Stupid Dice: An Epidemic

[Illetizgerg]

I am unsure of where to put this, however it has spawned from my own personal considerings and initial thoughts regarding a system, and thus I think it is most appropriate on this board. While people just starting to design their game can benefit from this, I think anyone who has designed a game owes it to themselves to read and consider their dice engine thoroughly. I hope this will help you in that endeavor.

Hello, my name is Gregory Zitelli, although you may call me Greg if you'd like. I realize I'm rather new here, however I feel that this entire community may be harboring a rather deadly infection. This disease is being transmitted rapidly, and I feel I am obligated, regrettably, to inform you that your RPG may be at risk of contracting "Stupid Dice". Fret not, for I am hear to inform you of the measures you can take to reduce your risk of infection, and of passing it on.

Despite the members of The Forge having a possibly unhealthy obsession with the Power 19 (though I do admit that the questions are incredibly poignant), people don't really seem to delve that much into items 10 and 11. I personally believe that item 11 is particularly important in game design, as the overall system probably transmits more about what your game is about than anything else (that was a confusing sentence). Still, people don't really seem to break things down, so allow me.

Doctor's Orders #1:
The mechanics of your system are there to set in the stone the relationships between members of the game, whether they be players or the GM (if one exists). Otherwise, your dice are stupid.

RPG design, similar to everything else, abides by the rules of KISS: Keep It Simple, Stupid. Complexity slows things down and detracts from the emulated world, something I think most of you crazy narratists are heart-wrenchingly concerned with. Consider the Tenacious Tres:

1) d20 System
2) Rifts d% System
3) Storytelling System

The d20 System uses a d20 with straight modifiers, adding a consistent 5%. What happens when you get an extra +1? 5%. What happens when a situation gives you a -4? -4 * 5%. What happens when-- You get the idea! The system is predictable, although it does have some obvious flaws, like an unimagined roll-vs-target combat system.

Compare that to something like the Storytelling System. While this system is still actually not bad, let's take a look at trying to predict it. Consider the chances of rolling a success using one die (a success being an 8, 9, or 10).

1 Die: 3/10, 30%

What happens when you include two dice? How about the range of success?

2 Dice, at least 1 Success: 51/100 = 51%

What about more?

3 Dice, at least 1 Success: 657/1000 = 65.7%

4 Dice, at least 1 Success: 7599/10000 = 75.99%

30% --> 51% --> 66% --> 76%? 21% diff to 15% diff to 10% diff? What exactly could the reasoning be behind this system.

Now, that isn't to say there isn't any reasoning behind this, or that bizarre systems do not have a place. What should be understand, however, is that systems need to meet certain criteria based on the games their running, and usually complex systems are designed for no good reason. This is the kind of thing you need to stay away from.

Let's talk about the Neuroshima core system. It's a roll-under dice engine using 3d20, where two dice have to roll under your target number. If you've got eyes in your head then you should immediately be able to witness the obvious seepage and discoloration, positively diagnosing the system with Stupid Dice. This system is not beyond predicting, I could try and possibly succeed if I so chose. The question is why? Why make a dice system with these weird rules, with no backing?

Doctor's Orders #2:
Your system should only use a complex dice/card system if it produces a specific, unique effect as far as the relationships between participating game members are concerned. Otherwise, your dice are stupid.

If your game is a classic style RPG, where the GM controls things and the players play their characters, then you have no use for a complicated dice system. The places that they do come in handy is when you are not simply using dice to make your world random.

Many narration-based games allow the players to heavily impact how the story is ultimately narrated. In these situations dice engines can be built around the idea of randomizing how successful the players are at controlling the universe. In these situation nobody really cares about reality.

Doctor's Orders #3:
Realistic systems do not allow players to mess with the universe. If you're building a realistic system then player's need to be able to make rational choices, knowing that the outcome is as predictable as it is in real life, and your system should reflect this. Otherwise, your dice are stupid.

Does this mean that realistic systems are good and narration systems are bad? Of course not. Different strokes for different folks. You cannot, however, design a system where the player's get to dictate what happens if they win a die roll, and then just say it happens, claiming that it's rational. That is simply not possible, although it could make for a more fun experience. You decide.

