D&D General Kicking the tires vs. puncturing the tires; being effective vs. breaking the game

[MichaelSomething]

That also depends if you're a "scrub" or a "try hard".

 

[EzekielRaiden]

Well, its kind of almost a tautology, that the less things you control about a character, the harder to cook the books on them it is. But the flip side of that is that it decides to block imbalance by forcing you into the game (and random rolls) idea of what your character is like.

Seems too high a price to pay to me.

This is a false dichotomy smuggling in an unstated premise: "The only way for a game to be interesting is for it to be breakable." That's why @Ruin Explorer brought up very popular games like chess; if having too little control guaranteed an uninteresting experience, people wouldn't play these games.

Instead, it seems your issue is excessive focus on the basic/easy way to make a game "unbreakable": make it trivial. But that means presuming that the correlation works both ways, that the only way to make a game "unbreakable" is to make it trivial. That's simply not true, as games like chess and go attest. The problem with tic-tac-toe isn't that it is unbreakable (though it is); it is not that it has few rules and few moves (though it does); it is the plain and simple fact that it is boring. Minimal skilled play from both players makes every game end the same way, a draw.

Of course making a non-trivial "unbreakable" game is much more difficult than making a trivial one. Even go, for all its beautiful simplicity, still has one small opening for issues, which is where the ko rule comes into play (some variation of "you cannot make a move which would return the board to a previous state.") There are also different rules regarding how points should be awarded, with some groups giving handicap points to white since it goes second, usually either 6.5 or 0.5 points. (The latter simply means that, in the event of an exact tie, white always wins; since this is difficult to achieve in go, this is a relatively small bonus.)

For something as sprawling, complicated, and open-ended as a TTRPG, the difficulty of creating a non-trivial "unbreakable" is yet higher. But "difficult" is not "impossible." It may be impractical, but that's a far lower bar. Hence why, above, I spoke of allowing a slightly looser range for "unbreakable," e.g. mostly or nearly unbreakable. With that extra leeway, quite a bit becomes entirely viable and the corner cases can be relatively isolated and dealt with as exceptions, rather than the constant refrain.

 

[Thomas Shey]

This is a false dichotomy smuggling in an unstated premise: "The only way for a game to be interesting is for it to be breakable." That's why @Ruin Explorer brought up very popular games like chess; if having too little control guaranteed an uninteresting experience, people wouldn't play these games.

Well, this also makes the assumption that something like chess's approach would be interesting in an RPG context. That's not an assumption I agree with. Especially since I'm talking about being interesting from a character definition approach. I'm pretty firm in my opinion that the more useful ability you have to define character abilities, the more potential areas you have to create problematic results.

And yes, I'm unconvinced that, at least by my standards, its possible.

 

[EzekielRaiden]

Well, this also makes the assumption that something like chess's approach would be interesting in an RPG context. That's not an assumption I agree with. Especially since I'm talking about being interesting from a character definition approach. I'm pretty firm in my opinion that the more useful ability you have to define character abilities, the more potential areas you have to create problematic results.

And yes, I'm unconvinced that, at least by my standards, its possible.

I don't think it does make that assumption. All it says is that we cannot presume a 1:1 connection of the kind asserted, that reliably games that are more "unbreakable" are also less interesting. That doesn't mean there's no connection at all. But it does mean that we can't leap from "this is more unbreakable" to "it absolutely must be less interesting."

And yes, certainly, the more potential areas you have to create problematic results. Keyword: potential. But potential need not become actual. Which was my whole point. It is quite difficult to design a good, intricate/textured, open-ended, balanced game. That's a lot of requirements, and all of them bring their own difficulties. But something being difficult doesn't mean that it's impossible. There are plenty of things that are exceedingly difficult but which get done every day--or things that were once "impossible," but which we now know to be possible, you just require the right tools, training, etc. (Consider AEDs, which save the lives of a couple thousand Americans every year--people who, even as little as sixty years ago, would have almost certainly been considered impossible to save.)

I can't speak to your standards, of course. But to blanket reject the possibility seems premature. The Mertens conjecture is a good reason to be cautious about heuristic reasoning like this. Large amounts of statistical, computational evidence suggested that the conjecture was true (that the absolute value of a certain very slowly-growing function never grows beyond sqrt(x) for any positive integer x). Then someone proved that a counterexample occurs somewhere in absolutely enormous numbers, we're talking numbers with many more digits than there are particles in the universe....and later, someone else proved that there are infinitely many values that are counterexamples.

