Again - it's a perfectly functional and good way to play RPGs. But not the only one.
...
Immersion is a very specific and limited agenda in roleplaying for a few reasons. Firstly, it requires the engagement and cooperation of all present; players who do not "play the game" are "disruptive" and GMs that don't take all the extraneous stuff off the players are "poor GMs". The style necessarily requires a good deal of GM control and empathy in equal measure - and the GM also has to handle all or most of the "system stuff", too, often leading to headaches or similar after long sessions of play.
I don't particularly disagree with any of that, except the "specific and limited" part. Any character can be played in this fashion, and any event can occur. I see no limits other than the processing power of the DM's brain. But no one ever said it was easy.
Now, is some other paradigm of gaming any easier, that's a question.
I can't say I've ever found shared narrative control easy, largely because of confusion and conflicting goals and roles. However, since I did not learn that way, I certainly allow the possibility that I am simply not very skilled at it.
There are, indeed, better games for a number of varieties of such play. But then, there are also better games for waking dream/immersive play, too. I wouldn't characterise either of these "types" of play as monolithic, however. There are varieties and also different nuances in both "player engaged on metagame level" and "player engaged purely on gameworld level" types of play.
Which to me is a problem. I don't characterize such fundamental divides as playstyle differences. They're really different games. To me, D&D or any other rpg has to have clear roles for the participants.
Thankfully 3e does that, and as a first generation gamer, I largely follow the books on this subject. That is, I learned Rule Zero by picking up a book and reading it. I don't think anyone who has the 3e books and can read could draw any other conclusion about this paradigm, and how it informs the entire system. Do other games, including other versions of D&D, specify differently, or not at all? Yes. If someone else wants to analyze that social contract that informs other versions of D&D, they are welcome to.
To my mind, building a game that supports a different role for the players (and whatever DM/GM/etc. is posited, if any) is fine, but requires a page 1 rewrite. Since 3e was built explicitly with the assumptions of a strict player/character bond and an omnipotent and interventionist DM, it seems that when people try to play it with a different social contract it sometimes does not work well. This could lead to lengthy discussions about how "balanced" the game is, but I think that's missing the point.
As I mentioned above, I think this is because you are using a mechanically heavy system to run a waking dream/immersive experience for the players. If you are unhappy with the state of affairs (and I understand you may not be) I suggest using a lighter system or moving away from a strong emphasis on immersion to some other flavour of character identification.
Well, there is a reason why I'm running CoC right now. But it is CoC d20, and thus is in some sense part of the 3e family. 3e, as constituted, has a lot of unncessary baggage. Since most of us started with 2e, there is a nostalgia factor that keeps us engaged in things like classes and spells. Frankly, I'm beginning to look at this as a collective character flaw among my group (and probably not just my group).
To me, 5e ought to be exactly what I'm talking about, something fundamentally simpler and more generic.
The aim of "producing a good emergent story" is actually at odds with immersionism. Trying to hit both targets at once is making a rod for your own back. In a sense, story generating games are a bit like particle physics experiments; you smash some character needs together and see what interesting particles emerge from the collision. Nobody controls or dictates what will result from the collision, as such - they just set the parameters for the colliding characters, drive them up to high velocity and see what happens!
I don't see how any of this is at odds with immersion. In fact, I find that as a DM, focusing more in in-world phenomena and less on metagame agendas is very helpful to producing a naturalistic and emergent flow of events.
It does, though, bring up the necessity that results that you really DON'T want should be made impossible by the rules. In other words, rules quality(1) and balance become important.
(1): Edit to be clear: I don't mean "quality" in the sense of "good/bad", here - I mean that the structure and the detail of the rules - what qualities the rules have - matters, simply because you are not intending to depart from them for any reason.
Well, that's the box I was talking about. For example, all the talk of legislating in or out Save or Dies or other forms of quick and random high-lethality mechanics. It may be that some people really don't want these things and that there are very meaningful problems that can arise from their application, but legislating them out of the game removes reams of possibilities and completely changes the tactics. There are cases where getting what you want is not a good thing.
And indeed, that's part of getting emergent events. It's not only the DM that decides whether the character lives or dies, or even the player. It's the dice. That's pretty scary, but it is emergent.
I don't see how it can be "legislated into (your) campaign". Just strip out the bits that you dislike. Run 4E with several fighter powers disallowed. Make oozes immune to being thrown prone, undead and constructs immune to striker bonus damage (or just Sneak Attack). If the balance doesn't exist in the system to begin with, however, putting it in takes far more work than it will ever be worth.
If I stripped out the bits that I disliked, there wouldn't be a game left, honestly.
Having said this, the easiest route for either way is to pick a system better suited to what you want to do. You can even port D&D worlds and monsters into it, as I did with DragonQuest, way back when.
To me, it's been a learning curve over the years with 3.5 revisionism. If I were starting again from scratch, I might start something else. But is there another system out there that's not only better for me than what I have right now but by enough to be worth making the transition? I'm skeptical of that.
I would have hoped for 5e to be exactly that (and indeed I hoped for 4e to be that before its release), but I haven't seen my hopes fulfilled thus far.
If waking dream/immersion happens to be the way the players want to play - of course! But if they want something else then "DM totalitarianism" can be at least as dysfunctional*.
If. yes.
In a similar vein, I think that "balance" can be oddly beneficial for a simulative, immerion-focussed game. The "real" world has a certain "balance". If it was as easy to become a CEO as it is to become a janitor, there would be very many more CEOs in the world, I would wager! It would actually make sense to demand, for instance, that Wizards and Clerics have much higher attribute requirements than Fighters or Rogues. Early D&D actually made an attempt to be balanced in this way; attribute rolls were random, and only those with very good rolls could get the "powerful" classes (such as paladin or monk).
That's probably the best argument for balance that I've heard. Darwinism. People will naturally do the best things that they can do, and the ones that get the best results will keep doing it. If there was a type of magic that only ever produced cantrip-level effects, no one would learn it, because it would be a waste of time. So it follows the magic that does exist must be worth the effort and opportunity cost for at least some people.
I'm on board with that. I don't see people talking about balance much in practical terms like that though.
If you look at how 3e handles class balance, it makes perfect sense on the naturalistic level.
In the real world, everything has its own value and the "system" of reality finds its own balance where cost and benefit come together (on the whole). That a game world might represent such a "dynamic equilibrium" is to be expected, not some sort of oddity.
To me, that's happened in the D&D world we've been collectively imagining for the past few decades as well.