This leads to the idea of gimmick dice engines, and where there place truly is in the RPG world. A gimmick system can usually be identified as a system where a player is rewarded for their ability to perform a specific skill or build strategy according to one tactic. This differs from more realistic RPGs where players are rewarded for accomplishing goals such as completing an ultimately randomized task.

To go into more detail, I have a friend who built an RPG where poker is played during combat. Attackers and defenders draw and redraw cards based on stats and skills, trying to make the best poke hand they can. While the game is engineered so there is some realism in the balance of the skills and stats, you're really judging a person's ability to play poker, thus the system is unrealistic.

Alternatively, a system like the d20 System is more straightforward. The advantage there is that you know what the probability is of success, and you also know what you can do in order to increase that probability. These kinds of games can actually promote more strategic thinking, because you can analyze the predictable rules of the game in order to increase your chance of success (like using cover in combat). The more gimmick is introduced into your system the less this can be done, although it does usually have the significant advantage of often being more fun.

[Vulpinoid]

So your basically saying that anything other than d20 is stupid???



I do agree that there are some dice engines out there that seem illogical, but if you look through a few of the design threads around here you'll see that a lot of people have a fondness of bell curves, and the best way to get a "realistic" distribution of results is to use multiple dice and add the results (or average them). Another result path developing an underground following is the logarithmic scale, which usually takes a complex die function to achieve with any decent realism. There are also a few other methods of randomisation, but I'll stick to the term "dice" for consistency with your posts.

I think that certain concepts you have posted in this thread are a bit naive and require some further investigation.

I'll address my immediate concerns to you "doctors orders" in turn.

• Doctor's Orders #1:
  The mechanics of your system are there to set in the stone the relationships between members of the game, whether they be players or the GM (if one exists). Otherwise, your dice are stupid.

  Personally, I'd always thought that a game's mechanics were there to define the way that the players and GM interact with the world rather than a method they use to interact with one another. The social contracts inherent in gaming set the structure for inter-player relations, the dice set the structure for inter-character relations (whether PC-PC or PC-NPC). If you're using dice to set the social contract, I'd say that was stupid.
  
  As you go further into mechanics, you forget that d20 might have a dozen different modifiers to a single roll which slows things down dramatically and interactive feats which react off other effects in the game to possibly produce bonuses. This can be just as slow, overwhelming or unrealistic as any other system out there.

• Doctor's Orders #2:
  Your system should only use a complex dice/card system if it produces a specific, unique effect as far as the relationships between participating game members are concerned. Otherwise, your dice are stupid.

  This sort of addresses the bell curve point, proving that sometimes it takes an interesting mechanic to get specific effects into play, but your editorial then runs directly counter to the point just made. Flat probabilities are unrealistic, true bell curves are closer to realism in some ways, logarithmic scales are closer to realism in other ways. Every designers has a belief behind why they place their die mechanics in place.

• Doctor's Orders #3:
  Realistic systems do not allow players to mess with the universe. If you're building a realistic system then player's need to be able to make rational choices, knowing that the outcome is as predictable as it is in real life, and your system should reflect this. Otherwise, your dice are stupid.

  Sorry...No...wrong.
  
  Realistic systems do not allow characters to mess with the shared imagination space in a way that disrupts the established reality. Players mess with it all the time. It's like quantum physics, observation makes a difference and nothing is fully predictable. A character's decisions should never result in entirely predictable outcomes because the characters never have all the information needed to make a 100% accurate assessment of the situation at hand. If you want a "real world" example of this, consider the information that meteorologists have available to them then consider how often their predictions are about the state of weather over the next week.
  
  I believe that if there's a 99% chance that an event will happen, or if it's good for the story and makes sense to the players then it probably should happen. But I'm also an advocate that if a failure could cause a dramatic twist in the story that might prove a bit more enjoyable, the leave it in the hands of the dice. The dice are a tool toward storytelling, simulation and game effects, not a straight-jacket that define reality in black and white.

I've also noted that you mention RIFTS, which I believe is an outdated system that probably should have seen a major overhaul ten years ago. (I would have liked to see Palladium turn it into a d20 game, since the Palladium system did just start as a D&D knock-off...but that's another story). Personally, I dislike RIFTS (despite owning dozens of books for it) because of it's mish-mash of systems...use a d20 here, use a d% there, half the attributes give benefits in some way, while others give special d% abilities...there's just no coherency in the system and that's a pet hate.

