2005-05-10 : How Do You Design a Mechanic?

Oh hey, just in case anybody's wondering: for the next couple of weeks I'm going to pull topics from the Game Chef comments instead of from the open house. I'll get back to the open house and some other outstanding topics after the competition.


In other words, how do you design a mechanic? As a model of reality (the keyword?). Based on the probability range you wish to generate? Based on the power it bestows on the player?

Never, ever as a model of reality...

...Unless it's the central, thematically charged reality your game is to comment on. Like, Sorcerer's demon rules are a model of the reality of abusive relationships. Dogs in the Vineyard's town creation rules are a model of the reality of communities in breakdown.

But, like, the details of your resolution mechanics? No reason for them to model reality, especially reality in a physical sense, no reason at all. In fact, designing your mechanics to model reality is a disaster, if that's all you're doing; it's only okay to model reality if your mechanics also thoughtfully structure the interactions of the players.

So! Design your mechanics to structure the interactions of the players (including the GM or GMs). This goes beyond "how much power."

Here's a very easy first step: whenever you write notes about a mechanic, in every stage of your planning, write what the people do. Don't write "sword attacks get +1," write "if you have your character attack with a sword, add 1 to the roll." That'll remind you that you're designing a thing for people to do, a process.

I'll post an example or three in the first comment.

Meanwhile, the point is: if you want to design a mechanic out of a keyword, think about the keyword in player terms, not in character or setting terms.

1. On 2005-05-10, Vincent said:


"Dawn," from last year's Game Chef. I was working on a game for that competition, didn't finish it, it was called BADASS the roleplaying game, subtitled "every man is an island." It paired dawn with dusk to represent how far along your character was.

Anyhow it had rules like these:

Choose a dawn/dusk number from 2 to 9, with 2 being toward dawn and 9 being toward dusk. Divvy a number of dice equal to your dawn/dusk among your skills. Give your "natural resiliance" a number of dice equal to 10 minus your dawn/dusk. If your character's an elf, you have to choose a dawn/dusk over 6, as you've lived thousands of years of murder and hate you undying Elric-looking bastard. If one of your human contact NPCs meets a bad fate, add 1 to your dawn/dusk.

So those rules happen to model reality, to some extent, with the badass character getting more world-weary and also investing in some human contact once in a while. But the real point of them, as you can see, is to structure my choices and actions as a player. And in the last rule, about "human contact NPCs," you can see how your choices and actions as a player may hook into mine - you can bet there'd be some rules about who gets to say what about the fates of which NPCs.

Or let's pull a good word from your answer to Andy, Tobias: "marriage." Let's apply it to Emily's cool future Brazil. And this time let's avoid modeling reality altogether:

You can declare that your PC's fate and another PC's fate are "married." The PCs don't have to have any particular relationship or even know one another at all. Once per scene, you can pull a die from the other player's pool and include it in your roll. If your roll's a win, return the die to the other player's pool; if it's a loss, throw the die away!

There's a rule that structures the interaction of the players without any real reference to the game world at all.

And for a third example, how about "marriage" in Em's Brazil again, but a) modeling reality and b) to the detriment of the inter-player interaction:

PCs who're married get +3 to their conflict resource total when their spouse is in danger. No rules about who can and can't be your spouse; that's left to the setting. Let's even go exotic, in keeping, and say that in the setting marriage is a flexible institution, with people marrying and un-marrying one another in all kinds of whimsical combinations.

What happens? Within minutes of meeting, all the PCs marry each other! It's a reasonable-seeming reality-modeling rule, but thoughtless: it has a bad (at least iffy and unintended) influence on the players' decisions.


2. On 2005-05-10, xenopulse said:

... and that's because players want to be effective? I assume that's not just a Gamism trait, it's that people want to be able to influence the story. And while +3 is simply an advantage and more effectiveness regarding the outcome of the story, Take A Die From Them is a choice with potential consequences, which is thematic.

Does that seem about right?

- Christian


3. On 2005-05-10, Ron Edwards said:


Christian, it's way more important than that. Maybe I can clarify by showing the counter-example, which unfortunately happens to be the norm.

It has always puzzled me (and lately disturbed me) that most RPG text examples use the *character's* name but refer to *real-game* actions.

"Tamara the Swordchick is battling a troll! They roll initiative, [details] and Tamara wins! She then rolls a 14 for her attack, which given her +2 sword and the troll's Defensive Target Number of 15, hits! Then the trolls rolls his Resistance against the damage ..."

See what I mean? Examples of game mechanics using the character names. The people are bizarrely absent.

I don't think this is mere convention. I certainly don't think it's "for convenience." I think it's strange, fucked-up, wrong, and so on. I think it's about hiding behind Incoherent play as a shield to protect one's fragile psyche. I think it's about pretending, in the sense of dishonesty. And I think it's about huddling together and rattling fetishes rather than actually socializing and sharing imaginations.



4. On 2005-05-10, xenopulse said:

Wow. I hadn't thought of it that way. But yeah, I need to remember that while the game is written as a text, it's played as a process among people—so the text should reflect that.


5. On 2005-05-11, Tobias said:


That idea about what 'marriage' does to characters - that's a nice one. Mucho ramifications. Especially if you add some 'upkeep cost' to marriage in those rules as well.


