rgrnwood:
1) Player investment: if the player personally feels some kind of emotion about defeating the boss, then every other method of making it epic works better.
I use the term "buy-in." It's a combination of trust, engagement, excitement and other things, and while it's hugely important, it's not something that is easily brought about by a set of rules. Any reasonably well-made system seems able to engage players, so it seems less about the system and more about the group dynamic.
rgrnwood:
2) Character investment. If your character has RELATABLE reasons to care about defeating a boss, then the encounter is more epic.
That makes sense. I think this is mostly a factor of player buy-in. Fights in games are really just decorative hoops to jump through to get to the next bit (even if that's just the credits scene and a new achievement), but we are more likely to want to jump if there seems to be a reason for it inside the game.
rgrnwood:
Is it cliche? Yes. Did it work for the player? Yes.
Well, cliche certain can work. That's why it becomes cliche. Bought-in players can lap up a lot of cliche, but everyone has a breaking point where their eyes just start to roll.
rgrnwood:
3) History with the boss: This is why recurring villians can work so well.
This begins to touch on rules, because not all games handle recurring villains well. They're a common trope in stories because it's generally easy for an author to contrive things to let the enemy escape, or to have yet another henchman standing in the path on the way to the enemy. In video games, boss escapes tend to be a tad contrived, often featuring a cut-scene in which the player can't do anything, and which sort of invalidates the skill they might have brought to bear on the fight.
When it comes to RPGs, I generally find that players
hate when villains get away. RPGs are freeform enough, that it's sometimes hard for players to see why they can't or shouldn't immediately start chasing an enemy who has escaped. GMs can contrive escapes, but it's tricky and can elicit groans from players. Again, buy-in is a factor.
I know I asked what you think of as an "epic boss fight" outside of RPGs, and I'm looking at your criteria through the lens of RPGs, but I'm trying to see how to bring what makes those fights what they are into the pen-&-paper RPG.
rgrnwood:
This is turning into a wall of text so the other events I can think of are: arena design, encounter design,
While this is somewhat independent of RPG design, it's important to know the rules you're using. Some RPGs are more about close fights and others are about very open fights, for instance. Setting one kind of fight in rules meant for another kind might lead to unintended results.
rgrnwood:
bosses having abilities to change the arena, bosses having abilities that are unavailable to the players,
Okay, these get into the areas of rules, somewhat. If we're talking destructable structure, does the game have easy rules for that, or rules that can be reflavored (an attack that imposes a penalty to PC defences for a certain amount of time could be seen as eroding their easy cover, for instance)?
Some games have enemies who are built on the same principles as PCs. GMs can usually state by fiat that the enemy has some aspect of their backstory that would plausibly give them access to enough abilities or gadgets to make them a challenge for the players, but may be somewhat constrained by a desire not to burn too much player buy-in by giving the enemy too much amazing stuff. That's mostly a matter of staying within the genre of the game, though. A wizard or a star-emperor might be able to shrug off projectiles, but a corrupt Nevada sheriff probably can't.
One important aspect common to epic fights is a terrain or location that works for the enemy. It's their lair or their base, or just their kind of terrain. So, one might want to make sure the ruleset allows for cool environmental stuff, or things like traps or automated defenses. If a fight is taking place in a factory, how can the rules accommodate the dangerous and random nature of active machinery to help or hinder both sides?
rgrnwood:
removing abilities from players that they assumed they would always have,
While this is mostly a storytelling element, it's something some games consider, though it's hard to do well. D&D is well-known for monster immunities and things that counter common magical approaches, but different editions handle it differently.
Removing player options also takes a lot of buy-in from players. Do they want to go in and duke it out after fighting to gain all their neat stuff, or do they want to have that stuff stripped away and have to use their wits more?
rgrnwood:
puzzle bosses (not just bullet sponges),
Do you have some examples of these? I like a fight that requires strategy, but when I think "puzzle boss," I think of something that isn't really a "fight," just a question that needs to be solved. Those can be cool or funny in stories (or not), but in games I tend to feel like they are either trivial or impossible. In video games, even if they work well, they tend to have diminished replay value, since the fight will be much easier next time. I suppose that's less of an issue in RPGs.
I suppose that as long as a fight is raging while a solution is arrived and implemented then it's likely to be cool. When Luke hit on tripping the walkers, it was still a dicey situation and not a sure thing. And they still lost, though they never really planned to win. But I tend not to want something where the players can simply stop and think and figure it out and win.
rgrnwood:
perfect difficult level (push players to their limits but not being their limits)...
A perennial issue in game design, right? This isn't entirely independent of the rules, but probably has a lot to do with the mastery of rules and of the tricks of game mastering, so that the GM can line things up well and then adjust them on the fly.
I thought of a few other things I like to see in epic fights:
Change. The characters or the world need to change before, during or after a fight. Rules can facilitate that, in terms of characters at least. If a character has an epiphany and suddenly an ability or set of shackles is unlocked (or locked) and this doesn't require a complicated reworking of a character sheet, then that could be pretty epic.
Real risk/cost. Maybe this is related to "perfect difficulty level," and my dislike of "puzzle bosses" that can just be "solved," as well as to the need for "change," but I think an ideal boss fight has to be one that
requires an important sacrifice. If everyone gets away unscathed, then it wasn't epic. I'm not sure how rules can contrive to make that happen though.
rgrnwood:
Oh, and D&D is designed for boss fights. No one-hit kills like they can get in Savage Worlds, plenty of mechanics to toy with, plenty of locales and enemies to choose from, players aren't too frail to make a couple mistakes or have a few bad rolls during a boss fight, etc.
I would tend to agree, though there's something about D&D that I think also works against epic boss fights. I'm not entirely sure what it is. Arguably, it's the degree to which it's handled as something to which optimal strategy should be applied, and the general idea of "fairness" that is generally assumed. That is, players tend to expect and be expected to win, and the game (depending on the edition) supplies enough options for players to lock down or "solve" what could otherwise be epic fights. I know that it's not unusual for people to feel like fights in D&D are just dice-rolling slogs, expecially when the fights go long, as we tend to expect boss fights to do.
All in all, I think what is coming from this discussion is that boss fights can be aided by the rules, but not easily brought about by them. I tend to wrinkle my nose at the idea of things coming about "organically," but (outside of a situation planned out with the players, which I'm not opposed to), I think the epic boss fight is something that is going to happen without obvious intention. There's just going to be a combination of buy-in, backstory, challenge, stakes and disadvantages that, when it's all over, will be recognized as having been "epic."