https://www.cemc.uwaterloo.ca/contests/euclid.html
I see online that the average score (out of a 100) this year was 51. I recall that my score was 87, and, that the guy who beat me had 94. In 2010, there were two 98's. The online results go back only to 1998. It's a university entrance sort of test, held in the springtime, near the end of the traditional school year.
The only thing I much recalled the name of was the name of the scholarship. Back in the early 80's, at Waterloo University, it was called a Descartes scholarship. My grade-13 marks, either, didn't hurt. I retained my university transcripts, as well, but, nothing about the contest(s), or scholarship. The Euclid papers as graded weren't returned to the participants. I have no idea how that works now, online or not. Interestingly, the contests were put on lightly colored paper, a different color each year. Mine was pink.
Probably the best way to prepare isn't to obtain such questions at a university level, but, to dig up the old textbooks from the 40's, 50's, and 60's, from before calculators/computers, and intellectual laziness in general. As much as philosophy was declared dead by Hawking, has vanished from scientific papers, after the early era of Einstein, I envision the same fate for mathematics over the next couple of centuries, in the sense that we've lost the spirit of its pioneers. It's gotten to the point now that you're institutionally expected to pick up on some overly minute problem, and, then, hopefully, add a step.
The postdoc thread is from a professional math forum for university student math problems, with some latitude, but, lately, even when the higher-ups post something a little different, in a bit new direction, the fun stuff, the questions are "closed", and soon "deleted" if not improved upon. I put a question up, a few years ago, that one of the higher-ups wanted to help develop. One of his deceased colleagues had already put out a paper on the subject. My question, out of nowhere from my own work, involved a generalization of the paper. So, the retired university mathematician from Australia rounded up four or five or his friends on the forum to vote to reopen the question. Something that I've never seen happen before, on either of the StackExchange forums. The (top) moderator was somewhat, nay, totally, pissed. Ever since, I'm not allowed to put up anything fun even in the standard New Year's question thread about who can find the most complex way to express the numeral that is the new year. Anyway, there's the analogous math forum for research questions. I can understand not messing around on that one.