I agree.
Why the hate for complexity?
I agree.
GrahamWills
Hero
So, as a statistician, a lot of my job is to handle complexity. One of the measures used for complexity is the number of parameters that affect the outcome. So if I am trying to predict an outcome Y using an input X using regression, then a quadratic model is one parameter more complex than a linear model. A cubic model is one step more complex again.
It's not immediate how to map that to a roleplaying game, but I think (as has been proposed up-thread) that it makes sense to take resolution of actions as the primary activity in-game and so try and evaluate that complexity. There is a separate out-of-play complexity around character building and the like that would need a similar analysis.
So, for me, when I am playing a one-shot with pre-defined characters, all I care about is the complexity of playing the game, not the complexity of prepping for the game. And since I usually spend more time playing than prepping, that's really what I'll focus on even for long-running campaigns.
With the mathematical definition of complexity in mind, our base criterion is: How many parameters (powers, stats, number, existence of feats, etc.) do I need to consider when resolving an action?
For the theoretical "coin-toss" game where every result is 50-50, the answer is zero. This game could be said to have zero complexity. For every actual game the answer is different based on what action is being done. So I think we need to at least have a rough idea of the distribution of types of action to say what the "average complexity" of an action resolution is.
Mathematically, we might weight the complexity of each action by how often it occurs and use that as an overall measure. Although every game and table is different, I think it's reasonable to make very rough estimates and make it work. In many genres we could probably get by by saying that 2-3 exemplars are good enough to judge a system of. Perhaps:
So a very rough idea of complexity might be to estimate the number of factors that you need to consider for each of the above, add them all up, and divide by three.
It's not immediate how to map that to a roleplaying game, but I think (as has been proposed up-thread) that it makes sense to take resolution of actions as the primary activity in-game and so try and evaluate that complexity. There is a separate out-of-play complexity around character building and the like that would need a similar analysis.
So, for me, when I am playing a one-shot with pre-defined characters, all I care about is the complexity of playing the game, not the complexity of prepping for the game. And since I usually spend more time playing than prepping, that's really what I'll focus on even for long-running campaigns.
With the mathematical definition of complexity in mind, our base criterion is: How many parameters (powers, stats, number, existence of feats, etc.) do I need to consider when resolving an action?
For the theoretical "coin-toss" game where every result is 50-50, the answer is zero. This game could be said to have zero complexity. For every actual game the answer is different based on what action is being done. So I think we need to at least have a rough idea of the distribution of types of action to say what the "average complexity" of an action resolution is.
Mathematically, we might weight the complexity of each action by how often it occurs and use that as an overall measure. Although every game and table is different, I think it's reasonable to make very rough estimates and make it work. In many genres we could probably get by by saying that 2-3 exemplars are good enough to judge a system of. Perhaps:
- Attacking in enemy with the most common form of weapon in an attempt to kill them
- Sneaking past an opponent
- Persuading someone to do something
So a very rough idea of complexity might be to estimate the number of factors that you need to consider for each of the above, add them all up, and divide by three.
Classic Traveller:
* Attacking requires knowing an attacker stat number, range, target armour, and target stats (to apply damage) - that's 4 parameters;
* Sneaking isn't a precisely defined action, but resolving surprise (which is a precursor to evasion) requires knowing the number in each group, the skills in each group, and the equipment/vehicle of each group - that's either 3 or 6 parameters (I'm not sure if I'm meant to count each twice because I have to go to do bits of info, the PCs and the opponents);
* Persuading someone requires knowing character skills and world population - that's 2 parameters.
* Sneaking isn't a precisely defined action, but resolving surprise (which is a precursor to evasion) requires knowing the number in each group, the skills in each group, and the equipment/vehicle of each group - that's either 3 or 6 parameters (I'm not sure if I'm meant to count each twice because I have to go to do bits of info, the PCs and the opponents);
* Persuading someone requires knowing character skills and world population - that's 2 parameters.
So that's complexity either 3 or 4, depending on how I was meant to measure the evasion case.
Celebrim
Legend
My usual example after having discussed the limitations of The World's Simplest RPG, is to introduce a slightly more complex game. It's rules are these.
1) If the proposition is trivially easy, it always succeeds.
2) If the proposition is impossible, it always fails.
3) If the proposition is doubtful, then flip a coin. On heads it succeeds, and on tails it fails.
Thus, we are able to distinguish between, "I step across the puddle." and "I leap across the Atlantic Ocean" as propositions.
One thing to note about this slightly more advanced game is that I'm actually explicitly stating something that isn't always stated in the rules of an RPG, but which pretty much all of them have in common - namely, that by fiat we can decide when some proposition fail and some succeed. Your very rough idea of complexity runs into the problem that the parameters you need to consider when resolving an action include those that aren't stated explicitly by the proposition->fortune->resolution rules, which is in this RPG only Rule #3.
Is the factor "doubtful" actually complexity 1?
