I think the most interesting point was that there are no (or very few) rules specific to each prize and that's it's mostly up to the idiosyncrasies of the unique small group of people who are part of the panel. So the prestige of a specific book prize is really superstition. It makes sense rationally, outside of having the same group of people judge the same book prize every year (and then they'd have to live forever?) there's no real way to keep consistent. And their tastes would I assume change from year to year. The other way would be to have a sort of constitution or set of rules, but that would be interpreted differently by each group.
I wonder if we'll have AI-judged book prizes in order to keep them more "value-free", you could have a group of authors write down a prompt file and then the AI could read hundreds of books in a day. But then there would be the implicit biases of the AI and I suppose most high-brow authors and readers would despise those particular biases.