Strategic rules

27 October, 2008 at 7:07 am (rpg theory) ()

I’ve been reading a collection of philosophical essays, titled “Tieto, totuus ja todellisuus”, for some months now (slowly but with certainty. A particular article by Jaakko Hintikka contained a bit of terminology I found useful. It is related to games.

Defining rules

In game theory, rules are what define a game. They tell what one can do within the bounds of the game.

For example, in chess: Turns, moving different pieces, winning, stalemate.

Strategic rules

The interesting part was strategic rules, which essentially tell what moves one should make in order to win (winning, in game theory, means maximising utility, and utility functions are something beyond the scope of rules; see, for example, playing against/with young kids, where you are likely to have different goals than you have in normal play). E.g. in chess, you generally don’t want your queen to be eaten.

Learning games

As a contrast to (most) roleplaying games, take a competitive game that has a winner. Assuming it is a good game, players will be making (strategic) choices, which will to some extent determine who wins the game. In my experience, it usually takes a bit of play to really understand these games, which is the same thing as learning some strategic rules. Simply playing the game may be sufficient, but maybe being taught by someone or reading books is more convenient or efficient. Be that as it may, once certain level of competency is achieved, then the intricate and interesting parts of the gameplay open.

Sometimes the learning process outlined above is interesting in and of itself, someties a nuisance. Personally, I only find gameplay meaningful after understanding what the game is about, in a sense.

This is far less true of roleplaying games (again in my experience); most of them are fundamentally the same game with different defined rules. There are two major exceptions: Intricate subsystems (combat and character creation are the most common) and the more divergent Forgey games.

The lesson here is that mechanical rules, in and of themselves, do not matter a whole lot. Maybe I roll 2d6 and add skill, or maybe I compare an attribute to value indicated by a table. The difference is minor, unless the way the game is played changes significantly. For example: If the way to solve problems is to have a character with suitable skill or spell, then the art of character building is important, but if the way the player approaches the problem is what determines the success of a given action, then wits and reading the GM’s/game desiger’s mind are more important, and the actual character played matters less.

In conclusion

Forge wisdom sayeth: System matters. The best way to investigate this claim is to play different systems and see if there is a difference. Here’s my refinement of the phrase above:

Take two games. Between these two, system matters to the extent that a different set of strategic rules is necessary for enjoying the different games.


The above has little to do with system as defined in Forge glossary (as it encompasses defining and strategic rules actually used in play) and even less to do with the content of the system does matter essay, which is focused on GNS and so on.


  1. the_blunderbuss said,

    I tend to agree with you, specially because we kind of come from a similar gameplay style (to give it a name), but I also think that the gameplay goal is highly important at regarding the relevance of system in the experience of play.

    Recently I was given a fairly interesting explanation of how my friends have approached D&D (and roleplaying in general), this might be useful to illustrate a point.

    (note to friends, this is an extrapolation of our conversation… errors are more than possible.

    My friends state that for it (an RPG) to be a game, there have to be winning and loosing conditions, even if they’re more abstract than your regular board game you can define certain conditions that you (as a player) want to accomplish and certain ones that you want to avoid. The game is set up (by a person) to provide fair (as in coherent within the game itself, not fair as in always fair to your character) challenges for the player to solve through wits and the use of strategy within the scope of the game.

    A lot of times (not all the times of course), this “scope of the game” can be related to combat. In this particular case, there are a number of very specific options (usually far more than in most boardgames, but not infinite nor anything that grandiose) that can be defined. The ability to “know what your options are” is usually a requirement for strategic play. I believe that this is what you’re saying when you talk about “knowing what the game is about.” This presents a strategic field of possibility, that is, a number of possible paths that are unevenly effective for the attainment of certain objective. The ability of the player to choose the most effective one is rewarded in game while the inability is punished.

    In this cases system is important. It might have to do with dices and stuff like that or it might not, but there is a need for a specific framework to choice, if the maximization of utility is the goal.

    PS: Been a long time Tommy! Loved your last posts.

  2. Tommi said,

    Hello Fred.

    The ability to “know what your options are” is usually a requirement for strategic play. I believe that this is what you’re saying when you talk about “knowing what the game is about.”

    Knowing the options means knowing the defining rules. It is certainly required (to some extent), but not sufficient. One must also know enough about the strategic rules to be able to make meaningful decisions.

    Game theory assumes that everyone is a rational actor who attempts to maximise their own utility. The analogy involving paths is good, as finite games can be represented as trees. Roleplaying is not a finite game in that sense.

  3. Callan said,

    Roleplaying is not a finite game? Or is that just one option?

    Another option is that you can make roleplay games under a finite tree structure.

  4. Tommi said,

    Hello Callan.

    The number of options a player (GM included) has at any given point is almost surely infinite. This is obviously true of all traditional games, and also true in the extremely formalised Forge games, because the fictional content is also relevant to the outcome of the game, simply because people respond to different content in different ways.

