In peacocks and peahens, the peahens prefer mating with peacocks having a large and bright tail. But how is having a large tail an indicator of genetic "fitness" (in survival terms) ? If having a large and bright tail is not related with genetic superiority, then what is the point of having the mating ritual?
Darwin thought quite a bit about this. He even said "The sight of a peacock makes me sick". https://www.ncbi.nlm.nih.gov/pubmed/10824193 He added to his ideas about natural selection to include sexual selection. Traits selected by females, even when they are non-adaptive for males, will increase the males fitness (as defined by his number of offspring). There is also the idea that perhaps possessing such an outrageous and non-adaptive tail proves that you a good mate, the same way we would think that a guy who could win a sprint while carrying barbells must be superior.
Evolution of traits under sexual selection is relatively complex subject that isn't easily summarized in a post. In short, these are the most important concepts you will want to read about; Fisherian runaway (aka. runaway selection), the sexy son hypothesis), the good genes hypothesis and Zahavi handicap principle.
All these concepts are somewhat related and the semantics can quickly become a little complicated and has been subject to a number of papers (e.g. Kokko et al. 2003). My understanding of the semantics details is limited. To my understanding, the Fisherian runaway selection and the sexy son hypothesis are very similar and the good genes hypothesis and Zahavi hanicap principle are related. What you will be most interested in is the Fisherian runaway selection.
I will below just cite some popular sources and I will leave to you the pleasure of understanding and misunderstanding the differences between these concepts :)
Imagine a bird population in which females choose mates at random. Males with slightly longer tails fly a little more adeptly, avoid predation, and so, survive better than males with slightly shorter tails. In this situation, a gene for female choosiness (longer tail = sexier) will be favored, since by choosing a long-tailed male she will have sons with longer tails. This trait will spread through the population until most males have long tails and most females prefer long-tailed mates. So far so good.
However, once this has happened, the process may run out of control, until the male trait becomes so exaggerated that it is disadvantageous. In other words, female preference, instead of survival advantage, may begin to drive the evolution of ever-longer tails, until males are encumbered by showy plumage that no longer helps them avoid predation.
Imagine another bird population in which females choose mates at random. Some males in the population have better genes for survival than others, but it is difficult to tell whether a male has good genes or not. In this scenario, long tails make it more difficult to survive�they are costly to produce and maintain. Because they are so costly, only males with good genes have the extra resources to produce them. In this situation, a long tail is an indicator of good genes. A gene for female choosiness (longer tail = sexier) will be favored, since�by choosing a long-tailed/good gene male�she will have sons with good genes. This trait will spread through the population until most females choose long-tailed mates and males that are able to produce long tails are favored.
If females choose males with “long and costly” tails, they are guaranteed to get good genes! If they choose males with “short and cheap” tails, they may get good or bad genes.
The sexy son hypothesis in evolutionary biology and sexual selection—proposed by Ronald Fisher in 1930—states that a female's ideal mate choice among potential mates is one whose genes will produce male offspring with the best chance of reproductive success. This also implies that a potential mate's capacity as a parental caregiver or any other direct benefits the father can offer the mother, such as nuptial gifts or good territory, are irrelevant to his value as the potential father of the female's offspring. Fisher's principle means that the sex ratio (except in certain eusocial insects) is always 1:1 between males and females, yet what matters most are her "sexy sons"' future breeding successes, more likely if they have a promiscuous father, in creating large numbers of offspring carrying copies of her genes
"Good genes" theory proposes that females select males seen to have genetic advantages that increase offspring quality. Increased viability of offspring provides compensation for any lower reproductive success that results from their being "picky". The good-gene hypothesis for polyandry proposes that when females encounter better males than their previous mates, they re-mate in order to fertilize their eggs with the better male's sperm.
Clearly, the good genes theory is not the one you are interested in with the peacock tail example.
The handicap principle suggests that reliable signals must be costly to the signaler, costing the signaler something that could not be afforded by an individual with less of a particular trait.
Note that the term "genetic superiority" is never used in evolutionary biology. In fact the term sounds very social science related and make me think of the ugly side of eugenics. We usually talk about "fitness difference", "low fitness" and, "high fitness".