I find that the largest problem about this competition is that the people who can actually do this probably have no motivation to compete.
We all have jobs, lives, and cool stuff to hack on in languages we actually like using.
That didn't stop me from competing when I was a student. I got into the finals. The training I did still helps me to identify and solve algorithmic challenges to this day.
What do you mean? That's what the tiebreaker is for. Problem writing is hard, especially if you're trying to distinguish between very close competitors.
The difference between those different entrants is a huge amount of time. The winning unis made correct entries earlier, and had less incorrect entries.
Compare it to a marathon where there are 9 finish lines. The more you cross, the better you did. Everyone who crossed the 6th line is ranked by when they crossed it - or how fast they ran. They probably also got much closer to the 7th line before the time ran out.