tshawytscha after a trap‐and‐haul program was implemented around three dams on the Cowlitz River, Washington. A multistate model was used to describe how factors such as origin, sex, release site location, and discharge affected transition rates to riverine areas where spawning habitat was located. Natural‐origin Chinook Salmon moved upstream from a reservoir release site and entered one of two rivers more quickly and in greater proportions than hatchery‐origin fish. Results from the multistate model indicated that transition rates from the reservoir to the Cowlitz River were 2.2 times higher for natural‐origin Chinook Salmon than for hatchery‐origin fish. About one‐half (49.6%) of the reservoir‐released hatchery‐origin Chinook Salmon moved upstream into the Cowlitz River or the Cispus River during the spawning period. The release of hatchery‐origin Chinook Salmon directly into these rivers increased the percentage of fish with river fates during the spawning period to 72.3–75.4%. Results from the multistate model showed that factors such as release site location, origin, day of year, and discharge were important predictors of transition intensities between specific locations in the study area. These findings illustrate the need to evaluate how salmon and steelhead respond to trap‐and‐haul methods, allowing for better management of reintroduction efforts in the future.