Rationale for random numbers in rpgs

[Thierry Michel]

Could anyone point me to a discussion on the rationale of the use of random numbers in rpgs ?  What does the roll of a die represent exactly ?

[Le Joueur]

[What is] the rationale of the use of random numbers in rpgs?

I've heard quite a few following the release of Amber: The Diceless Role-Playing Game (by Erick Wujcik and Roger Zelazny) ranging from "that's how it's done" on up to some of the most obscure.  I've playtested various permutations of diceless and dice-based games and came to a simple conclusion.

Random numbers make role-playing games seem less arbitrary.

Take a moment to let that sink in.

What does the roll of a die represent exactly?

Nothing.  Everything.  Anything.  There is no single answer to this question.  Throughout the history of role-playing game design, the roll of a die has been used to represent just about anything you care to name.  And that's the problem with trying to 'figure out' what random numbers are for.

Some say that they represent factors 'too small,' coincidental, or too innumerable to be listed in a portable role-playing game.  Others believe they represent luck, chance, or taking advantage of situations too specific to parse out in rules.  Still others want them to generate details, boost creativity, or vend the unexpected.

It was from that last one we (I and my partner, soul-mate, and wife) found most telling.  Traditional role-playing games have a gamemaster who pretty much does everything; take away the dice and the players will tell you, in 'old fashioned games' it feels like the gamemaster is simply leading them around (not always a bad thing, by the way).  In other words, it feels arbitrary.  Because the gamemaster decides everything, no matter how objective they really are; it looks arbitrary.

The dice on the other hand have no 'hidden agenda' nor seem to.

That is what we concluded random numbers were for.

Fang Langford

p. s. Sorry this is so clipped, I've got an emergency...

[Ron Edwards]

Hello,

I believe the diversity of randomizers' uses doesn't lie with the diversity of randomizers, but rather with the diversity among role-playing goals.

Here are some of my thoughts about it.

Contrast how the dice results are used relative to the dialogue and establishment of "what happens."

1) Setting up starting conditions. A lot of character creation used to operate in this fashion; I'll present one of the simplest examples of starting cash in old-school fantasy play. You roll 3d6, multiply by 10, and that's your resource for outfitting your character.

The Fortune starts the process, and you take its results and apply them with a free hand.

2) Springboarding. This is the essence of Fortune-in-the-Middle during resolutions in many games; I'll present an example from Hero Wars play. In this game, during a clash or conflict, players' and GM announcements are made in general terms, providing only enough information to convey the desired goals/outcomes and perhaps the general style or approach of attack. When the dice hit the table, perhaps favoring character A, the events are interpreted "backwards," including such things as character B never managing to get his attack started in time before being pre-empted.

The Fortune provides "meat" to the process - it must be consulted and its outcomes cannot be ignored, but it does not resolve already-stated actions, but rather permits those actions to be (now) produced through dialogue.

(Many people use this technique while playing games which don't specify it or whose text implies otherwise; in Hero Wars, you have to use the Springboarding interpretation or the game makes no sense.)

Fortune-in-the-Middle resolution is only one application of Springboarding. Also, it's often combined with metagame modifiers which permit one to adjust the dice outcomes according to a resource (e.g. Hero Points in Hero Wars).

3) Final-arbiter. In resolution, we've taken to calling this Fortune-at-the-End; I'll use an example again from solid Simulationist play. In GURPS, in an exchange exactly like the one above, before the dice hit the table, all the "moves" of the opponents are already known to everyone and are considered immutable. The dice deal with the moment-of-truth only; they "take over" narration, if you will, and tell us how it comes out.

The Fortune is the outcome-determinant, as well as the procedural final step in the process - the imaginative or narrational contribution is essentially complete prior to the dice's "this is what happens" contribution.

I submit that any and all comparisons of Fortune's uses, goals, techniques, and related matters have to start with the distinctions I've drawn above - crossing the categories will cause major confusions in trying to compare anything.

Best,
Ron

[Thierry Michel]

Just to clarify: where would, for instance, "wandering monsters" tables fit in that classification ?  # 1 ?

[Emily Care]

Thierry,

Excellent question.

As far as the role of randomization in  roleplaying, I stand squarely with Fang on this issue.  IMO, random numbers, and all mechanics exist, in order to enhance a participant's ability to engage in the game.  I refer you to this thread.

There is a legacy from the wider world of gaming to contend with: randomization in card and board games allocates resources (ie. what cards you are dealt, how many squares you can advance),  and deters cheating (ie cutting the deck).  I see the use of it in start up conditions as an attempt to address both of these issues.

But the resources potentially available to you in role-playing are much broader  than in non-roleplaying games. They include all of the rights generally given to gm's: world creation, plot and narrative creation, application of mechanics, outcome resolution, etc) and in part due to this wider field, the  issue of cheating is much less clear.  What people consider "fair" or "appropriate" in a roleplaying game varies wildly based on the assumptions you're operating under and what kind of play you are looking for.  A sim-oriented player might consider the narrativist-oriented players dynamic character development to be breaking the initial contract of what one can expect from the character in questions.

