We really have no idea how much wild fish stray. Normally, all we mark are hatchery fish. We sample fisheries, we sample hatcheries, we sample spawner index areas.

But, how many streams have a significant number of wild smolts marked? How many streams (rivers/creeks) have traps to observe upstream migration? We sample a miniscule amount of where wild fish go.

Yes, we do know that they home, and home "well". We know, based on how the fry migrate (and what controls the migration) that sockeye have to home very well. The rest of the species not so much, as their freshwater requirements are much more general.