Any guess on who is in consideration of the 2018 prize? I imagine those under consideration are already at top tier institutions.
Is no noteworthy mathematics being done at less prestigious institutions? (Loaded question I know) but it would be cool to see some undiscovered mathematician doing ground breaking stuff at a lesser known university.
I would argue that medalists' universities have been reasonably diverse. For example, in the US since 1990, medals have gone to mathematicians from Rutgers and three separate University of California campuses. These are prestigious universities to be sure, but it's not like all the medalists are from the same two or three places.
> it would be cool to see some undiscovered mathematician
I think "undiscovered mathematicians" are very rare, for the reason that mathematics is an inherently social activity. We are constantly reading each other's papers, visiting each others' universities to give talks, and meeting each other at conferences. If you do excellent work, then in general your peers will learn about it and spread the word.
People with high-potential are detected early on and can choose to work in the most prestigious institutions.
It's like a pro-athlete. You don't expect Michael Jordan to be playing with a small local team.
However I do think that there might be just a little too much focus on contest-math when it comes to identifying the country's best mathematicians. It's easily seen that the best contest mathematicians usually become the best professional mathematicians. In aggregate, though, most contest-math participants come from a very particular demographic (i.e. they live in big cities in wealthy areas, have very supportive/tiger parents (this is the key), go to magnet schools, obviously practice a lot, etc.). These people are still probably much more likely than average to become professionals, but it filters out a lot of bright people from less wealthy families or areas, or in regions where culturally math competitions are barely even a thing.
In general I think placement into top math undergrad/PhD programs (e.g. Harvard) should be less focused on contest math.
See Srinivasa Ramanujan for reference.
Because of this, Princeton offered Manjul Bhargava (another Fields winner) a tenure job right away without putting on the rigma role of tenure-track.