Because you have no idea which of those ideas would have come to the brain of the debaters nor if your wording of it would help them word it better (or ironically worse) than they originally would.
The help you offer each side assists the enemy too, to structure and prepare for rebuttals.
But that's not really addressing my point, which is that the point at which assistance is provided matters, and not to whom it is provided. I'm not sure what you're talking about when you say "the enemy", but if you're informing both debaters, then the assistance you're providing should be even less concerning, since everyone is equally affected.
If you're concerned that the debaters aren't going to come up with their own ideas, I'd also say that that's problematic on a couple of fronts. First, debaters get ideas that come from the brains of other debaters in a dozen other ways, including PMs (if you make public comments about this off-limits, that's where these would go), forums, looking at previous debates, discussions IRL, etc. I don't see how this adequately prevents any cross-contamination in this regard. Second, I don't see the problem. If debaters were copy-pasting ideas from other debaters, that would be one thing, but there's recourse to punish that in voting. If they're taking basic concepts from others and finding personal ways to expand or reinterpret those concepts, why is that less valuable? I can speak from experience when I say that, when I was being coached and a coach would feed me an argument verbatim, I'd pretty thoroughly fuck up any explanation of that argument. So I'd say that the value of that information depends on an individual's ability to process it and convert it into a form that they can effectively utilize, which means it comes from "the brain of the debaters".
Additionally, the opponent of each may have structured things specifically to exploit that debater's style and flaws, failing to do so directly due to your assistance of the opponent which opened their eyes to something.
I don't get this. You're saying that an opponent can use a debater's style and flaws against them, which is true regardless of what information they're given. Style and flaws have little to nothing to do with the information that is provided to them, otherwise a source or two could substantially alter both. It seems like you've more got a problem with someone actually providing a structured argument that imposes their style and flaws on the debater, which would mean that they've taken verbatim (or close enough as to make little difference) the arguments that someone else gave them. Again, there's plenty of room for voters to punish that kind of plagiarism, so I don't see why moderators have to do anything about it.
Every Chess website that I know of rules out using a computer engine and getting live coaching from another user during a match with advice and move-suggestions specific to the game being played, since the elo rating is meant to represent the individual's skill alone.
Chess is very different from debate. There may be a massive number of possible moves for the game, but there are clear tried-and-true strategies that are used by top chess players and they work pretty much across the board. For that matter, you can impose that kind of strategy on a relatively new player and it can be nearly as effective (lacking some of the nuance) as with an actual master player. The same does not apply with debate for many of the reasons I already described, in that debate strategies can't be so easily superimposed on a new debater, copy-pasting debate arguments already has effective routes to be punished, and we're only talking about providing initial strategies before the debate (which, at best, would be like telling a chess player what the first few moves of the game should be based on the first move of their opponent). I don't see these as equally harmful.
