I was under the impreswsion that both Down's syndrome and sycle cell were defects. Firthermore, if genetic accidents, somehow turned into biological features by "natural selection" is a strawman, all you have to do is explain the real origin of biological features according to NeoDarwinism.
Your impression was (slightly) incorrect.
Biological features are not inherently positive or negative. That's a subjective judgment that can change depending on the environment.
sickle cell is the perfect example; while it causes problems, it
does prevent malaria. If you're in an area rife with malaria-carrion mosquitoes, sickle-cell is a
positive feature. If you're in an area where malaria basically does not exist, sickle-cell would be a
negative feature.
You could make the same judgment with animal camouflage. A leopard blends into its normal environment, so its fur patterns are considered a "positive" feature. But if you put the leopard in a completely different environment like, say, a rain forest or a city street, it would stand out like a sore thumb.
Natural selection does not "turn" a genetic "accident" into a feature. Genetic "accidents" simply happen, and are the direct cause of any given feature. Natural selection determines whether any given feature is passed onto the next generation, but has nothing to do with the expression of a genetic trait - the genetic change
is the cause of a change in a feature, all by itself. Multiple successive mutations passed on will build on each other to make more and more meaningful changes over time. This means that while individual mutations are relatively slight and sometimes have no real observable effect, over many successive generations those changes can build into something
very observable, like a change in camouflage patterns.
Natural selection is an incredibly easy idea to grasp, so I'm surprised at how many people fail to do so. Natural selection is a blind process by which features that help an organism survive to reproduce will gradually become more prevalent in a given population compared to features that do nothing or actually hinder the organism's survival. This determination is not intelligent, but is rather wholly subjective dependent on the organism's environment, and a change in that environment can make formerly "negative" features into helpful ones and vice versa.
The
Peppered Moth is another good example of this in action.
The moth originally had a light background with dark speckles camouflage pattern to blend in with the trees it tended to land on. Over a few generations, the British population of these moths actually changed colors - the soot from the Industrial Revolution was so thick that it darkened most surfaces including trees, and so the moths former camouflage suddenly made then stand out. The moths with mutations that made them darker were suddenly harder for predators to find, and over just a few generations the camouflage pattern of the entire population changed to one that is almost uniformly darkened.
Even
more interesting is that, as environmental laws cleaned up British industry and the soot began to disappear, the moths have begun to change back to something like their original color patterns!
It's literally an observed case of natural selection guiding random mutation.
So again - you were incorrect to say that natural selection turns genetic accidents into features. Genetic accidents (mutations) are the cause of features themselves. Literally changing a gene (or several, as most features are not determined by a single gene in their entirety)
will change a feature in the organism regardless of whether that organism survives or not.
Natural selection is what we call the process by which those features are either encouraged or discouraged by environmental pressures like predators or food availability. Mutations cause the new or altered features, and natural selection causes those features to become more or less common in the population as a whole. The changes in frequency of given features in a population is evolution.