The role of fortune

[Andrew Morris]

Hmm....okay, let me make sure we're talking about the same thing here. The d20 roll in the example is accepted by consent, because the players agreed (explicitly or implicitly) to abide by the rules as printed. That agreement continues for the course of the game, and covers all the individual incidents. It's not a matter of all the players agreeing whether or not to accept each die roll. Likewise, in the photo example, it's a matter of whether you have the authority to introduce character X's existence and appearance. If so, it's because the participants have agreed on it ("Sure, he's the GM, he can bring in characters and state what they look like.").

I'm not talking about mediation, just agreement and authority.

[contracycle]

OK.  Well I have not disputed that some act of consent is required.  All I have suggested is that a clause in the explicit contract can say, up front, "we will accept the decision of of the dice without challenge".  Thats not necessarily, and is not dependant on, the printed rules.  Once that is done, the results of any given role do not necessarily require further consent.  Again, I am not claiming the dice have an authority of their own - but that they can be granted authority by the social contract.

[Andrew Morris]

All I have suggested is that a clause in the explicit contract can say, up front, "we will accept the decision of of the dice without challenge".  Thats not necessarily, and is not dependant on, the printed rules.

Okay, it sounds like we're saying the same thing, then.

[simon_hibbs]

As a long-time Amber player, and having discussed ADRPG online many times, much of this discussion is very familiar, but as ever with some interesting Forge twists.

As a historical observation, I think randomness in RPGs directly orriginates from it's growth out of diced wargames. The dice in a wargame are there to arbitrarte outcomes because wargames are direct competitions between players and are often played without a referee at all. While tactical board games don't have to use dice (e.g. chess) one advantage they offer is providing uncertainty, and a sliding scale of possible results, in each interaction between pieces. Contrast with Chess in which all interactions between pieces are entirely predictable and produce only binary results.

Roleplaying games are played in a completely different context. They are not streightforward direct competitions between different 'sides'. They have a referee for arbitrarting many conflicts and making decisions not covered by the rules. Therefore many of the reasons for using dice that orriginated in wargaming nolonger necesserily apply.

Most tabletop roleplaying games give the referee wide latitude to make arbitrary decisions without consulting dice, and the degree that they do this varies. In some games "Wandering Monster Tables" are an important element, while in other games it would never occur to anyone that any encounters should be randomly generated. It's just assumed that all encounters are decided by the GM.

Simon Hibbs

[Joshua A.C. Newman]

Boy, do I not agree with that, Simon.

First, Matt Mearls' comment to the contrary, an RPG in no way requires a "referee". Polaris doesn't, Shock: doesn't, Prime Time Adventures, Dogs in the Vineyard, Mountain Witch, and piles of others don't. That's because they rules cover the kinds of situations one will get into in those games. In fact, one kind of situation at which they're all particularly poor is competitive, wargaming-type conflicts.

The difference between a referee and a judge is important here, too. In many mainstream games (GURPS, D20, L5R, etc.), there is a judge. That's the person who decides whether to reward or punish player input. Like the judge in, say, figure skating, that person's job is to determine how much the players' actions conform to a predetermined chain of events. Sometimes, they just have to conform aesthetically to the GM's vision and sometimes they have to conform point by point. If they don't, they don't gain the resources they need to continue: XP, information, whatever.

A referee, on the other hand, is a person who determines if the players are playing by the clearly stated rules. If they are, the scoring is obvious and not subject to arbitration, as in the case of hockey. The referee's job is to have a clear, unbiased perspective from which they can make sure that they can arbitrate close calls and enforce rules.

Now, in Amber (with which I'm not familiar with beyond some quick research just now), I would wager that there is such a judge, because the effectiveness of a given action toward obtaining given stakes is decided on by the GM, same as, I dunno, D&D. This may just address the fact that the dice in D&D can be removed and the game still works, so long as you're not playing a competitive, wargaming-style scenario (aww, snap!). Since, in Amber, you're presumably setting up characters who want something, they're still in conflict (just like in a wargaming situation), so you still need righteous opposition. If that's done by the GM reliably setting up antagonists that seem like they can effectively oppose the protagonist, then maybe that works. I dunno.

