You can be right but have an unpopular opinion, similarly to how you can have good taste but maybe nobody knows about your art and therefore you can remain "undiscovered" forever, so experts might never have a chance to review your work and deem it good. You can also be right under most circunstances but a bigger factor making your judgement sound but wrong.
In engineering (programmers, managers, product managers) you have this a lot where you're trying to promote people with good judgement to own large codebases, or big projects, or teams of people, but identifying those is something that is hard to do.
It'd be interesting to have art experts quantifying art in similar ways that we quantify engineering work. If experts can classify art, surely there's a mental checklist they go over, even if they never tried to put it to paper. I imagine it'd be a mixture of from 1-10 "how much of a sense of awe did the piece cause you" and from 1-10 how "good are the porportions" or whatever, but this is where I struggle. Perhaps it's more complicated than that and there's a decision tree of each "criteria" but surely there must be one, the same way we'd assess a software engineer or a product manager or an engineering manager for good judgement in the quality of their work.