Use of random numbers in final-arbitration of conflict resolution also enhances credibility by operating to deter the semblance of cheating.  The participants can agree before hand on the specfic operation by which a given type of encounter is resolved, and thus the gm's (or whoever interprets the results) narration of results is accepted as a credible addition to the narrative.  Springboarding does the same thing in a different order.  

The use of randomization gives the sense of an independent authority determining exactly which version of events is accepted as a valid play.  This helps everyone agree, ie facilitates concensus.

Wandering monsters sounds a bit like springboarding as Ron described it: the gm gets to choose when there will be an encounter, and the choice of type of encounter is pre-determined (what's on the table).  The roll determines which possible outcome gets chosen.  It would be part of starting conditions if the gm rolled on the table while creating a map of the dungeon, or what-have-you, and wrote into various spots what the players would encounter when they reached the spot.

--Emily Care

[clehrich]

Emily brings up an interesting point about cheating, and the ways in which it is and is not the same thing in RPGs and other game forms.

I have to say, though, that the majority of times I've seen cheating in RPGs has been about die-rolling: someone rolls 2d10, for example, and then says "Oh, well, actually THIS one is the 10s die, not the other one," or just rolls repeatedly (as though out of habit) and then suddenly says, "Hey, I'm doing X, and here's what I rolled."  And so on.

Cheating is a lot harder with cards --- you have to have some skill, really.

One other way of looking at the randomization issue (Fortune) is in reference to what is sometimes seen as GM cheating, where (in some games, at least) the GM is supposed to be bound by the dice.  I think, though, that this is not a common issue, since even in D&D most DMs I think use a screen, behind which they could roll horribly and still announce that they rolled wonderfully.

[Walt Freitag]

Why determine or modify outcomes using random numbers in RPGs?

Let me suggest a surprisingly simple answer: because it's fun.

No, seriously. Although fun is supposed to be so subjective as to be outside the scope of what can be rationally discussed here, there's a simple principle that I believe applies to most players and most types of game (and other forms of play) most of the time: reacting, with at least some degree of effectiveness, to events beyond one's control, is where the fun is.

While there are many possible sources of events beyond one's control, most notably including the actions of other players (which is why strategy games with no randomness, and diceless RPGs, don't lack in fun), random numbers are the most available and straightforward.

Now, there's a lot we could say about the ex post facto implications and consequences of particular uses of randomness in particular styles of play (with Ron's post providing a starting road map). But if you're looking for as basic and universal a "why" as possible, I think the answer is "because it's more fun to have certain occurrences that are beyond any participant's control." (And why is that? I think it has to do with the adaptiveness of play as a part of mammalian development; particularly, explorative play that exercises the use of higher brain functions to survive unexpected circumstances.)

- Walt

[Bankuei]

Walt,

I cannot agree with you more.  You've said just what I was thinking.  

Part of the joy of games is that they contain controlled surprises.  Unlike things like finding out that your girlfriend is pregnant, getting into a car accident, or finding a 20 dollar bill on the street(all of which are exciting, in their own special way),  when you play a game, you are playing in order to provide yourself with the entertainment that comes with unknown outcome, but in a limited range of potential outcome.

Certainly there are other means available for randomizing things(coins, cards, spinners, pulling names out of a hat, reading goat intestines, etc.), that have been used for years for both divination and gambling, dice are easy to find, transport, and calculate odds for.  

Dice also are a means of taking some power out of the traditional godlike GM.  I think even at its earliest points, roleplaying in general recognized that a  "GM only" power structure doesn't necessarily lead to random outcome, hence the use of dice.  

Chris

[Le Joueur]

Thanks everyone,

I was trying to get at all these points but nary had the time.

Ron,

[list=1][*]Randomized "starting conditions" are simply one opportunity to begin with something other than the arbitrary "100 points."  (Would that be Fortune-as-the-Start, FatS?)
[*]In a FitM event, a 'vegetarian diet' (one having no "meat") will require many arbitrary choices on somebody's part.  (Randomizers that don't produce numeric results - like the dice-and-chart - also give 'middle events' without being necessarily FitM.)
[*]The same is true with FatE, where the dice make the call.[/list:u][/list:o]

Thanks for clarifying some examples of

when randomization can occur.

Emily is right about cheating, fairness, and credibility.  All of which support my point about arbitrariness (how credible is it to say 'you get to play' and then make choices for the players effectively not letting them play).

Clehrich hints at something I wanted to address, but hadn't the time (no worries: emergency handled), the idea that, by choosing when to invoke randomness and when not to, a traditional gamemaster can covertly overturn the 'objectivity' of the dice.  That's why I always say "the feeling of arbitrariness;" true objectivity is almost impossible.

Walt turns the issue around, taking my rather negative stance 'it suppresses arbitrariness' and reversing it, 'because it is more fun.'  This stands to reason as the real reason to do anything with game design, but I like the way he puts randomizing it.[/list:u]I think it also speaks to how a designer can use the limitations of randomizers to create a 'safety zone' for the fun to exist in; I've found that most people aren't comfortable with 'no limits.'