So I disagree: wargaming is simply one form of conflict, but many of its requirements are the same. As stories are made out of conflict, removing what dice seem to give you — uncertainty, objectivity — without replacing them with another mechanic means that your conflicts lose punch. In the worst case, your conflicts lose meaning because their outcome is either predetermined or based in the unsatisfactory system of social discourse.

Now, Amber with a system that gained uncertainty in the form of other players throwing challenges in your way (á la Polaris or Shock:), with reliably challenging opposition (á la Dogs in the Vineyard), and a guaranteed objectivity of the GM (any functional RPG)... that's a different thing. It could still be "diceless" insofar as there are no random elements, but you get what dice give you without meaningless randomness polluting an environment where it's irrelevant.

[lumpley]

J, I think you're ... well, I think you're hung up on "what randomness gives us," as though it were one thing or a list of things. In fact, what randomness gives us in a well-designed game depends precisely on its place in the overall design.

In Dogs, for instance, what the dice give us is context for the pressure to escalate. Take away the randomness just there, and you short circuit the game's escalation of conflict.

But you can't generalize to "in RPGs, randomness gives us context for the pressure to escalate." Most games don't even care about the pressure to escalate! Some games, like Polaris, use non-randomness to create context for the pressure to escalate, because they organize escalation differently than Dogs does.

"In RPGs, randomness gives us suspense," "In RPGs, randomness gives us inspiration," "In RPGs, randomness gives us irrelevant pollution" - those all seem just as nonsensical to me. Maybe in some individual RPGs randomness does those things, but we'd have to look at the RPGs case by case to really know.

In RPGs, randomness (when present) serves a particular function within an overall game design, widely variable from design to design.

If there are trends in what function randomness serves, families of functionality across game designs, then the place to start is with the designs not with the randomness. Don't say "what functions can randomness serve?" say "here's a game; in it, what function does randomness serve? Here's another..."

-Vincent

[Valamir]

That's pretty much what I consider to be a fundamental truth of game design, Vincent.

Randomizers are just a design element like any other design element.  One could well ask the same question "what does a character sheet do for us", "what a DM screen do for us", "what does having to give our characters a name do for us"...etc.

If you have a design goal you can identify any of a number of paths to get you to that goal.  Those paths consist of a package of techiques assembled together in some fashion, one of which may well be some randomizing element.  If that randomizing element helps make that path work, and that path fulfills a design goal, then the randomizer is a good thing and what purpose it serves depends on how its used in that design.  If the randomizing element doesn't do these things...then its just not good design.

That's why so much of this thread boils down to what random elements CAN do, but no amount of discussion will derive what they DO do outside of the context of a specific application.

[Joshua A.C. Newman]

OK, so, great, what function does randomness serve in:

• Dogs in the Vineyard

• Trollbabe

• Shock: (I ask because I want your input, not because I don't have an answer)

• Prime Time Adventures

• The Mountain Witch

• D20

• GURPS

... and, since Ralph just joined the conversation, why are games like the following so unusual in their lack of randomness:

• Universalis

• Amber

[lumpley]

I'm only comfortable talking about a couple of those games in these terms. You'll have to ask the experts!

I've already told you about Dogs.

In Primetime Adventures, the cards serve two functions I see. First, they make it so that the person whose decisions we abide by isn't always the person whose character gets her way - without randomness just there, you could base your investment in any given conflict on your foreknowlege of who would be making the decisions about the details of the outcome.

Second, they regulate the flow of resources from the producer's hand through the players' back to the producer's. This has important consequences in the pacing of the session, I think; without randomness just there, you could base your investment in any given conflict on your foreknowledge of how much episode is left.

Matt, correct me if I'm missing something or wrong.

Universalis is massively diced.

Add Capes, Breaking the Ice and My Life with Master to the list!

-Vincent

[Joshua A.C. Newman]

Ah, vis-á-vis Universalis, that wasn't what I meant. That just pooped out of my fingers because Ralph posted while I was writing that. I'll get back to you when I remember what I was thinking about.

Since I've read, but haven't played, Universalis, I'd like to, at some point to see how it works in action. Someone asked me once why there were dice when player control was so central and I don't think I can answer that until I play it.

[lumpley]

In Universalis, the resources you have at your personal disposal are in constant high flux. Thus, the decision whether to invest heavily in this conflict or squirrel your coins away for a more important future conflict isn't too substantial - you can't count on that kind of prioritization to make it so that the person with the most resources won't always win. Randomness serves that purpose instead.