    It is possible to only consider the mechanical elements or to have some other means of lumping moves together, and with luck such classification is useful, but it still combines infinite options under a finite set of them, hence simplifying the problem.

    See also the use of game theory in economics, where problems with infinite options are simplified to a smaller set.

  5. Callan said,

    Well, if the responding person only has a set number of mechanical choices, then it can still be mapped as a finite tree.

    How the person responds to content might be infinite, but the options they have to respond with can, as an option, be fixed.

    I don’t think anything forces it to be an infinite tree.

  6. Tommi said,

    Assume system where you roll opposed combat checks when your character tries to kill another or disarm another. (I don’t know if such have been published, but I have played under a few.) Does this mean that trying to disarm or kill the opponent is essentially the same move? Maybe a living opponent could be used as a hostage or maybe the fight is a misjudgement or maybe the enemy is deadly even without weapons.

    Here there are two mechanically similar actions that have vastly different effect in the game.

    The principle at play here is that an rpg is a game where the fiction is part of the rules, as Brand put it.

  7. Callan said,

    Well no, because in the end the effect will again come down to choosing a mechanically supplied option. Hostage? Bluff/intimidation/diplomacy check. Fight was a missjudgement? NPC’s apply attack rolls. Deadly without weapons? Spot check, initiative roll, etc, etc.

    The imagined side of the game is infinite and yet at the same time the mechanical side can designed as being entirely finite. You might like to visualise it as the infinite being sandwiched between finite mechanical rules. Infinity in a box.

  8. Tommi said,

    The imagined side of the game is infinite and yet at the same time the mechanical side can designed as being entirely finite.

    (Emphasis mine.)

    A roleplaying game can be designed so. Not all (or most) roleplaying games are designed so. When discussing roleplaying games in abstract, I see no reason to limit myself to highly formalised ones. (Also, I find the more freeform games to also be more interesting, in general.)

    Well no, because in the end the effect will again come down to choosing a mechanically supplied option.

    I don’t think it is relevant that some rules might be eventually used. Let me give another example. Your character’s archnemesis is at your character’s mercy. Kill him or not? In most games this is not a mechanical choice; assume this is one of those games. In fact, let us say that the game in question only has rules for combat. Your choice will have significant consequences later on. Some of them may be about combat, some about diplomacy. It is not possibly to codify these as mechanics within the scope of this game.

    Or, to take even more explicit example, consider freeform play. Players make plenty of choices, they have effect, yet very little, if anything, is codified as mechanics. (The situation may be very game-like, even something with a clear victory condition, like who gets crowned a king.) I am also assuming this is freeform play where the choices people make actually have an effect, but given that such play is possible, this assumption is justified.

    Your note about boxing an infinity is very appropriate. To be more exact, rules of many roleplaying games are constructed such that there are mappings (or functions) from the fiction to the rules, and people are assumed to take a fictional situation, map it to a particular rules-level procedure, get a rules-level result, and then map it back to the fiction. (Universalis is an extreme example of an rpg that has rules which are not like this. For contrast, see, say, GURPS.)

    In case of freeform play the set of mechanics is empty. All the fiction is mapped to the empty set. The mechanics of freeform play simply say nothing about the fiction (as there are no mechanics that could say anything). Or consider a game with only combat rules. There is no sensible place to map non-combat situations to. This means that the fiction itself is engaged. And, as you seem to agree, the possibilities within the fiction are infinite.

  9. Callan said,

    “A roleplaying game can be designed so. Not all (or most) roleplaying games are designed so. When discussing roleplaying games in abstract, I see no reason to limit myself to highly formalised ones.”
    And I see no reason to limit myself to unformalised ones. :) Thus when it’s asserted that roleplay is not a finite game, I question it and suggest it’s a choice, rather roleplay games always being infinite when expressed as a tree.

  10. Bruce said,

    Interesting discussion and thought-provoking post to spark it off Tommi. I find it intriguing that though system can make a lot of difference, individuals are frequently employing exactly the same system to achieve totally different objectives.

    In most games like chess there are set victory conditions e.g. checkmate, whereas in roleplaying one player may be playing to experience an emotional connection with their fictional character while another may be playing because they enjoy the tactical challenges. Arguably your reasons for playing chess may also be different but what ‘victory’ is is much more clearly defined.

    I don’t know if this is true of you Callan but it seems to me that someone who enjoys the tactical and mechanical aspects of the game is more likely to view the game as offering finite options, whereas someone who is in it for the emotional experiences might feel there are infinite possibilities.

    Do you think there’s any truth to this?

  11. Tommi said,


    So, miscommunication. Okay. (In math, it is sometimes said “Integrable functions are not contunuous.”, when the actual content is “Not all integrable functions are continuous.”. Lazy language use, granted.)

    How about I make a stronger claim: Even in strongly formalised games there are significant components of the game that are missed if the fiction is ignored. See, for example, this discussion about DitV:, particularly points such as the following:

    Infact, Giving is the most thematically powerful tool you have. Giving is the way you acknowledge the players have struck at the emotional core of things.