Sorry, but d20 isn't the "be-all-and-end-all" of gaming, and there are a decent number of aspects in that system that are also Stupid!

The community here may not be too overly concerned with the die rolling of a game, and that was something that struck me as odd when I first showed up 12 months ago. I wondered how a game design forum could have such a lack of interest in the minutiae of the mechanics, but then I realised that a lot of people around here are more interested in the big picture stuff. The dice are just a part of that whole.

V

[greyorm]

Many narration-based games allow the players to heavily impact how the story is ultimately narrated. In these situations dice engines can be built around the idea of randomizing how successful the players are at controlling the universe. In these situation nobody really cares about reality.

Hi Greg. I admit I almost stopped reading right there because this is...well, nonsense. It isn't true.

I could state in counterpoint that maintaining the established reality, despite players holding powers of narration, is central to all narrative games. Everyone does, in fact, care about reality. But I don't know if you're talking about internal narrative consistency or trying to claim something else, here.

It seems to me you are assuming the role of the dice in a game are to function as a measure of the specific capabilities of a given character or other game-object in the imagined space, rather than as tools that can and do serve roles outside that function (providing pacing, dramatic tension, metagame resource levels, or effectiveness). Would I be correct in suspecting this is because you have only seen/played in games where that is how the dice work, and have no frame of reference outside that idea, and maybe are having difficulty imagining how dice could serve any function other than as a set measure of character skill and character-success probability?

I will note that your statement might be true if you stop to define what you mean when you use the term "realistic". What, exactly, do you mean with this word? Consistency with established physical principles? Faithful reproduction of source material despite known physical reality? Adherence to probabilistic outcomes? Unfortunately, there are so many different meanings for this word that it is impossible to understand the nuances of your argument, especially when your argument confuses/conflates immersion, physics-emulation, and measuring probability by lumping all these disparate notions into the word "realism".

You might find this LJ entry by Rob Donoghue interesting. It is about the different sorts of "realism" one can strive for and that the mechanics can reflect if you are using the dice to measure probabilities of success.

Does this mean that realistic systems are good and narration systems are bad? Of course not. Different strokes for different folks. You cannot, however, design a system where the player's get to dictate what happens if they win a die roll, and then just say it happens, claiming that it's rational. That is simply not possible, although it could make for a more fun experience. You decide.

Sure! It is completely rational depending on what the dice or the results of the dice are actually measuring.

Let me explain: you've just argued certain systems -- let's take one of mine: ORX -- are "irrational" because the dice do not emulate and reflect specific character-centered probabilities, only dictate a win-or-loss and narration. In ORX, this is not irrational because the attributes being rolled with the dice do not emulate either the amount of effect of the attribute based on "reality" or the actual current state of the attribute (ie: "injury" is not considered actual injury).

It is not irrational because the dice are not being used to measure or contrast the reality of the game-world, except in a highly abstracted sense; it only appears "irrational" if someone mistakes the dice as doing so, and in that case the fault lies with the individual failing to grasp the nature of the game and the way the dice function differently in the set-up. "What happens" and "what is likely to happen" and "what is a reasonable addition to the narrative" is maintained at the social level, supported by the social contract at the table, the choice of game style, consistency with the existing narrative and background, and with what is known about the characters, with some minimal narrative detail/constraints provided by the roll itself (the type of roll and the success or failure in relation to the Stakes).

That is: you don't make a roll to arm-wrestle. That isn't the function of the dice. You make a roll to narrate the events of a scene relying on your physical nature and whether or not you get what you want out of it. But how strong you are? The dice have nothing to do with that part of play. It is simply not a function of the dice roll.

Probabilities? Realism? You can easily guess your chances of success and failure, and strategize your choices in terms of the metagame resources the dice do represent and measure. Nor do you suddenly become Superman because you decide to narrate you have lifted a building (not unless you ARE Superman and CAN lift a building), nor does a scrawny nerd become Superman and punch you through a wall because you failed the roll.

Make sense? Probably not. Play some more games first.