Correct me if I'm missing something or wrong too, Ralph! It's been too long since I played the game.

-Vincent

[Emily Care]

As I see it, in Trollbabe randomness is used to make you weigh out your choices for success by putting your own safety & that of your resources & characters' connections into risk.  If it wasn't random, you'd be choosing to torch your kit & kin with a known consequence to it or yourself, or none. 

In the Mountain Witch, the randomness of conflict resolution does several things: 1) makes the narration switch hands, giving a wider variety of input from the players, 2) weights the input of others in your conflict via their Trust and 3) (this gets at the inspiration thingee, too, J) creates a range of outcomes that give people a direction to point their narration, ie the degree of success creates a smaller pool of outcomes that you get to choose from among when deciding what happens.  That, I think, is actually a very important role of randomizers. 

And, of course, Ron & Tim know best what they intended and how it works out in play.

In BtI, the dice are there to make an incentive for you to narrate various things, while creating tension about the outcome.  The dice are also used specifically to make the players collaborate by how they are awarded, what dice are available & how they are stacked against you.  You could probably just as easily narrate all the bits for free, then flip a coin to see what happens, but you would probably get less investment in the final outcome because it hadn't been added to by the players in each succeeding bit.

This may or may not be a side issue, but something that nar rp eliminates--along with most randomness--is the ability for anything to be decided about one's character without the player's consent.  Randomness clusters around that in most tabletop rpg. 

yours,
Em

[Josh Roby]

OK, so, great, what function does randomness serve in... • GURPS

First off, as an iterative artifact, GURPS does not use dice in a consistent way.  You can't simply say "GURPS uses dice to mirror the unpredictability of the real world" or whatever.  Dice have been added to the design of GURPS for a number of reasons through the years of its development, often for very different reasons.

GURPS uses dice to determine:

• character success or failure in task resolution -- binary result
• degree of character success in combat (ie, damage) -- numerical result
• character ability to overcome character flaws and fulfill player-initiated actions -- binary result
• character ability to overcome physical damage -- binary result
• consequences of critical successes or critical failures -- selective result (item from a list)
• magnitude of success in magic, usually in terms of how many of X the mage creates -- numerical result
• whether a character can add an Ally, Patron, Dependant or other NPC into the action via advantages of the same name -- binary result
• whether an unreliable feature or flaw manifests itself (a car's finnicky starter, an uncontrolled and fickle superpower) -- binary result
• the results of hitting random buttons on alien technology -- selective result
• details of the setting, assuming the GM is using tables to generate such (GURPS Space gives tables on creating solar systems, frex) -- selective result

Overall, I would say that GURPS uses dice to support the game part of roleplaying game -- it adds an element of chance that theoretically makes the experience at the table more enjoyable for the participants.  It is building off of solid wargaming roots in this sense.  I would not say that GURPS dice support any specific flavor of play -- it simply does not have the intentional design foundation to support such an application.

[Josh Roby]

Crap, I forgot (because I never used):

• results of a failed Fear Check -- selective result

[Valamir]

The use for randomness in Uni serves a couple of specific design purposes.

First they prevent Complications from devolving into pure auctions.  While I love auction mechanics and enjoy many auction based board games, there are certain behavior patterns prevelent with auction mechanics that I didn't want to see in Uni Complications.  So instead of a pure auction where high "bidder" wins the right to narrate, the bids translate into dice and the dice decide.  This makes players far more likely to "bid" up what they consider a reasonable amount and be satisfied with their total even if they're just close to their opponents rather than perpetually try.

Also, philosophically, Uni's design was based very much on an Adam Smith-esque economic model.  Complications are investments and in order to get investor like behavior from the players the return on your investment has to be predictable yet uncertain. For instance, its reasonable to say that over the long term stocks return 10% on average.  Over the short term, however, its anybody's guess.  Similarly in Uni I can say with some statistical certainty that over the long term Complications return 25% on average (i.e. a 10 die Complication should net you 12.5 Coins on average), but for any single Complication you could come out ahead or behind.

The payoff motivates players to start Complications which is a desired behavior...initiating Complications is a profitable activity.  But the uncertainty means that the player's behavior around Complications is much different than if that payoff were a guarenteed, known, non randomized event.