I've heard this argument a thousand times now.

Why does rating exist, then? If CF is not about rating, what's the point of keeping such a feature?

I'll tell you why. Many people come here to train. Ad that involves not only solving problems, but also seeing their progress. When one sees that his/her rating starts increasing after some time, it gives a good feeling. "Solving problems helped me improve", they think. And "I can get even better if I keep this up", further fueling them in training and doing serious competitions.

The "cheater" accounts do the exact opposite. One finds out that despite solving several problems successfully, the rating only decreased. "What's the point of trying if I only drop?" is the question that appears instead. And this can even lead to someone who has the potential to become an excellent coder (but, for example, has difficulty implementing their ideas, which such sites can help greatly), never utilizing their potential.

So there you have it. Humans are competitive creatures. We always strive to get to a better place, and this is not an exception.

It's more complicated than this. I'm not an admin so correct me if I'm wrong, but this is what I know about it:

1. Contestants are divided into rooms at registration. Which means: if there are many fakes, no hacks.

2. After the registration ends, an expected place for every participant is calculated. Then, rating is calculated based on the difference between actual and this expected place. So registered users not in the ranking change the rating. This is fine as long as it's just a few accidental cases, but imagine what thousands would do.

So there is more to it than just submitting.

There are other ways to make your point. Like, saying it directly instead of spamming the contest system just to be noticed.