Literally everything conventionally regarded as a hook - being a URM, a faculty brat, a development case, from a square state, with unusual athletic ability or a legacy - is something you’re born with and did nothing to earn. As for your second sentence, I would argue that legacies, on average, are actually in a much better position than, say, first-gens, to be high achievers in college, since they tend to be far better prepared academically and have superior support networks. A first-gen may transcend their background and, through grit and initiative, be admitted to a top-tier school, but plenty of them fail to adjust to what’s a dramatically different environment and either underperform or drop out. I’ll bet we all know people in that kind of unfortunate circumstance.
As @SatchelSF says, filling their classes with the “best” students academically, however defined, isn’t what these places are about. If they did that, they wouldn’t be able to satisfy various other institutional imperatives - there’d be a lot fewer URMs, athletes, first-gens and people with special talents, for starters. These places are looking to educate tomorrow’s leaders in many areas, not just academia/research. They also need access and resources for their mission, which alumni engagement is very helpful in providing.
Regarding MIT and Caltech, so often cited here, if, as with them, the preponderance of your operating budget is met by research grants, 90% or more of your undergrads are STEM majors, you’re focused almost to the exclusion of anything else to educating the leading scientists and technologists of tomorrow (a niche in which you have a dominant position), traditions and alumni relations aren’t major factors for you and, as such, you’re prepared to sustain a much narrower range of activities and departments than HYPS, diversity and legacy considerations probably won’t be very relevant to your thinking.