Or, if you wish to be more pithy:
There are more things in heaven and earth, Horatio,
Than are dreamt of in your philosophy.

 

[Thomas Shey]

I can't speak to your standards, of course. But to blanket reject the possibility seems premature.

I'll be pretty blunt; when my opinion has been formed by more than 40 years in the hobby, not limited to one system or family of systems (albeit leaning into trad games pretty heavily), its pretty hard for me to consider my view "premature". It may be incomplete (because it hard for that not to be not knowing every game that does or can exist) but that applies to any conclusion one can ever come to. So I'm pretty comfortable saying I don't think the one I've reached is premature in any meaningful sense.

(Though you do reference a legitimate point; there may well be people who's standards of what meaningful character differentiation and choices therein does not produce the result I refer to. But based on my own standards, I stand by my position).

 

[Pedantic]

I'll be pretty blunt; when my opinion has been formed by more than 40 years in the hobby, not limited to one system or family of systems (albeit leaning into trad games pretty heavily), its pretty hard for me to consider my view "premature". It may be incomplete (because it hard for that not to be not knowing every game that does or can exist) but that applies to any conclusion one can ever come to. So I'm pretty comfortable saying I don't think the one I've reached is premature in any meaningful sense.

(Though you do reference a legitimate point; there may well be people who's standards of what meaningful character differentiation and choices therein does not produce the result I refer to. But based on my own standards, I stand by my position).

My hot take, based on about half that experience: TTRPGs are often badly designed by the standards of other kinds of games, and the culture of play surrounding them has made this normative. When they break, it tends not to be because of the oft-discussed "absurd combinatorial" but because of simple iterated probability. Failure states are generally obvious, and don't require extensive play time at the table, so much as a solid math review.

The kinds of errors that crop up in TTRPGs simply wouldn't make it into a published board game, and we're very quick to jump to "this is a unique medium" to explain away why that happens. I'm certain there is a limit of design perfection that's possible, but I'm equally convinced we are not even close and not really trying. I want hours of logged same game tests, I want to hear about iterations across 5 play groups monitored for monthly campaign play, I want game designers coding simulations of monster fights and likely town interactions and glumly staring at spreadsheets, and, given the existing culture of play that so readily encourages tables to modify the rules, I want the second half of the DMG to be design notes on the impact of various likely to be considered house rules.

 

[overgeeked]

My hot take, based on about half that experience: TTRPGs are often badly designed by the standards of other kinds of games, and the culture of play surrounding them has made this normative. When they break, it tends not to be because of the oft-discussed "absurd combinatorial" but because of simple iterated probability. Failure states are generally obvious, and don't require extensive play time at the table, so much as a solid math review.

The kinds of errors that crop up in TTRPGs simply wouldn't make it into a published board game, and we're very quick to jump to "this is a unique medium" to explain away why that happens. I'm certain there is a limit of design perfection that's possible, but I'm equally convinced we are not even close and not really trying. I want hours of logged same game tests, I want to hear about iterations across 5 play groups monitored for monthly campaign play, I want game designers coding simulations of monster fights and likely town interactions and glumly staring at spreadsheets, and, given the existing culture of play that so readily encourages tables to modify the rules, I want the second half of the DMG to be design notes on the impact of various likely to be considered house rules.

That would be amazing. If only.

 

[Thomas Shey]

My hot take, based on about half that experience: TTRPGs are often badly designed by the standards of other kinds of games, and the culture of play surrounding them has made this normative. When they break, it tends not to be because of the oft-discussed "absurd combinatorial" but because of simple iterated probability. Failure states are generally obvious, and don't require extensive play time at the table, so much as a solid math review.

The kinds of errors that crop up in TTRPGs simply wouldn't make it into a published board game, and we're very quick to jump to "this is a unique medium" to explain away why that happens. I'm certain there is a limit of design perfection that's possible, but I'm equally convinced we are not even close and not really trying. I want hours of logged same game tests, I want to hear about iterations across 5 play groups monitored for monthly campaign play, I want game designers coding simulations of monster fights and likely town interactions and glumly staring at spreadsheets, and, given the existing culture of play that so readily encourages tables to modify the rules, I want the second half of the DMG to be design notes on the impact of various likely to be considered house rules.

Well, I think that's true but a bit separate from what I'm talking about. Let me expand a bit.

While I've done a number of other genres over the years (fantasy in particular, but also post-apocalypse, space oriented games, and others) a rather good part of my GMing career was running supers games.

