To tell you the truth, I really don't know much about UOttawa math. I'm fairly certain that Carleton's is better, though, and Carleton isn't much compared to UofT anyway, let alone Waterloo.
And I'm sure you're very well prepared (it certainly sounds like it!). I just criticize the Ontario high school math curriculum whenever possible, because it's really terrible. I was lucky that my school was actually fairly good, although ironically it's being closed now!
As to why I chose Carleton, there were a few reasons. I'm actually a math-physics double honours student, and while UofT has a strong physics program, I wasn't all that impressed with Waterloo's from my research (although I actually think that Waterloo's is probably more undergraduate-oriented than Carleton's - half of Carleton's research money goes to physics, but there are only sixty undergraduate students at all levels, and only ten in math-physics). As I said, I never really considered UofT all that seriously, although I'm not sure why not in retrospect. Its facilities are highly impressive compared to those at other universities in Ontario, and as I mentioned before its programs are strong throughout. I visited the Waterloo campus and absolutely hated it, but this was obviously a subjective judgement. I strongly suggest you visit any campus you're considering before deciding to attend, though. As well, I wasn't interested in co-op, something that 80% of Waterloo students participate in.
I also realized that I would be doing all the math that I wanted regardless of where I decided to go (and don't worry, my main interest does indeed lie in math, not physics). I also get paid ~$5000 a year above my tuition to go to school here, though I couldn't care less about it. Carleton's math program is by no means substandard regardless. There are roughly 200 students in the department, though most are in computer mathematics or something similar. While I'm sure that the classes would be more rigorous and quicker-progressing at Waterloo or UofT, and I do get bored quite frequently, it doesn't pose much of a problem, since there are always others to help with math, and I can of course develop my own problems (which my professors are more than willing to discuss, if only to evade the tedium of teaching the slow-progressing classes).