We typically don't think of metarules like Rule #1 and Rule #2 as being part of the rules or part of the proposition->resolution cycle, because they skip over the fortune mechanic. Still, a huge number of considerations often go into deciding whether to make a fortune check at all, and these metarules by virtue of not usually being formalized will differ between tables. A good example would be to query people who play variants of D&D 3.X as to what the minimum DC that they commonly will make a skill test at. My impression is that the vast majority of tables do not test (regularly) actions that have a DC less than 10, whereas I suspect you'll find some tables that commonly make tests as low as DC 5 or DC 0. The first set of tables are probably resolving the DC's under 10 by fiat in some fashion.
And further, back on me harping about how none of my examples have actually defined what an RPG is or how to play one, the above rules set does not specify who gets to decide when a proposition is easy or impossible. But it's a reasonable assumption that most people reading the above rules assumed the presence of a GM in the form of a referee, something which The World's Simplest RPG does not technically need, and which this game does not specify but which is common to most RPGs. However, for the purposes of complexity, the additional step of needing to vote on whether or not a proposition is easy or impossible would in practice make the game feel more complex even if the number of factors in the fortune mechanic (heads or tails, no modifiers) were the same.
GMMichael
Guide of Modos
Yes, if the GM simply says "no" after a tails, and "yes" after a heads.
Allow me to clean this up a little bit, for the argument's purpose. The question(s) should be:
How many parameters (powers, stats, number, existence of feats, etc.) do I need to consider to gain my desired outcome?
or
How many parameters (powers, stats, number, existence of feats, etc.) do I need to consider to cause damage?
I'm assuming a player's perspective on this, since players consider powers, stats, feats, etc. (GMs consider story.) Since "resolve an action" isn't something that players typically do, unless the game involves some sharing of GM duties or is GM-free, you can consider parameters based on what the GM could say (former question), or what the rules could say (latter question). I hope it's fairly obvious that, once you include GM persuasion in the question, it's pointless to discuss how complex the former question is.
So we're left with: how do I cause damage (or similar rules-only question). Now you can look at the complexity of the action, because your typical rules will explicitly state "if you do X, the target takes Y damage."...and Celebrim can't step in and complicate things with questions of "easy" and "impossible"...
Probably, yeah.
Yes, but it follows that, since these factors are in existence for most games, they are a "background" of complexity that will be roughly even across games. These decisions probably aren't more complicated to make in D&D than they are in, say, Call of Cthulhu. There will occasionally be a game (like, say, Nobilis) where the metagame rules take a bit more thought (and are so more complicated).
Yep. For me, again, the complexity is relevant for two reasons - 1) the time it takes to resolve events, and 2) the mental gymnastics required to get through resolution. As these go up, the number of players happy with the system will drop.
Resolving the meta-rules questions is typically done lightning fast, and typically requires little mental gymnastics on the part of the player, right? So, they don't effectively add to the issues that complexity raises in play, and should be ignored for these purposes.
Unless you are spending several minutes resolving whether or not the GM should just fiat the result, that is. Then they should be totally part of the discussion.
Celebrim
Legend
I think you are missing my point. In the above system, how complex are the factors that go into decided whether this is case #1, #2, or #3? How many parameters go into deciding whether an action is easy, impossible, or doubtful? Like say, the character proposes something like, "I attempt to kick open the door." or "I attempt to bash open the door with my fists." Depends on the qualities of the door and the qualities of the fists.
Back in the old days, these often became table arguments over "realism", as in, if this proposition could be made in "the real world" what would happen? And that happened precisely because so much of the system was built on basically DM fiat, that in practice worked a lot like the hypothetical "simple" RPG above.
What I'm trying to get at is that the shortness of the rules isn't always a good judge of their complexity and that fiat is actually a very high complexity feature of a rules set.
Celebrim
Legend
That's precisely what I'm getting at.
As a GM, whenever the rules for resolving some doubtful proposition don't exist, that for me creates a massive complexity issue. A complex procedure is to me less complex than no procedure. When a proposition is made that no existing procedure covers, then for me it's like having an exception thrown in the code that has to be handled gracefully. I have to stop and decide how it should be handled. Should I just let it work (say, "Yes")? Should I just let it fail (say, "No")? Should I just flip a coin (In D20 this is usually roll a D20 and add a some bonus, on a 15+ you did it)? Or, do we need some more complex procedure in order to be fair about things? What are the long term implications for the game of setting this resolution methodology as precedent? Am I risking creating an absurdity? Am I risking unbalancing the game? Am I risking stealing spotlight from a player? If I extend the methodology to the NPCs, will it rob the players of agency? Is this going to derail the fun or enhance it? There are a huge amount of things that normally need to be considered.
