This is really hard for me to say...but Theo is correct on all counts.
The fact that competition will weed out inferior genes is completely true. Keep in mind however, that inferior genes in one pond or lake may be slightly different than what constitutes inferior in another pond, based on variances in water chemistry, prey type, etc. This is why it's so important to have at least a little variability in the original genetics. If you were to start with only two parents than the inferior gene is on the same allele of ALL progeny, so fish are competing with other inferiors. Competition doesn't work at all in that situation. Now if you have maybe three or four of both males and females and they "mix" well in the first spawn, then the defective genes are on different parts of the DNA strands depending on which parents you came from. Therefore, you have competitive pressure from some fish that are superior genetically. I'm no geneticist, but I suspect that just a few parents of each sex is enough to set competition into effect and eliminate inbreeding depression as a concern.
Theo's also right about inbreeding as a form of reinforcing some postive traits, but this is impossible to build into pond reproduction dynamics because you can't put effective pressure on defective individuals. You'd have to be able to examine every single progeny AND have the knowledge to know what constituted a good and a bad offspring. Not very practical when dealing with tens of thousands of offspring from a single parent.
