These types of programming contests are all about whether you can come up with a clever algorithm for solving the problem. With the right algorithm, CPU time limits should not be a problem for any of the languages; but if you're using the wrong algorithm, you're going to run out of time regardless of what language you use.

(Note: I was once a contestant in the Australian Informatics Olympiad as well as the International Olympiad in Informatics, and in subsequent years I have also tutored kids participating in the contest. None of the kids were at the beginners level, however.)

I have competed in both the informatics olympiads and the ACM collegiate programming contests, and in my experience there was a much greater emphasis on algorithms in the informatics olympiad. Whether your algorithm was O(n^3) or O(n^4) did make a difference to the results.

That said, the choice of programming language only adds a constant factor to the runtime, so your point still stands.

Back when I was taking math classes, I used a program called TeXmacs for taking math notes. It's a (almost) WYSIWYG editor with a well-designed system for inputting equations, using sensible key bindings for all the common mathematical symbols and operators.

Despite the name, it's not related to either TeX nor Emacs. But it does use TeX's fonts - which result in equations that are almost as good looking as the ones from TeX.

It worked fast enough for me to keep up with all my math lectures.