All right, let's try this in a bit more depth.
Hussar said:
Flavour First: The designer comes up with an idea first and then tries to create mechanics to best fit that idea.
Game First: The designer looks at how the game is being played at the table and creates mechanics to best facilitate that. Flavour is then added afterwards to justify the mechanics.
That's all well and good, but the division works at an even more basic level than that: what are you willing to give up at the table? Are you "flavor-first," willing to give up wonky mechanics in pursuit of the plot or character development or whatever? Would you be okay with something super-powerful if it felt right? Or are you "game-first," willing to give up your idea of a character or a world or a plotline if the mechanics work against you? Would you be okay if your character couldn't exist because they're too super-powerful by the numbers?
It's not a simple design philosophy, really. It's fundamentally the same question of style that has haunted D&D and other PNPRPG's forever. It's Crunch vs. Fluff. Every game finds a different middle ground. 4e has found its middle ground as well, and certainly isn't looked at from purely one direction.
For most of the history of D&D, we've had flavour first mechanics. Going all the way back to the beginning, (I'd argue especially back at the beginning) elements were first dreamed up as fitting with the genre or the game and then bolted on using mechanics that hopefully worked at the table. Now, I would like to state that this can work. It most certainly can. Frequently we get some very good stuff out of this approach.
Actually, historically, D&D has always used a "mix." What is a wandering prostitute table if not "Game-first" thinking? The idea that there's a difference between an expensive doxy and a brazen strumpet that should be encoded into the rules is because it's fun to roll the dice and spontaneously adopt the persona at the table, not because if fullfills some archetype.
In 4e, D&D still uses "Flavor-First" design. The process of generating an NPC begins by defining what role the character will have in the story (are you going to fight them? are you going to rescue them? are you going to shop for items with them?) as the basis for what stats you need.
1. The paladin.
From a flavour standpoint, there is probably no more archetypal element you could add to a fantasy game than a holy knight out to right wrongs and lay the beat down on evil. This tunes into some very old memes. Heck, if it worked for Arthurian tales, it can work for D&D right?
But, is there a class that has caused more problems at the table than the paladin? From day 1, the paladin has caused all sorts of issues at the table. Frustration and outright hostility quite often. The flavour says that the holy knight should only work with other good characters. Right off the bat, you're telling the entire table what they can and cannot play the second you put this down. I was just told in no uncertain terms that a player most absolutely cannot tell a DM what races should be in play in the DM's campaign, but, if I play a paladin, I've just told the entire table that they can only play good characters.
Never mind all the restrictions that came with the paladin. The whole "paladin's code" is informed by a very narrow view of what constitutes a holy warrior. Why should a paladin automatically mean Teutonic knight? I cannot have a paladin from any other culture than faux-European?
This is actually a DM/Player communication problem, not a problem of design, because the problem doesn't lie within the
rules of the Paladin class, but rather in the fact that the DM can revoke the players' power if the DM thinks the player isn't "playing it right" (even if the player does). The reason for the existence of the code/alignment requirement/ally requirement in early editions was to keep Paladins from being played because they were otherwise clearly superior. By 3e, it had become purely a flavor issue, and thus purely a DM/Player communication issue, just like trying to shoehorn a dragonborn into a strictly medieval Europe-style setting is.
2. The Ranger.
Again, the flavour here is great. The edge-walking warrior who patrols the borders between civilization and the great unknown. Very archetypal, very inspirational. But, again, when the mechanics hit the table, you get huge problems.
Take the ranger's favoured enemy ability. Back in 1e and 2e, the ability was very limited - basically only giants and giant-kin. Great, did loads of damage against them. But, if you're playing in a game that doesn't feature giants, it's just taking up space on your character sheet. 3e went a little ways to helping, by broadening the ability and allowing players to choose, but, again, the player is pretty much at the mercy of the DM to provide those opponents.
The only way the ability is of use is if the DM tells the player what to take as a favoured opponent. Otherwise, it's just taking up space.
You've got a bit of something here. However, there's no reason to think that the reason Favored Enemy doesn't work is because it is flavor-based. It is an example of regular old bad design. Other games have taken the flavor idea of a hated enemy and have dealt with it admirably (drama/fate/karma/hero/action/extreme points come first to mind). D&D lost by tying it to a class, and thus consuming one of the rarest of resources as well as tying it to excess baggage.
You're missing the target there. The problem isn't that they wanted to come up with mechanics for a hated foe, the problem was the specific implementation of that mechanic. Other mechanics work just fine.
3. Prestige Classes
A 3e issue, because there were no PrC's in earlier editions. Many PrC's relied on role play, or in game elements as balancing. "To become the Yellow Blade Master, you must study with a master for six months". But, what does the rest of the group do for that six months? What if the campaign is on a bit of a time limit? Most of the time, these restrictions meant that either the PrC languished in obscurity, or they were outright ignored.
The other problem was in PrC design that narrowed the focus based on the flavour. Knight of the Chalice (at least the 3e version) is a poster boy here. A PrC that gives you massive bonuses against demons, because that's their purpose - kill demons. But, what if the DM doesn't use a lot of demons, just some? Or he uses demons sometimes, devils another time and other evil outsiders a third? This is so campaign specific that the only way it works is if the DM specifically tailors his game to cater to the PrC. It's essentially the same problem you see with the Ranger's Favoured Enemy.
The first part of this fits into "not really a problem" territory. Not every PrC needs to fit into every campaign. Some campaigns can't handle tieflings, some campaigns won't be able to handle 6 months of downtime. I guarantee 4e will have many things that you won't be able to use in it. I bet it already does. PrC's went with the same approach: there's a lot of them, we don't expect you to use them all, but we'll fill a couple of different niches with them.
The second part of this is just an exacerbation of the Favored Enemy problem mentioned above. It's an "all your eggs in one basket" problem. 25-75% of your characters' power shouldn't be limited like that. But there are successful ways to implement the idea that don't rely on those mechanics.
4. Rogues
Ohh, backstab, sneak attack, how I love thee. I love rolling all that damage. Pulling it off was a thing of beauty. From a flavour point of view, limiting backstab or sneak attack, makes sense. After all, how do you kick a jelly in the vulnerables? Stabbing a golem in the kidney doesn't do a whole lot. Makes perfect sense right?
But, then you hit the table. And the poor rogue player is sitting in the corner twiddling his thumbs because his one main combat ability is nerfed entirely by about a quarter of the creature types out there. In a tomb raider scenario, quintessentially D&D, the rogue is reduced to the role of well armed commoner by pretty much any monster most likely to inhabit a tomb.
And this makes sense?
Again, this is just the Favored Enemy problem. Bad design, okay, but the flavor has also caused good design.
I could go on, but, I'll stop here. There are many more examples and I'm sure you can think of a few of your own. And, I'm also sure I'm going to be told that the above examples are not really problems with the game, but with the players. To me, these are GAME issues. Anytime the game forces this sort of thing on the players, this is a game issue.
We've had flavour first mechanics for the past thirty years. I think that's more than long enough of a kick at the cat. It's time to let game first rule the table. Find out what works at the table, what results in the most fun (yes, I'm going to use the "f" word here) and then design to that.
You're missing your mark. This isn't baby-with-the-bathwater territory, this is baby-instead-of-the-bathwater territory. I guess the baby's no longer in dirty water, but you've hardly addressed the real problems.
This bogeyman of "flavor-first design" that you've conjured rears its head in 4e, and has in all editions of the game. Our potential savior of "game-first design" has likewise popped up all over the place.
The metric of success of a mechanic is how well it help you play the game.
The metric of success of a game (of D&D, anyway) is how well it helps you evoke the flavor of a good fantasy story.
A rule is usually poorly designed for D&D if it makes the game bad (by preventing you from telling your good fantasy story) or if it makes the play bad (by getting in the way somehow).
A paladin's ally restriction and a ranger's favored enemy only really make the play bad. Both can be solved without making the game bad. Mostly by getting new mechanics that still represent the archetypes and genre tropes that the players are going to enjoy.
Certainly for some, the changes 4e has made to make the play better have made the game worse, because it erodes the very reason some people play the game. They have more trouble telling the fantasy stories they like to tell. So while the game might be a breeze to play, why would you even play it in the first place?
It comes back to Fluff and Crunch. Some people play D&D for the former and feel it has been violated by a focus on the latter. Some don't feel it's really been violated. Some people play D&D for the latter and are pretty happy.
We'll see in a few years if things like second wind, or daily abilities, cause anywhere near the gaming anguish that paladins have.
For some, they already have, because while paladins could have been overlooked by a group that had a problem with them, everyone has second winds and daily abilities, and those can't be overlooked.
...and yeah, my first run was better...
