Why that assumption? Is there any reason why there should not be detrimental traits in any population? Its not like there is any direction to natural selection. Isn't assigning cause retroactively a flawed premise?
Because natural selection does give a direction. It is the very definition of it. Seriously, this is about the most basic concept you will find in evolutionary theories. Again, what natural selection does is reducing the allele frequency of detrimental alleles. Thought example: there is a an allele that causes slowness in some prey animal (and does not give any other advantages). Selective pressure in form of predators would result in more of the slow prey dying than the fast ones. Eventually the slowness allele traits would be reduced in frequency, as given everything else remaining constant, those with the alleles have a harder reproducing.
The detrimental traits do not need necessarily vanish completely, however. There are number of additional mechanism that may result in the persistence of a low frequency of the given detrimental traits (depending on how strong their influence on fitness is, for instance) that I do not want to discuss more before the basis is clear. I should have made clear that I did not mean that there are no detrimental traits at all, but that e.g. in cases of certain spiders they actually occur in higher frequency than they should if the fitness of the organism is more important than passing on genes. This can then be easily explained from the viewpoint of the genes. As long as they get passed on effectively it does not matter that they may have apparent detrimental effect on the organism.
Taken the above example, if the slowness allele also results in increased fecundity, it may not only remain in the population, but may actually increase in frequency, if the increase of fitness can off-set the fitness cost of being slow. Eventually then, for example, the prey in question may become a slower but fast reproducing species.
The important point is that natural selection is one of the major factors why not all organisms have identical reproduction successes. Understanding of this is essential to understanding evolution and it is one of the central aspects of Darwin's theory that have survived until now.
And btw., historically neo-Darwinism is an old concept. It arose shortly after Darwin and essentially refuted the Lamarckian mode of inheritance.
But I assume that this discussion is bound to be fruitless as long as the basics are not clear (textbooks anyone?).