The starting point ever since about 1974 has been to build on-with-around that which came before.
Er...no, not really. 3e did away with essentially everything mechanical that preceded it. 4e did away with nearly everything mechanical that preceded it. 5e sutured together stuff that came before it, but very much with a completely different spin.
The mechanics have not been one contiguous thing for ages. At absolute most, you could argue that the most barebones mechanic of all, the d20, is the one and only thing that's lingered...and even then, it only really took that level of prominence with 3e. Before 3e, plenty of very, very important things were resolved by d6 or d100 or a variety of other methods, meaning resolution mechanics were all over the place.
How so? From my viewpoint 4e and 5e share a lot more similarities than differences; in design philosophy, intent, and intended playstyle if maybe not in precise mechanics.
From my viewpoint, they only share a very few points of similarity, and 5e actively worked to destroy as much of 4e's legacy within it as the designers could humanly get away with. I mean that very, very sincerely. 5e is
ashamed of its 4e roots, covers them up whenever it can, and works as hard as possible to really be "3e mark IV, we'll get it right this time, we promise."
For many, being able to admit when trial and error has come up "error" is the hard part.
Uh...
If that's the "hard" part, then what on earth is the appropriate adjective for describing how difficult it is to create a structure that is actually mathematically sound, while also actually interesting to use,
and flexible enough to account for a reasonable diversity of choices?
Because
that part sure as hell ain't
easier than admitting you didn't do it right.
Designing things at all is hard. Once you've got over that hump, "well" and "poorly" take IME about the same amount of effort.
I'm sorry, I just...I can't engage in any constructive way with this belief. "Painting at all is hard, once you get over that, it's as easy to paint like Leonardo as it is to paint like a five year old." No. Just...no. That is simply not true, and never will be.
I'll caveat that, however, by saying "well" does not by any means equal "perfect", 'cause we ain't gonna get that on any scale larger than a single DM designing for a single consistent and known-well group.
Perfection is never the goal. Achieving the design goal within a reasonable statistical range is the goal. You set what you consider the range of reasonable success (e.g., "the damage output of a damage-focused Xth level monster should be within +/-10% of average output of (X-1)th to (X+1th) monsters" or the like), and then test and iterate until you achieve this. It's a long, slow, and often boring process as a designer, and requires extensive testing. That's one major part of why it's so hard for individual DMs to replicate. Getting enough testing data as a singular DM is extremely difficult. Designers can tap dozens, hundreds, potentially even
thousands of groups to get much, much more data, which makes the mathematical structure
far easier to test.
And if, after (say) a couple months of doing this, you find that you genuinely can't get within that range, you go back to the drawing board to find out why your reasonable design goals aren't reachable. Perhaps you were actually unreasonable and didn't know it. Perhaps you missed something that's messing things up. Perhaps you failed to account for the actual degree of variability, and +/-20% should've been used. There are myriad ways why your design might fail to meet the goal you set for it.
Agreed to the latter: everything being identical isn't the goal. That said, balance isn't the holy grail of game design and treating it as if it is means all the other factors - playability, long-term engagement, short-term appeal, relative simplicity or complexity, fun, etc. - get given short shrift.
....balance is what
enables all of those things, other than maybe short-term appeal.
An unbalanced thing becomes less playable, because it has degenerate solutions and unexpected failure points (such as the "ghoul surprise" that caught 5e's own designers off-guard
during a demonstration that was supposed to show off how fun it was to play 5e.) An unbalanced system reduces long-term engagement, either because it can be solved, or because it defeats skillful play too often with random bull$#!t. Balance is how you
make things that are simple but still provably effective, and other things that are complex but still worthwhile.
"Fun" is a useless standard. A game that genuinely cannot, even in principle, be enjoyable to play doesn't exist. If it did, it would be a powerful and extremely dangerous psychological torture tool. Now, that doesn't mean you should not be checking that the goals you've set are in fact enjoyable for the people you're wanting to sell the game to. That's very important! But getting those goals to be actually fun--to have something where prediction is possible but never certain, where there are many different valid strategies but no clear winning (aka degenerate) strategies, where doing the thing the rules indicate is a worthy experience people will seek out--then
pursuing balance IS trying to make that happen.
Unbalanced systems have short-term appeal for the same reason puzzles have short-term appeal, because they are effectively identical: "Find the solution." Once you find the solution, the puzzle is done. There is no more enjoyment to be had. Maybe you hang the completed puzzle in a frame or something to look at later, but that's really not
solving the puzzle anymore.
Balance is just one factor among many, and there's many different ways of both defining and achieving it.
But balance is either one of the best, or is the only, way to achieve many of those things. You are correct, however, that definition and method both admit a lot of variety! That's another huge part of why game design is so difficult. Being able to wisely pick design goals, to set meaningful standards for whether they've been met, and correctly interpret what the data is telling you when you get it--all of these things are
hard. You are an engineer.
The problem IMO isn't that martials are too weak, it's that casters have been made too strong by the slow steady removal of all the restrictions they used to have in the BX-1e era.
I mean, I don't completely disagree with you there. BX/1e still had stupidly powerful max-level casters; 5e didn't invent
wish, after all.
The bigger problem is that a lot of those limitations, restrictions, and complications that early-edition casters had weren't
enjoyable difficulty. Difficulty, like balance, comes in a lot of forms--and, unfortunately, just like balance, most of the easy forms of difficulty are crappy. TVtropes has a term for this, "Fake Difficulty," and the page has some lovely examples of the concept. First listed one applies to platforming games. Having a difficult jump that requires precise timing and coordination is perfectly fine. Having a jump that is merely difficult because the camera switches orientations halfway through, so you have to suddenly shift which direction you're "pointing" or else instantly fail? That's fake difficulty.