    What this illustrates is that the fictional content, meaning your narration, is an element that clearly affects the mechanical side and outcome of conflicts, yet still the scope of narration is (for all practical purposes) infinite.

  12. Tommi said,

    Hello Bruce.

    Game theory is constructed with the concept of utility. Precisely speaking, utility function assigns a numeric value to each and every possible outcome of a game. Utility is assumed to be an accurate measure of what people want, and people are assumed to be rational in that they try to maximise their utility.

    So, someone might play chess with high utility for winning. Someone else might have high utlity for an exciting game, with winning barely a consideration. It is further possible for someone to have high utility for the other player winning (e.g. playing with a kid).

    (Game theory breaks down in reality because people are generally not fully aware of the full rules of the game in the sense assumed by the theory and further are not aware of their exact utilities, much less those of their opponents.)

  13. Callan said,

    Hi Bruce,

    As a mechanical object, I don’t think a game becomes infinite if you view it as infinite (or finite if you view it as finite). It is whatever it was to begin with. What a person views it as, in itself, isn’t worth noting.

    So whether there’s any truth to whether some people people seeing it as infinite and some people not, I don’t think it matters to begin with. There’s just applying some critical method for determining what it is, or deliberately avoiding applying a critical method.

    Hi Tommi,

    “Even in strongly formalised games there are significant components of the game that are missed if the fiction is ignored.”
    Well no, it’s not missed at all. The players aware there is this mechanical option called ‘Giving’ and they may decline to use the option. That’s like declining to move a pawn forward. (or they forgot the option, but that’s not about narrative, that’s just bad memory). It’s part of mechanical play. Not making a certain move is as much part of play as making another move instead – it’s not missing the move entirely.

    Now if your talking about seeing thematic weight in giving, that’s a different ball game.

    But I have to tell you, thematic weight doesn’t exist natively, at all, in giving. It’s like, say, a puppet of a dog. Yes, the pupeteer might make it act like a dog, and in watching you start to watch it for the signs of its dogs life. But it is not a dog – it is a puppet. It is just paper mache. The ‘give’ mechanic is just paper mache. The infinite narrative/SIS does not imbue it with any life or anything infinite. It’s just a finite mechanic.

    It’s just a puppet, waiting to be embued by a pupeteer and the audience with ‘life’ or ‘infinity’. But that infinity is not native/inherant/part of the mechanic at all.

    Without a pupeteer or audience granting it life, the give mechanic just sits there being very, very finite. Like a puppet dog with its strings cut.

  14. Tommi said,


    Giving is indeed a mechanical choice, but whether one gives or not is greatly influenced by narration; as some people in the linked thread mentioned, given a sufficiently emotionally powerful narration, they’ll give, even if the by the mechanics they could stay in the conflict or even win it.

    Hence the narration matters a great deal as it affects the factual outcome of conflicts.

    Likewise, take Capes. It is advantageous to push the buttons of other people, because that way they will oppose you, which will give you resources. Hence, some narration is simply more effective a move than narration other players care less about.

  15. Callan said,

    Whether you give IS greatly influenced by narration? Or CAN be greatly influenced?

    There’s nothing reaching out into the brain of the player and controling his synapses, of course. I would say it is merely ‘can’.

    Indeed, the default is not to be influenced by narration at all – giving or not giving are just two buttons, press one. That’s it. No further influence.

    Being influenced by narration, I would say, is an abberation. In a good way, but abberation none the less. It’s an abberation of how things really are – that there is zero narration influence on whether you give, by default.

    What I usually run into though is that gamers see zero narration influence as the abberation (abomination, usually), and being strongly influenced by the narration as the default of how things/reality really is.

    “as some people in the linked thread mentioned, given a sufficiently emotionally powerful narration, they’ll give, even if the by the mechanics they could stay in the conflict or even win it.”
    If they had two buttons in front of them, one marked ‘Give’ and one marked ‘Don’t give’, prove to me something physically stopped them from pressing ‘Don’t give’ and I’ll see the merit in your claim.

    Tommi, there’s nothing there. There’s only the listeners decision to be sympathetic to the narration. Which is a wonderful thing! But that doesn’t mean he’s greatly influenced by narration. It means he chose to be greatly influenced. Very different. It’s a decision by a listener – it is not how roleplay games actually work by default.

  16. Callan said,

    Oh, just had a quick example pop to mind.

    If you had a roleplay system where you replaced the buttons with abstract terms – like replace ‘give’ with ‘#532’

    Okay, run the person through it with abstract terms. THAT’S how a roleplay game works by default. No sympathy, no soul at all. THAT is mechanics in play.

    I think I might put this on my blog as well :)

  17. System matters, but system can’t demand sympathy « The CDS Weblog said,

    […] this discussion with Tommi over at Cogito, ergo ludo. Tommi […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: