I, for one, will quite vehemently argue that LC is actually good. A part of the reason is personal, that I’m far better at problems when I get a lot of time to think about them: for instance, I can do many med-hard problems in a day but many times even medium level problems take me a full hour to code. To someone like me LC is the highlight of the month.
But alright, I agree that’s not much of a reason. Let me come the the other reasons I have for LC to stay. The level of final few problems you’ll find on LC far exceeds most of the problems you find in other contests. This means that while in cook-off etc the solution mostly involves “standard” techniques which are “expected”, in the long challenge every time there’s some new lesser known techniques exhibited. This means that the learning is better in LC.
Furthermore, it is a long challenge: it’s expected some people will “cheat” on the challenge (I’m not justifying it, just saying it is inevitable). While I’m totally against a copy pasta type of cheating, I believe that if a person reads an editorial of some other problem and gets the idea to solve a problem in LC, it is not “exactly” cheating. As long as you aren’t copying someone else’s code you are learning something in the process. And in most cases, especially for the harder half of the LC, there arent many articles/blogs that immediately tell you how to solve it. For instance you may not be so fresh about segment trees and may read up something on it, but the fact the you realized you had to use segment trees was the major hurdle of the problem. This implies that unless you know what you want to do, you will not get any material online that helps you do it.
This of course, assumes you do not copy-paste the solution from someone else’s code. Those who do usually never make it to the top 100 or so, so the people who come to LC for competing don’t lose out much in the process.