This doesn't mean that these things couldn't have been reworked into legitimate challenge rather than BS. They probably could have. But the designers (again...) took the easy route and just axed them. They then did not
Playing a caster, particularly in combat, IMO should be a mix of great satisfaction when your spells work and immense frustration when they don't; with "don't" happening a fair percentage of the time unless the caster is very cautious, and with unpredictable consequences following. In other words: high risk, high reward. 1e got this right other than not having potentially unpredictable consequences.
Unfortunately, this sort of thing just isn't enjoyable for most players. Remember your standard of "fun" earlier? Your goal is not a badly-chosen one....except in the sense that it just
isn't fun for most of the people who play. So where does that leave you? Are you going to do the thing you said was unacceptable, and limit your audience not just a little bit but very sharply? Or will you return to the drawing board and find a different approach that would be fun for more players?
Note that I am
not saying there shouldn't be challenge. There has to be, that's part of what makes fun. But challenge is like bittering a beer. If you hop it too much, it drowns out the maltiness, the residual sugars, the aromatics from the malt roasting process, and the subtle flavor compounds produced by the yeast. Yet if you don't hop
enough, the beer will taste cloyingly sweet. Just the right amount of bitterness is required...or, one might say, a
balance of bitterness and sweetness.
Balance is what allows the whole to be greater than the sum of its parts, whether it be in brewing or in game design.
What if I don't care how they fall as long as they all end up in the box instead of one or two ending up on the floor? I ask because in this analogy I suspect their all being in the box would be good enough for me, with further fine-tuning not required.
I mean, you can care or not care as you like. I am purely and exclusively using this as a physical demonstration of the general principle: balance is rare, it does not happen unless it is created, and other than obviously trivial cases, it is
always--I repeat, ALWAYS--harder to create balance than it is to create unbalance. ("Extremely trivial" being things like "there is literally only one option". Trivial in the mathematical sense.) Unbalanced arrangements of objects have infinitely more infinite variety. Just throw the items in however you like, and you are essentially guaranteed to have an unbalanced setup. Conversely, starting from an unbalanced setup, there may not even
be a way to balance it at all! Even if there is, it may be extremely un-obvious, even in this physical metaphor. Move it into the space of abstractions and ideas, and the difficulty increases tenfold.
No.
Do I think that when converted to game design, even trying to achieve that sort of spatial symmetry is overkill? Yes.
No.
Must all, or even most, chaos and disorder be designed out of an RPG? Again no.
Okay so you basically did the same thing for a second and third time here, and it's really, really annoying. I would really appreciate it if you stopped.
That is, you are turning something which I have explicitly called, and which very obviously was, an
example and metaphor for a generic principle, and distorted it into a COMPLETELY unrelated claim, as though I had made those claims myself. Please stop doing this. I gave the clear, specific questions I did because they were examples of the general principle: It is
always more difficult to set up a balanced structure than it is to set up an unbalanced one, outside of the cases you and I both agree are (a) trivial, (b) pointless, and (c) bad.
An unbalanced system is one where
chaos rules. A balanced system that includes probabilistic elements, such as dice, is one that necessarily has uncertainty. I've never once said otherwise. Putting perfectionistic goals in my mouth is bad argumentation; it's you creating a strawman out of my position and then pretending you've beaten it.
So I will focus on your answers: You agree that spatial symmetry (one example of balance) is difficult to create and easy to disrupt. You agree that a random assemblage is almost never an orderly structure, and that in order to
get an orderly structure, we must exert effort to create it.
The exact same principle applies to rules. In order to get rules that WORK, that do the job for which they were designed, you have to put in work! A lot of frequently very hard and boring work.
I agree that balance that offers no choice at all is pointless. I also agree that balance that offers merely an illusory choice is pointless.
Good. I've never said otherwise and have repeatedly and explicitly said both of these things (albeit in different words, usually too many) on many occasions.
But balance can and does take many forms. In post elsewhere I've outlined some of them: short-term balance, long-term balance, party-vs-opposition balance, character-vs-character balance, spotlight balance*, wealth balance, etc.
I mostly reject spotlight balance as an awful and pernicious thing, so I won't directly address that here. Other than that, I certainly agree that there are many different elements in play, which make the process of design more difficult! That,
again, is why I say game design is
difficult. It is not some trivial task, unless you are okay with doing it in a pointlessly bad way--as you have seemingly agreed above.
Nailing down any one of these is almost certain to mess with some others; the trick is to not nail any of them down but instead get each of them vaguely in the ballpark and then stop trying.
No. The trick is to
do the work to make them harmonize. It is not to
give up, which is what you're explicitly advocating: "stop trying." No. Don't stop trying. Keep at it. Keep testing and iterating and, if necessary, re-evaluating. You will either learn that your goals weren't achievable (whether in general or just practically), or figure out how to achieve them.
* - by far the least important from a design perspective; spotlight balance will always sort itself out at the table no matter what the game says, and my preference is that the players are fighting for that spotlight like hungry dogs fighting for a piece of meat.
As noted above, I find the very concept of "spotlight balance" pernicious and dangerous. Making players fight over it is one of the best ways to create open hostility in groups and to
create the very "casters and caddies" relationship you claim shouldn't be a thing.
Because, in the end, what "spotlight balance" does
isn't getting everyone an equal share of the spotlight. Instead, it teaches players that the game they should
actually be playing is "how to manipulate where the spotlight points, so it's always on me."
And that, that
precise thing, is why casters are and remain a problem, even in 5e. Because they have
both the most powerful tools,
and the greatest ability to keep the spotlight pointing only at them.
Plus? Controlling where the spotlight points is metagaming. I don't recall if you personally dislike metagaming, but I know a lot of folks around here hate it rather passionately. Folks who dislike metagaming may want to re-evaluate whether they like spotlight balance.