I mean, there's a reason why I mostly post on the social forums. Competitive Pokémon can be fun to mess around with, but I would never take it super seriously for exactly the reason that you're alluding to.youre literally on a competitive pokemon forum
Regarding luck and skill: This is a fair point, and something that I wish I had acknowledged in my original post. In most of the roguelikes that I've played, the game is fundamentally skill-based; that is, you can always theoretically win with sufficient skill, even if you get a relatively poor set of starting circumstances. This isn't a universal truth — in Dicey Dungeons, for example, it's possible to lose before you have a chance to do anything about it because the first enemy got insanely good dice rolls — but it's consistent enough that I'll concede the point. However, I find that being put at an automatic disadvantage due to a poor seed still feels miserable, even when it's theoretically possible for me to overcome said disadvantage through some old-fashioned getting good. Given the popularity of the genre, this is clearly not an issue that everyone has, but it massively damages the appeal for me. One of my good friends used to be really into The Binding of Isaac, and I would watch him spam the reset button on the first floor until he got a great starting item when he was trying to get deep runs. It seemed like an awful time.Roguelikes aren't built around their procedural generation. They're built around a certain unique game mechanic. Procedural generation is the medium by which the mechanic is enjoyed. Any given roguelike game would be an objectively worse experience if you were playing through a set map, and the skill you have through your experience in a roguelike is rarely going to reach a point where your runs are based 100% on luck. Early on, luck doesn't matter because you've bad at the game and are bound to die anyways. Once you've cleared the game a few times, you'll have the skill where you can consistently make good runs/clear the game regardless of whether the items you get are good or not.
Take for example Noita. The game features 2 prominent mechanics: The wand system and environmental interactions. If the game had been built around a set map, there would be a 100% consistent way to beat the game every single time, and it would get really boring really fast. However, because your spells/wands and the environment is random, the game will provide a slightly new experience every time you play, and due to the nature of the game being that you kill everything fast (including yourself) the only thing stopping you from winning is your own skill. You can't blame anything on luck.
I've never played Noita, so I can't speak to how much worse it would be if it wasn't a roguelike. However, I reject the implicit assertion in your post that replay value is inherently good. A game might be able to provide me with a slightly new experience on each replay, but that doesn't excite me if none of those experiences are especially enjoyable. I'd much rather play a game with one set experience that's really good, even if it doesn't quite hit the same on replay. I'm also not convinced that there are any game mechanics that absolutely cannot be explored as well through deliberate design as they can be through procedural generation, but I haven't played every video game, so this isn't something I can really argue.