Oh, BTW, you state "the overall system probably transmits more about what your game is about than anything else" -- and I must say congratulations and welcome! You have independently stumbled upon a core principle of design-theory here at the Forge. We call it "System Does Matter" and it is the reason why all the color-text or admonitions to role-play in the world won't save the actual mechanics of certain games from dispensing D&D-type dungeon-crawl analogues using what end up being super-heroes without heavy modification.

[Illetizgerg]

So your basically saying that anything other than d20 is stupid???

Absolutely not. What I'm saying is that the most basic groundwork of the system is far easier to use than most other systems I've seen, and I think people tend to shy away from straight single die rolls plus a modifier solely because they prefer their system to look unique, which is ultimately pointless.

I do agree that there are some dice engines out there that seem illogical, but if you look through a few of the design threads around here you'll see that a lot of people have a fondness of bell curves, and the best way to get a "realistic" distribution of results is to use multiple dice and add the results (or average them). Another result path developing an underground following is the logarithmic scale, which usually takes a complex die function to achieve with any decent realism.

First of all, you have to realize that I was one of those kinds of people not too long ago. What nobody ever seems to address, however, is the fact that weighting the outcome can be done through the manipulation of numbers outside of realtime. For instance, a system like the Storytelling system (where you add more dice to a pool and take the highest) approaches a number; however the advantage that each new die gives you decreases the more you have. This can be imitated in a linear system by using a number of different methods.

The first method is handled by increasing the amount of experience or character creation points needed to increase a skill by one point. For instance, you could say that in order to increase a skill by one point you must pay the number of points currently in the skill plus one. This would mean that getting a skill of 1 would take 1 point, a skill of 2 would take 3 points, a skill of 3 would take 6 points, and so on (this specific example exhibits a quadratic increase, however you could make it whatever you want). Using this method every single point of character advancement does less and less, however the gradation compounds in an odd way.

Another method, which is far simpler and can solve other problems, is simply changing the way that target numbers progress as a task gets more difficult. Instead of simply increasing linearly (an easy task is like a 5, a medium task is around 10, hard is 15, etc), you can curve the difficulty in order to produce basically anything you want.

Personally, I'd always thought that a game's mechanics were there to define the way that the players and GM interact with the world rather than a method they use to interact with one another. The social contracts inherent in gaming set the structure for inter-player relations, the dice set the structure for inter-character relations (whether PC-PC or PC-NPC). If you're using dice to set the social contract, I'd say that was stupid.

Control of the game world must be partitioned is some way between the GM and the players. Usually the GM holds most of the power, however it varies from game to game. It seems that people on The Forge enjoy pushing their players into more narrative driven roles, and so dice mechanics can be used in order to balance out power.

In games designed to be realistic, the game world is "ruled" by the GM. It is his/her job to maintain the world, however the dice engine usually does not effect this all that much. Generating areas, determining treasure, and other things that are commonly randomized can be done using the engine, however that is not really its purpose.

When it comes right down to it, I do think that a dice system is ultimately designed to regulate how the GM and the players interact. This isn't really contradictory to what you said, as the GM controls and represents the game world. Consiquently, whenever a player attempts to interact with the game world, the dice engine is used to determine how the interaction is handled. The reason that I chose to distinguish between the GM and the game world is because many systems allow the players to circumvent the game world in order to effect other parts of the game, usually through an interaction with the GM.

Realistic systems do not allow characters to mess with the shared imagination space in a way that disrupts the established reality. Players mess with it all the time.

I don't really understand with what you mean here. These two statements would appear to mean (when combined) that systems are not realistic.

It's like quantum physics, observation makes a difference and nothing is fully predictable. A character's decisions should never result in entirely predictable outcomes because the characters never have all the information needed to make a 100% accurate assessment of the situation at hand. If you want a "real world" example of this, consider the information that meteorologists have available to them then consider how often their predictions are about the state of weather over the next week.

Well, the quantum physics thing doesn't really apply to the situation. You also seem to be ignoring the random aspect of the situation.

Random dice are used to emulate just how unpredictable a situation is. When I talked of predictability, I was referring to the player's ability to predict the statistically likelihood of success, which is entirely realistic. If I go to a shooting range and fire a billion shots at a target, and half of them miss, then I know that (statistically) I have about a 50% chance of hitting my desired target. If I fire the exact same gun under the exact same circumstances do I know if I will succeed? No. But I know about how likely I am to succeed.

