The problem with having this discussion is that most people tend to assume the possible negative impacts of hatchery raised fish can be seen after a couple of generations. Most of the predicted problems, which are genetic related, cannot be identified in a decade or less. The evolution of a population is shown in a temporal (time) trend, not in comparisons of the physical attributes between hatchery and wild fish.
Wild fish are not more "superior" than hatchery raised fish. They are simply better because they are products of natural forces. Organisms that are selectively bred by human (yes, we conduct selective breeding on them. I doubt we already have a full understanding of sexual dimorphism of salmonids and can conclude that large males will always breed with large females, etc) are products used to match our demand and best interest.
The juvenile phase between the egg and fry stages is an important phase in the entire salmonid lifecycle. During this time in the rearing streams, weak fish are eliminated due to predation, inability to hunt as well as others, competition for space, etc. These factors are absent in a rearing pond where hatchery raised fish are pellet fed. By minimizing mortality rate and raising the fish to the smolt stage, some weak fish that shouldn't survive during the rearing stage are receiving a free ride. When these fish return to spawn, the "weak" alleles (kinds of genes) might be passed on, which may shift the genetic frequency of a population and weaken it.
For those with a genetic background. If a disease that only occurs in juvenile salmonids is the product of a recessive allele, then offsprings that are homozygous recessive would not survive past the juvenile stage in the wild, but would do perfectly fine with the aid of hatcheries. If there is a higher number of fish that are homozygous recessive return to spawn, then it might result in a higher juvenile mortality in the next generation due to higher number of homozygous recessive offsprings.
As mentioned earlier, the clear effect on the genetic components of a salmonid population cannot be seen in one generation, but can be identified by looking at the trend over ten, twenty, thirty generations. Well, by that time (40, 80, 120 years), it would be pretty hopeless if there is indeed a problem. That's why the issue of hatchery vs wild fish continues to emerge in the scientific community.
So why do we raise more salmon than steelhead in hatcheries? Pretty easy answer really - demand. The sportfishery for salmon is much more heavily participated than for steelhead.
From an angler's point of view, more fish is always better. From a biologist's point of view, that's not necessarily true. Some hatchery proponents tend to lack a good understanding of fishery biology and push for higher production of fish to satisfy their own self-interest. If a population has already reached carrying capacity, then no matter how much more hatchery raised fish you dump into it, the stock will still not revive itself. We should really focus our money and effort more on producing and enhancing spawning and rearing habitats that have been lost due to pollution, urbanization, development and logging.