Supers games, unless they're narrow in scope or written in a sufficiently hardcore narrative focus that the distinctions are mostly cosmetic, virtually demand about as much customization as is at all possible. They also tend to demand the ability to, in one fashion or another, provide the ability to do things that would be considered well beyond the pale of balance in almost any other genre.

And one of the things that they reveal is that as the number of possible combinations increases, the less and less possible it becomes to foresee how some of those combinations will work (and this doesn't necessarily just refer to specifically the character bits themselves, but structural elements which can work okay with most of the system but reveal their risks when applied in particular places.

And yet, you don't really want to remove all those elements, as they're not only genre-appropriate, they often aren't hazardous to the game when used with most other elements. So you either end up with (potentially a large number) of special cases (if the design has been properly playtested either up front or from being in the field enough years) or a non-trivial amounts of houseruling and/or ad hoc decisions.

This is a separate problem from math that's just bad from the get-go. Naturally, supers is probably the far end of the spectrum when it comes to this (as very few other genres will demand the same degree of customization), but that just says its a matter of degree; the more moving parts a system has in terms of how characters constructed, and the more freely those construction bits can be connected, the harder it is to upfront balance everything in practice.

Its relatively easy to balance a simple modern action game, because there likely are only so many significantly interactive pieces in play. You may get failure states of some attributes being too worthwhile compared to others or excessive breakpointing,, but that should be visible, and if you want to, something you can address up front (this is why you hear relatively little about people "breaking" BRP based games; since about all most of them have is attributes and skills, either they break on misdesign in magic systems, or because of the aforementioned overweighting of a couple attributes and the like). The more special talents, paranormal abilities and other devil-in-the-details there is, the less true that is.

 

[Pedantic]

This is a separate problem from math that's just bad from the get-go. Naturally, supers is probably the far end of the spectrum when it comes to this (as very few other genres will demand the same degree of customization), but that just says its a matter of degree; the more moving parts a system has in terms of how characters constructed, and the more freely those construction bits can be connected, the harder it is to upfront balance everything in practice.

Its relatively easy to balance a simple modern action game, because there likely are only so many significantly interactive pieces in play. You may get failure states of some attributes being too worthwhile compared to others or excessive breakpointing,, but that should be visible, and if you want to, something you can address up front (this is why you hear relatively little about people "breaking" BRP based games; since about all most of them have is attributes and skills, either they break on misdesign in magic systems, or because of the aforementioned overweighting of a couple attributes and the like). The more special talents, paranormal abilities and other devil-in-the-details there is, the less true that is.

My rant on this topic has become more unhinged as modern design has tended toward increased simplicity but not actually gotten any better, but this should be the selling point for class-based systems, and even in the case of wide open point buy, I think we've still got a lot of space for more effort to pay off better dividends.

On the other hand, I do think games need to do a way better job of preparing GMs for their tolerances. How much of this tension in D&D comes down to not adequately telling DMs what kind of gameplay/encounter/adventure design is appropriate to dealing with the abilities PCs can marshal at various levels? Most of that discussion has focused on math, which is absurd; it doesn't matter how accuracy/defense scale when the players are deploying magic and planning to prevent encounters from happening at all. If that's the gameplay loop, then the GM should know that's the gameplay loop.

I worry that in addition to a lack of design rigor, you have GMs holding an unreasonably low baseline for what player power should be, and ending up surprised and frustrating the game has given the PCs the ability to do things. Ideally, that should neither be surprising, nor unwanted.

 

[Lanefan]

I want hours of logged same game tests, I want to hear about iterations across 5 play groups monitored for monthly campaign play, I want game designers coding simulations of monster fights and likely town interactions and glumly staring at spreadsheets,

To what end result, though?

Is it your goal that the system be designed such that those five same-game tests produce very close to - or exactly - the same results on each play-through?

Is it your goal that those endless monster fight simulations and data-filled spreadsheets lead to a system where said fights are 100% predictable, as in "here's the precise bell-curve of what'll happen when characters a-b-c-d take on monsters x-y-z"?

'Cause if so, I'll get off the bandwagon right now. I don't want combat to be predictable, nor adventures to come out the same every time they're played.

and, given the existing culture of play that so readily encourages tables to modify the rules, I want the second half of the DMG to be design notes on the impact of various likely to be considered house rules.

Now here we agree, though I'd go a step further and suggest they actually propose some of what they expect to be common house rules as options, and discuss the design knock-ons on that basis.