I believe that if there's a 99% chance that an event will happen, or if it's good for the story and makes sense to the players then it probably should happen. But I'm also an advocate that if a failure could cause a dramatic twist in the story that might prove a bit more enjoyable, the leave it in the hands of the dice. The dice are a tool toward storytelling, simulation and game effects, not a straight-jacket that define reality in black and white.

I don't really see how this disagrees with what I've said. If someone wants to make a system where events are altered according to what makes a better narrative then a system can be made to accommodate that, and the system (which regulates the relationship between the GM and the players) would simply remove the dice from the situation in order to ensure that an event happens.

I've also noted that you mention RIFTS, which I believe is an outdated system that probably should have seen a major overhaul ten years ago. (I would have liked to see Palladium turn it into a d20 game, since the Palladium system did just start as a D&D knock-off...but that's another story). Personally, I dislike RIFTS (despite owning dozens of books for it) because of it's mish-mash of systems...use a d20 here, use a d% there, half the attributes give benefits in some way, while others give special d% abilities...there's just no coherency in the system and that's a pet hate.

From what I know, Rifts uses a d% for skills and a d20 for combat. That doesn't seem all too psychotic to me, but I've never played it, and I'll certainly allow you your pet peeves considering I have so many of my own. I brought it up because most gaming circles I'm familiar with have, at the very least, heard of it, and its system is a rather straightforward roll-under d%.

Sorry, but d20 isn't the "be-all-and-end-all" of gaming, and there are a decent number of aspects in that system that are also Stupid!

Not to bite your head off, but just because I don't openly smash-talk Wizards doesn't mean that I'm not annoyed or spiteful about their products. Please don't assume that I'm in love with the game.

[Illetizgerg]

It seems to me you are assuming the role of the dice in a game are to function as a measure of the specific capabilities of a given character or other game-object in the imagined space, rather than as tools that can and do serve roles outside that function (providing pacing, dramatic tension, metagame resource levels, or effectiveness). Would I be correct in suspecting this is because you have only seen/played in games where that is how the dice work, and have no frame of reference outside that idea, and maybe are having difficulty imagining how dice could serve any function other than as a set measure of character skill and character-success probability?

I did talk about how systems may be designed in order to resolve things related to narration. In fact, I specifically mentioned that this kind of system is the one case in which simplicity is usually sacrificed.

I think I need to clarify how I attempted to break down games based on their adherence to realism.

A game centered around perfect-realism establishes a game world, decides on rules for that game world, and then enforces those rules. In this situation dice can be used in order to reflect aspects of this world, as far as statistics and probability goes. Realism, as far as I was using it, is with respect to this world. Things happen because the game world has decided that there is a specific chance that they will, and chaos "chooses" them.

When games give players access to things other than just their character's choices it loses this realism. If a game is centered around measuring character's abilities, what justifies a player's ability to manipulate things outside of their character's choices and actions?

Hi Greg. I admit I almost stopped reading right there because this is...well, nonsense. It isn't true.

I could state in counterpoint that maintaining the established reality, despite players holding powers of narration, is central to all narrative games. Everyone does, in fact, care about reality. But I don't know if you're talking about internal narrative consistency or trying to claim something else, here.

...

Let me explain: you've just argued certain systems -- let's take one of mine: ORX -- are "irrational" because the dice do not emulate and reflect specific character-centered probabilities, only dictate a win-or-loss and narration. In ORX, this is not irrational because the attributes being rolled with the dice do not emulate either the amount of effect of the attribute based on "reality" or the actual current state of the attribute (ie: "injury" is not considered actual injury).

It is not irrational because the dice are not being used to measure or contrast the reality of the game-world, except in a highly abstracted sense; it only appears "irrational" if someone mistakes the dice as doing so, and in that case the fault lies with the individual failing to grasp the nature of the game and the way the dice function differently in the set-up. "What happens" and "what is likely to happen" and "what is a reasonable addition to the narrative" is maintained at the social level, supported by the social contract at the table, the choice of game style, consistency with the existing narrative and background, and with what is known about the characters, with some minimal narrative detail/constraints provided by the roll itself (the type of roll and the success or failure in relation to the Stakes).

