Explicit and Emergent Rules

[TonyLB]

As I design, I increasingly encounter the choice between making rules that explicitly say "The game will be played in such and so a manner" and building a mathematical system that has emergent properties that I hope will lead to the type of play.

For instance:  I wanted to make a game where players often took two important facets of their personality (Love and Justice, for instance) and made statements about when they were willing to sacrifice one in pursuit of the other.

In my first iteration I simply said "You have these stats, and you get bonuses to your roll when you reduce one and raise another."  Folks had a real hard time getting their heads around it, and it intruded on play as often as it created neat moments.  That strikes me as a fairly explicit rule that (perhaps coincidentally) didn't work real well.

Second iteration I'm giving people the same numbers, but making the rules more emergent.  They can raise a number by being effective when it's "at stake" (objectively defined, but beyond the scope of this discussion).  They can increase their effectiveness by reducing other numbers.  I think this drives the same phenomenon (reducing one number in order to increase another).

I like the power of emergent rules to encourage players to create a pattern of behavior, rather than having it imposed on them.  For instance, in old-style D&D, neither clerics nor elves carried axes.  But the players of elves were happy with their long-swords, whereas the players of clerics tended (IME) to whine about their maces.  I think that making a choice fosters a sense of involvement in the structure of the game.  The structure is one that the players have created, even if the mechanics led them there.  

Today, right now, I'm inclined to believe that if you can get a behavior as an emergent property of a simple mathematical system then it will always be better than an explicit rule.  But that is almost certainly just because the concept is shiny and new in my mind.  It would really help my learning curve if somebody could disabuse me of that notion with clear examples of where explicit rulings are superior to emergent ones.

[Andrew Morris]

In my first iteration I simply said "You have these stats, and you get bonuses to your roll when you reduce one and raise another."  

...

Second iteration I'm giving people the same numbers, but making the rules more emergent.  They can raise a number by being effective when it's "at stake" (objectively defined, but beyond the scope of this discussion).  They can increase their effectiveness by reducing other numbers.

I'm not following your examples here, Tony. They sound the same, except that the second allows you to raise a stat when it's "at stake." Could you clear that up for me?

[timfire]

I increasingly encounter the choice between making rules that explicitly say "The game will be played in such and so a manner" and building a mathematical system that has emergent properties that I hope will lead to the type of play.

Could you elaborate a bit more on this? Honestly, I don't see how either of your examples differ, they both seem 'emergent' to me.

If I may use different termology, it sounds like you're discussing front-loading vs. developing-during-play.

I also wanted to say that you can never force anyone to play a game a certain way. People will always play the game the way they want to. The only thing you can do as a designer is write rules that encourage a certain type of play. It sounds obvious, but I believe it's actually kinda hard to implement effectively.

[edit: crossed posted with Andrew.]

[TonyLB]

First example:  You raise one stat and lower the other at the same time, in the same action.  That's the rules.

Second example:  To raise a stat you need to succeed in an overall situation.  You can achieve this with or without lowering any other stats, but lowering the stats increases your effectiveness, so it's very likely to be useful.  But trying to increase the stat and lowering other stats are two wholly separate actions.

[Andrew Morris]

Okay, that clears it up a bit for me. If I'm understanding correctly, the difference is that in the first example, the raising of one stat is directly and immediately tied to the lowering of the second, and vice versa, while in the second example, you can raise or lower stats independently.

Personally, the first is easier to wrap my brain around. I think I'd have more confusion using the second example in play. But then, I'd need to see all the rules to get a better picture of this in context.

As an aside, the second example leaves me scratching my head. To become more effective, I lower my stat? Does that mean I become less effective when I raise my stat? Again, this would probably be clearer in context, but at least with the first example.

[Shreyas Sampat]

Emergent rules have a few properties that you need to think carefully about when you're using them; they're a sort of double-edged sword.

[*]You only get the emergent rule when all the rules that imply it are being applied correctly. In this world of gamers who either don't pay attention or like making house rules, this is not entirely reliable.
[*]People will not always notice the emergent effect.
[*]People who do notice it might think you didn't mean to put it there.
[*]I had another but my brain froze. More later.[/list:u]

[Valamir]

Today, right now, I'm inclined to believe that if you can get a behavior as an emergent property of a simple mathematical system then it will always be better than an explicit rule. But that is almost certainly just because the concept is shiny and new in my mind. It would really help my learning curve if somebody could disabuse me of that notion with clear examples of where explicit rulings are superior to emergent ones.

I'm inclined to agree with you as a general principle.  Simply ruling that X goes up when Y goes down is in many ways not as effective as setting up a situation where there is an advantage to the player for choosing to do this.  So the situation of X going up when Y goes down emerges organically from choices the player made due to built in advantages in the reward and currency system.

However, I would STRONGLY encourage you (either in the text or designer notes) to highlight, explain, and draw attention to this interrelation, point out that it is intentional, and that it is an important part of what the game is meant to be about.  Make that explicit, even if it seems you're hitting them over the head with a +3 Hammer of Obviousness.

This is, in fact, at the root of my beef with the Sorcerer rules.  Sorcerer is very nearly 100% emergent.  If you sit down and play it as written...play of the sort Ron envisioned...just happens.  The interplay of Need and Price and Humanity and Ritual and Kicker and descriptor, and Relationship Map all...just work...really well.

But there is almost no supporting text to suggest that this is the point, to draw attention to how the parts all work together...there is no +3 Hammer of Obviousness to be found at all.  And so the number of people who find the game opaque, difficult to understand, etc. etc. are IMO much higher than it needs to be.


So my advice is, yes, make the rules emergent.  But then do a good job of explaining how all the gears mesh together.

[Callan S.]

As I design, I increasingly encounter the choice between making rules that explicitly say "The game will be played in such and so a manner" and building a mathematical system that has emergent properties that I hope will lead to the type of play.

This sounds like something you might call 'Fractal design', since your using a simple formula to create a complex result. It occurs to me that as much as it can have great results, it can be really hard to design the formula just right.

Anyway, hard or not I think such methods are probably going to end up being the best way of compressing RPG's in size. Err, excuse my drift. :)