To the OP:
I think it depends on how the game is set up for whether or not that randomness works. My initial thought was "Does this level of randomness make combat frustrating or interesting?". Without knowing much about your game, I thought "frustrating" might be more accurate. However, I don't have any details on the core combat mechanics, so I can't judge for sure. Changing the "weakness" of an enemy doesn't really mean much if there's a lot of stuff that just does straight up damage without being resistant or weak to an element. Or, changing the weakness could be absolutely everything if it's the only real reliable way to deal damage.
I just don't have enough information to work with here in terms of how "replayable" something like that is.
My knee-jerk reaction is simply that it doesn't sound like it will make the game replayable. But, again, I know nothing about your game at this point.
I don't mean to pick on you over this, but I feel I need to weigh in here.
Yes, you have the ability to to do anything in rpg maker. Do 99.99% of rpg makers make these battle systems like this? Definitely not. I can count on 2 hands all the rpg maker games I've ever played that had battle systems that weren't extremely plain and boring, and even for those - they weren't replayable for me.
To be fair, this is true of most games. Not just RPG Maker. And, hey, we should be fair. I haven't played an RPG, AAA or otherwise, that had a "fun combat system" in the last 20 years. The last time I had fun in a combat system, we're talking Secret of Mana, because it was so unique for the time. Combat systems in RPG's have mostly just languished in this area of "Dark Souls Hard" or "Pointlessly Easy" and none of them have really been designed "for fun". Nobody is QA Testing their combat systems and going, "is this still fun 10 hours into doing this?".
What I was getting at, is if you are making a long form classic jrpg (which is the vast majority of what the engine is used for), replayability shouldn't be anywhere near the top of your list, and thus focusing resources on making things random to increase this replayability isn't going to do much for your game. Whereas all the effort you spent into doing that being put into polishing up the graphics, adding quests, refining the battle system, adding secrets etc - would vastly improve the game.
I don't really agree with this. The main reason being... It's the entire premise of my game. Well, sort of. Put simply, the "replability" is basically baked into the core design of the game since it's all about Player Choice. A game trying what I call a "True Choice" system, in which every decision actually changes the story, and not by an insignificant amount... like... an entirely different story in most cases.
Replayability, can be an important aspect of your games design. It isn't always something you consider "at the end".
But, that's why triage is important (as my signature line says). Is it important to the specific design of the game? If it's something you can remove and lose nothing, it's probably better to remove it. If you remove it and lose something vital, then you should do everything in your power to improve it.
I'd also like to state up front:
Polished graphics doesn't necessarily make your game more enjoyable or interesting. It can help, but it's something you do at the end of your dev cycle after all the "fun" is already in place. It's done to convey mood and aesthetic. You shouldn't worry about this until your game is already complete and ready to ship.
More quests also doesn't necessarily make your game more enjoyable or interesting. Quality over quantity. Players remember the most fun and interesting quests they've been on. They don't remember all the ones that were boring and slogs. In some cases, "less is more", as well.
Refined Battle Systems doesn't necessarily mean "fun" either. You can refine a boring Battle System to perfection and still have it be boring. "Finding the fun" is more important than "it works really well".
Finally... secrets... Eh. Implementation of the secrets as well as content of them is what is important. Not just having them.
I can name 1 time in my entire life that I ever played an jrpg, beat it, then immediately kept playing after beating it, and that was Chrono Chross. Beat the game, then messed around for a few hours in NG+ with the time remote thingy and then was done.
That's too bad, I have several. Chrono Trigger, Mass Effect 1, 2, and 3. Elden Ring. There's probably more, but those are the big ones.
I'm not really sure what your tastes are in games for why you'd decide to play New Game+, especially since I couldn't get through the opening of Chrono Cross before just shutting it off out boredom and you managed to enjoy it to the point of completion and beyond. So, I don't know what to recommend to you for good RPG's to play.
But, yeah, a good chunk of games are a "one and done", and I think that's okay.
All I'm saying, is that the effort put into making a game replayable, should really be just put into making a great game in the first place instead. Unless of course, this is just a really short game, in which case making it random is fine, but until the op clarifies, we don't know exactly what he's making.
It is important to make the game fun to begin with, but if "replayability" is part of "Core Design" of that experience, then it needs to be considered. As you've stated one good example of why it might be core design. Short game. It could also be a Roguelike. It could also involve time travel (a la Majora's Mask).
Replayability might be incredibly important to the OP for reasons we don't yet know.
The rub with this, is that making a battle system engaging / different from everything else requires you to know how to program. Most people use rpg maker because they don't know how to program, so it goes hand in hand. I'm not saying you can't make a balanced battle system, but it's going to be very basic unless you know how to program.
It actually doesn't require that at all. Here's all you need:
A combat system that revolves less around stats and more around gimmicks. There aren't all that many RPG's that operate that way, and do it well. You instantly "break the mold" doing that.
I'm using the "vanilla" engine. Here's what I've got for plugins:
Party Manager (so that I can lock character one into slot one and make them unable to be swapped out)
Here's some examples of interesting things you can do with combat:
Enemies can act/react based upon status inflicted upon them, based upon stats of the player, based upon conditions of the battlefield, etcetera. You don't even have to go into "troop commands" for this stuff either. This is just what you bake into the creatures on creation.
The engine is incredibly versatile for anyone who tried to accomplish things in it without immediately and instantly resorting to grabbing plugins for "ease of use".
Heck, I'm doing some "trickery" with my "Revenge" system that is on bosses. Everything with an element in the game inflicts an invisible Status on enemies. All enemies are immune to all of these Statuses, except the bosses. Who are weak to just one. It inflicts, always, when hit with the correct element. Then, based on that status effect, the boss uses a specific attack. The status effect lasts a single round, just long enough to get the attack off.
The player sees: "Fire Elemental hit with Water, 200% damage, "Fire Elemental seeks revenge!", "Fire Elemental slams the steam cloud into the party!", Entire party takes 18-25 damage (per character) and has a 30% chance to also be burned.
It's just an example of something I've done with the engine without programming at all. You can do a TON of stuff in the Troop tabs and within just the creatures to make things interesting.
You can even jump into your combat formulas and the stats to make even more interesting mechanics.
There's a lot here, if one goes looking. No programming required.
Every other part of the game is pretty easy to get by with 0 programming knowledge, but to do anything outside the box with battle system isn't possible without at least tweaking a bunch of yanfly plugins.
I find this is the attitude that makes a lot of people just seek out plugins as their first solution. Which, in turn, just leads to most people not knowing what the engine itself is capable of. I find this to be sad.
Doing something "outside the box", just requires the dev know how their tools work, what it's capable of, and then working within the confines of the restrictions. It also requires the dev have a pretty good imagination and have "problem solving skills" or, rather, "critical thinking" abilities. To put more simply, you need to be able to take what you learn in one place and apply it to another. The art of iteration. Without it, even with all the plugins and programming in the world... you won't create something "outside the box".
I can't count how many RPG Maker games I've played that use dozens of Plugins for combat and who are utterly all "very similar" and do nothing "outside the box". AAA games have the same problem.
If you want new and unique, you need to actually know how your tools work and have the ability to accomplish random nonsense within it. You also need to know what's been done and try to do something different.