I admit that what I had outlined does not cover these specific kinds of games, and I think this really boils down to how RPGs choose to describe what the player has control over.

If a game has the player taking control of a specific character, with measurable stats determining the likelihood of certain (usually positive) situational-based outcomes, then the "realism" of the game is tied into what is being measured and then randomized. When a game then gives a player partial control over things unrelated to their character's choices then it needs to be identified as a separate responsibility, which it practically never is. This does mean that the universe is going to be skewed towards the personal goals of certain individuals, which is where I got the idea of irrationality from.

To sum it up, a game does sacrifice realism and introduce irrationality if a player's responsibility lies with control of their character's actions, and then the player is also given control of something else. If a game is designed with the players taking control of narration, then it is correct that what I said does not apply.

Make sense? Probably not. Play some more games first.

Well that took the top layer of skin off. I suppose I might have come across as being rather antagonistic, so it was probably deserved.

[Illetizgerg]

Looking back on the way that I chose to present my ideas, and the way that they were interpreted, I actually feel rather embarrassed, and I would like to establish that I didn't intend to come across as a jerk or anything. This isn't to say that Vulpinoid and greyorm did not pose appropriate responses; exactly to the contrary, in fact. I enjoy dialog like this greatly. I just wanted to say that I'm sorry, because I'm pretty sure that my initial post could be interpreted as being very negative and disrespectful, and that's not the intent that I had when I clicked the New Topic button.

[Vulpinoid]

I'll editorialize again by saying...

If you didn't want to be antagonistic, why did you give the thread it's name? And why did your opening dialogue include the words...

however I feel that this entire community may be harboring a rather deadly infection. This disease is being transmitted rapidly, and I feel I am obligated, regrettably, to inform you that your RPG may be at risk of contracting "Stupid Dice".

Sounds pretty antagonistic to me.

Hence my response.

Similarly, in your response to Greyorm, you state...

I admit that what I had outlined does not cover these specific kinds of games

What sort of games did your issue cover? Most of the games around here are independent and are trying to explore unusual and non-traditional forms of roleplaying. Your quote as mentioned at the beginning of my post seems to indicate that your attitude is aimed squarely at the games produced by independents, such as those here at the Forge. Please, back-pedal all you will...

I'm not going to comment much further on this thread until you give us a little more to go on.

• Please tell us your rationale for why narrative games cannot be realistic.
• Tell us in your word why control of the game world must be partitioned between the GM and the players.(I have ideas on this but I'd like to hear your perspective.)
• What sort of games do you enjoy playing? What are you aiming for in your discussion. Surely it has to be more than just slagging-off exotic die mechanics.

V

[Ron Edwards]

Hello,

It's time for me to moderate this thread.

1. It is not a First Thoughts thread. It's not about a game idea. It cannot continue here, if it is to continue at all.

2. It raises, or at least points toward, some interesting questions. However, since it's not describing any play or context for play, the content simply pushes people's buttons and becomes a non-constructive argument.

3. It's presented as an opinionated rant in a "take it or be wrong" manner, which isn't acceptable here. To post at the Forge means you are willing to let the responses of others affect your own views.

Illetizgerg, when you begin a thread with "I am uncertain where to post this," then ask me in a private message, or ask generally in the Site Discussion forum. That is the only reasonable option. What you've done is like saying, "I'm uncertain which counter is used to pay for this item, so I shall put it in my pocket and walk out of the store."

Michael and Raven, he pushed your buttons. I do understand your efforts to turn the thread into a discussion, but that line about "uncertain where to post," and the off-forum topic, should have tipped you off. Once off those tracks, the button-pushing is almost inevitable.

The thread is now closed. For people who haven't been here long, that means no one is to post to it again. It's not locked; physically, you could post to it - but do not. If you can't understand that, then you aren't grown up enough to be here at all.

That does not shut down the discussion, or rather, the reasonable discussion that might still occur. I invite the same topics to be raised in the Actual Play forum, by anyone, in the context of actually having played a game, during which those topics were relevant. Illetizgerg, that's the way to go for a good experience here and a fair consideration of your ideas.

Best, Ron