I think there are some misconceptions about the role good breeding practices play in the defects common to certain morphs.
Spiders are almost exclusively outcrossed. No one bothers to breed them together anymore, because no super-spiders have been produced. As a result, the wise assumption is that homozygous spider is lethal, and none ever hatch. Why lose 25% of a clutch by breeding spider X spider?
The spider head-wobble, head-tilt, and other ataxia signs continue. They crop up in the offspring of spiders that don't have the trait, too. At the moment, there is no good evidence that this trait is not directly linked to the spider gene itself--if it is, that means it cannot be bred out, period. You can breed spiders with less wobble to possibly minimize the trait, but never eliminate it.

The cinnamon duck-billing is ubiquitous to cinnamons--like the spider wobble, it's part of the cinnamon gene, and cannot be bred out. Like spiders, cinnamons are heavily outcrossed. The homozygous cinnamon naturally shows the trait to a more marked degree than the heterozygous form. Duckbilling seems to be harmless.

The kinking in caramels is not well understood, but it ALSO appears to be part of the caramel gene, as many attempts have been made to breed it out, and all have failed so far. However, I have recently heard that someone MIGHT have tentatively figured out how to minimize or eliminate this problem by incubating caramel eggs at much lower temperatures than most people use--86 to 87F. It's worth a try. If it doesn't work, I envision that caramels will probably slowly be replaced in favor by the very similar-looking Criders, which aren't prone to kinking.

There is no evidence to suggest that irresponsible breeding has anything at all to do with any of the above problems or quirks. Genetic mutations sometimes have more to them than appears on the surface alone.