And that's where

Chris comes in and ties it all together.  (Well said Chris.)[/list:u]That leaves me pondering one thing though...why did Thierry ask such an open-ended question in the first place?  I put it in terms of a comparison to 'randomless' systems for lack of a better standpoint, but what is the real driving force for his curiosity?  It's certainly peaked mine.

Fang Langford

[Cassidy]

[What is] the rationale of the use of random numbers in rpgs?

I've heard quite a few following the release of Amber: The Diceless Role-Playing Game (by Erick Wujcik and Roger Zelazny) ranging from "that's how it's done" on up to some of the most obscure.  I've playtested various permutations of diceless and dice-based games and came to a simple conclusion.

Random numbers make role-playing games seem less arbitrary.

Fang hit the nail on the head for me although not everyone will agree that conveying a sense of [/i]arbitrariness to players is an important aspect of play or even necessary.

For me it is though.

I like being surpised. I like the fact that the fickle hand of fate can throw GM and players a curveball and swing play in an unexpected direction usually when least expected.

I love the suspense you get at those pivotal points in a game when the roll of a die or the turn of a card has everyone on the edge of their seats and no-one really knows what's going to happen next. Those moments are priceless.

[Thierry Michel]

That leaves me pondering one thing though...why did Thierry ask such an open-ended question in the first place?

I thought it was a reasonably narrow question, actually. I was interested specifically in how self-conscious was the choice of using random numbers.
In particular, did designers have in-game justifications for using them ?

The answers are interesting, first because they all point in the same direction, second because they clash with some of my assumptions.

As I'm both a statistician, a gamer and interested in random (and combinatorial) literature, I'm particularly curious about the way people see random numbers.

[talysman]

As I'm both a statistician, a gamer and interested in random (and combinatorial) literature, I'm particularly curious about the way people see random numbers.

hi, Thierry.

as someone who is also interested in random/combinatorial/surrealist literature, and also because I have recently been generalizing the concepts in my Co9C game and another rpg I am designing (a game meant to emulate rogue-like games,) I've been thinking a lot about the question of randomizers in roleplaying. I've pretty much come to the same conclusion as Walt, but allow me to expand on the concept.

I have taken to defining rpgs as a form of group fiction (not group storytelling, because there are some nonstory forms of fiction, like the vignette, which could be used effectively in roleplaying.) in contrast to ordinary fiction, which has the goal of entertaining and enlightening others, we play rpgs to entertain ourselves.

randomizers and group participation are methods of introducing an element of surprise into the group fiction, because being surprised is fun. also, randomizing events and outcomes helps avoid blocks to creativity; instead of everyone in the playgroup needing to brainstorm quality entertainment off the top of their heads, they use the results of die-rolls or expand on the comments of other players. more surprising detail means more fun for all.

now, whether someone can make an rpg based on markov chains remains to be seen...

[ edited for grammar ]

[Thierry Michel]

I find it counter-intuitive that a random number is seen as less arbitrary than a human decision. I can't think of anything more arbitrary than a die roll.

In fact,  if I understand correctly your answers, the general function of randomness is "as a plot input beyond the control of the participants". Fair enough ? If so, it is precisely the arbitrary nature of a random draw that makes it useful.

[Le Joueur]

Hey Thierry,

Yeah, using common English, I can see where the problem is.  In fact, I had to dig into the dictionary to find any terminology to discuss this whole issue (I did this a while back).

Here are a few of the more telling definitions:

ar·bi·trar·y - adj.

Exercised according to one's own will or caprice, and therefore conveying a notion of a tendency to abuse the possession of power.[/list:u]
ob·jec·tive - adj.

Uninfluenced by emotions or personal prejudices.
Based on observable phenomena; presented factually.[/list:u]
ob·jec·tiv·i·ty - n.

Judgment based on observable phenomena and uninfluenced by emotions or personal prejudices.[/list:u][/list:u]That's how I've interpreted die rolls versus 'anyone's fiat.'  No matter how objective the person making the decision appears to be, there is always room for "a notion of a tendency to abuse the possession of power;" no such notion applies to the rolls of dice because they are objective.

In other words, the utility you see in "plot input beyond the control of the participants" is inherently objective, not arbitrary.  I hope this clears up what I said earlier (which should have been terribly confusing with arbitrary mistaken for objective).  I suspect this is the "clash with some of [your] assumptions;" am I right?

Unfortunately, in the broader world of role-playing game design, few designers make a conscious decision to use random factors.  Most are simply following tradition; I follow tradition

consciously to make a game both accessible to traditional players and be able to use random factors more appropriately.

Fang Langford

[Thierry Michel]

I suspect this is the "clash with some of [your] assumptions;" am I right?

Mostly, yes. Note that for me using the dice only displace the argument since in most cases the GM still gets to choose the difficuly of the roll or even  if there is a need for a roll at all. So it's partly an illusion of objectivity.

[The other assumption I had was that someone would defend the "rpg as a model" argument, where randomness is used to represent noise or tangential factors, as you explained in your first post.]