It's crazy the sort of questions that come up in games once players start doing things like (for example) deforming the terrain. Propositions come up like, "I'm going to spend an hour chopping on this wall with my magic axe to see how much progress I make." which not only involve questions of endurance and durability but sound production, or questions like, "I'm going to cast 'rock to mud' on the stone outcropping, does this cause the whole outcropping to collapse?" which involves figuring out just how the spell works, how finely the caster can control its shape, what portion of the rock face becomes mud, what happens to that mud, and figuring out whether the resulting damage is enough that the rock can no longer support its own weight. Or you get into questions in a game like Call of Cthulhu were someone says, "My character is to going to run to the nearest town, how long does it take me to do this?", which involve questions of what is a realistic pace for a person to cover over a long distance and what level of athleticism does the character's stats translate to and to what extent the characters current physical condition should be taken into account. And if you don't get this at least close to right, or if your answers don't match what the player is expecting, then you have to be able to justify your answers.
I suppose you could say I'm over complicating questions like these, in that I could as a GM answer the question with just, "What do I want to happen?", but I'm not the sort of person or the sort of GM that usually knows what I want to have happen, and that question for me is no more trivial than the rest of them (and probably, since it is me, equivalent to the rest of them).
These sort of "game stoppers" are annoying - like when the PC's ask for the name of an NPC you've not bothered to name or otherwise ask for details you've not bothered to construct. Winging it for me is a high complexity situation - much higher complexity than consulting a table or looking up a rule. Even things like, "Rule of Cool" are fairly high complexity, because now you have this subjective thing where the player might think it is cool, but other players at the table might not, and you might be ambivalent, and you are trying to decide, "Hmmm... is that cool.". Really what is going on is a sort of table negotiation, and the more players you have and the more varied their tastes the more complex that negotiation actually is. Sure, it might be obvious some of the time that everyone thinks its cool and you are all, "I'm happy to say "yes" to that." but in practice that rule gets really complex. Then like the second time it happens, is it cooler or less cool, and ect.
What I find is that the less "game stoppers" come up, the less complex the rules seem to me. I'd much rather have a game system where when unusual propositions happen, a subsystem exists to handle them, so that at most what you are asked to do is flip to the right page and play a little mini-game to resolve them (and ideally, you the GM already know those rules and simply run the minigame quickly). A system that has a lot of these minigames is less complex for me in actual practice than one that has none, because I'm never asked to make up the rules for said minigame or consider all the implications of a ruling on the fly. It might happen that we eventually notice problems in the minigame if we use the rules frequently, and that's a problem, but that's still likely to happen if I smithed out a ruling on the fly (and perhaps more likely to happen).
Last edited:
L
lowkey13
Guest
*Deleted by user*
Celebrim
Legend
The relationship between "rules" and written law and "rulings" and common law is something I've thought about before.
I've always been a critic of the phrase "rulings not rules" because ultimately both seemed to me to be much the same thing - regardless of whether the law is derived from a Constitution or a body of common law, it's still going to be law and fodder therefore for lawyers.
I've never been a lawyer, but it seems to me that the great difficulty in being a lawyer always come from the common law, which in any legal system soon vastly outweighs the written law both in its volume and its influence over how a case is adjudicated. I've read the written law and it's often struck me how the written law is just as badly written as RPG rules, and often much more so, so that it doesn't actually address the question or provide the answers you'd want to have in practice.
I don't think it is possible to write law or rules that are so complete that no common law arises from the interpretation of them. But I think it is possible to write law or rules to varying degrees of quality so that on the whole, the situation is more ruled by the law than fiat and almost everyone reading the rules has some close sense regarding what the laws say and how they'll work in practice.
I likewise think that there is a more or less ideal state where the rules are short enough to be comprehensible and yet comprehensive enough that they seldom give rise to the need for rulings touching on things that the rules do not cover adequately.
My suspicion is that that ideal state is for a rather large body of written rules, both by practical experience (trying to apply different rules set) and theoretically in that I think that the complexity of a rules set which involves simple operations tends to grow at a less than linear rate. That is to say, I tend to think that doubling the page count less than doubles the complexity - at least for certain types of writing (the sort I prefer).
In fact, I think you can increase complexity by shortening the rules. For example, consider the following variant:
1) If the proposition is trivially easy, it always succeeds.
2) If the proposition is impossible, it always fails.
3) If the proposition is doubtful, then the GM decides the outcome.
For most traditional RPGs, since the GM is the sole authority on resolving propositions, so this reduces to a game with only the following rule:
1) The GM decides the outcome of propositions.
This is in fact a game which contains only the rule frequently referred to as "Rule Zero". Despite having the same number of rules as "The World's Simplest RPG", the complexity of "The World's Simplest RPG" defined by the number of factors that touch on the resolution is zero. While the complexity of game based solely on "Rule Zero" has a complexity that approaches infinity. Since the "Rule Zero" game is the only rule in a Braunstein, I see the entire history of RPG rule development as an attempt to reduce the complexity of the rules compared to that of a Braunstein.
Last edited:
Similar Threads
