I think you're under estimating the relationship between reputation and users. What I imagine people were trying to get across is that you shouldn't up-vote an answer because, say, Jon Skeet wrote it, and you recognize his name. You should up-vote or down-vote a post based on the quality of that post, and every post should stand alone.
In essence, this is a means to enforcing that any person is the sum of his or her actions. If you make a lot of good posts, you probably deserve a lot of reputation points. If you make a lot of bad posts, well, not so much.
So yes, reputation is meant to result because a post is of high quality, and not because of anything based on the user, but that's just a way of making it more accurately representative of the user.
Note, of course, reputation isn't perfect. We all know this. But again, it's the best we can do. Quantifying understanding of a field, the SE system, and the general Q&A format is hard, and reputation turns out to be, on mass scale, a pretty effective way to do it.
As for your main question, then, about why there's a rep limit per day, that mostly has to do with privileges. Privileges focus on the user's ability to understand and work with the Stack Exchange model. "You understand what makes a question on-topic, so now you can vote to close." However, if someone pops on and posts a single, super-high-quality post that gets a hundred up-votes, that suddenly skyrockets that user up to 1000 rep points, after they've been with SE for only a day. It's unlikely that a user with that little experience on the system will have very good knowledge of how to use the privileges awarded during that time. By putting a limit of 200 rep points per day on there, we can be certain in a majority of cases that nobody will jump through privilege levels without having at least a basic understanding of the system.