6. On 2005-05-12, anon. said:

I don't think this is mere convention. I certainly don't think it's "for convenience." I think it's strange, fucked-up, wrong, and so on. I think it's about hiding behind Incoherent play as a shield to protect one's fragile psyche. I think it's about pretending, in the sense of dishonesty. And I think it's about huddling together and rattling fetishes rather than actually socializing and sharing imaginations. And I think it's about age old conventions of identification with characters and nothing terribly kinky. Many roleplayers when talking about their characters shift between direct identification and third person reporting with no embarassment. I don't think gamers hide behind "incoherent play" so much as they've never heard of it. The model works, and if it doesn't fit with your view of gmaing (or mine even), why bash it?

I have visions of the roleplaying police be marching up to tables up and across the country, hitting gamers with rolled up copies of Sorcerer saying, "Stop this! It's inchoherent!"


7. On 2005-05-12, GB Steve said:

Sorry that was me. Missed out the name.


8. On 2005-05-12, Vincent said:

Steve: Clearly there's some kind of history between you and Ron that I don't know about. That's okay, but consider my bemusement.

As it happens I agree with Ron. He can answer for himself but I'll answer for me. I have a general point and a specific point.

My general point is: I consider myself a critic of our gamer culture. I think it's broken in important ways, and I think that we game designers have some responsibility for both how it's broken and how we're going to fix it.

The attitude of "The model works, and if it doesn't fit with your view of gmaing (or mine even), why bash it?" is incompatible with my interests in RPG theory and design. I am, at heart, in intent, and by will, a basher.

Which comes to my specific point: the model doesn't work, not for healthy play. The model works to perpetuate play that marginalizes us, isolates us, shames us and keeps us closeted - play that makes us fetishists.

This would make an excellent open house topic, if you're interested in really discussing it, by the way.


9. On 2005-05-12, GB Steve said:

My last point, whatever my history with Ron might be, was to be taken as a joke. I'm sorry if my flippancy was out of place.

But then I am genuinely surprised that what most people call roleplaying is seen as unhealthy and possibly even damaging, by other roleplayers. Sure, it's not the way I play but to marginalise them as fetishists seems a bit far fetched to me. I'm all for challenging the model and doing things differently myself but not because what they are doing is something inherently bad.

I want to do things differently through personal preference for a different style of gaming, to see how far the roleplaying idea can be pushed, to explore new boundaries.

I can see that one of the boundaries you might want to cross is mainstream acceptance but I wouldn't want that to come at the cost of marginalising happy and harmless gamers.


10. On 2005-05-12, Emily Care said:

Which comes to my specific point: the model doesn't work, not for healthy play. The model works to perpetuate play that marginalizes us, isolates us, shames us and keeps us closeted - play that makes us fetishists.
Oh ho ho, Vincent, you said it now.  Can't wait to see that discussion.

And [personal aside]: Ha!  For perhaps no reason, I feel vindicated.  : )


11. On 2005-05-12, Sydney Freedberg said:

As I read this—and Ron, Steve, Vincent, anyone slap me silly if I'm wrong—what Ron's objecting to is NOT players "shift[ing] between direct identification and third person reporting" but rather muddling, to mangle Vincent's terminology, the shared-imagined-stuff and the mechanical "cues" in the real world (e.g. dice).

I.e. if real person Bob is playing Tamara the 12th Level Swordchick,
1) in the actual space where the game is played, Tamara's not rolling a 14 to hit, Bob is. Tamara doesn't even exist at the gaming table.
2) in the imagined reality of the game, Bob's not slashing the troll with a sword, Tamara is. Bob doesn't even exist in the imagined world.

And I think this is the point where Ron says "duh!" and Steve says "so what?"

Well, compare Ron's hypothetical muddled narration above to these two examples, of narrating the same thing "pure" on each level:
1) "I need to roll a 12 or better... damn... okay... HA! 14! I rock!"
2) "Tamara charges the troll. She raises her broadsword and ... hits! Green ichor spurts over the cavern wall!"

Here we have a more lively description of the imagined stuff, as opposed to just slapping SIS labels on real-world activities ("Tamara gets at 14!"); but we also have, more subtly, a full acknowledgement of the real-world human player's emotional investment in the game.

(Now there's nothing to prevent you from doing (1), then (2), in sequence, as long as they're both done some justice.)

Ron, is this a reasonable gloss on your point, or just pointless hair-splitting?


12. On 2005-05-12, Sydney Freedberg said:

P.S. Whoops, massive crossposting.


13. On 2005-05-12, Neel said:

So, of all the games I know, the game whose text most comprehensively blurs the difference between the players and their characters is Nobilis (I mean, jesus, the GMing section is written as an in-character essay by an NPC). The idea that it was done that way to codependently shame and fetishize roleplaying is, honestly, risible.


14. On 2005-05-12, Vincent said:

Anybody thinks this is about identifying with your character is missing the point.

This is about willfully, self-deceptively denying the role of human creative input. This is about rules that thoughtlessly - or hell, maliciously - provoke stupid and unhealthy human interactions when you play them, then deny responsibility with "we just model reality."

I DO NOT CARE HOW THE RULES ARE PRESENTED IN THE TEXT. I care how the designer understood them and how the players use them. In many cases, how the designer presents them reflects how she understood them, but of course there are exceptions.

Neel, I suspect that Nobilis' rules do, in fact, thoughtfully structure the players' interactions, and that Ms. Borgstrom knew it all along.


15. On 2005-05-12, Neel said:

Shoot, I'm commenting on a total side-issue. Anyway, this bit:

Meanwhile, the point is: if you want to design a mechanic out of a keyword, think about the keyword in player terms, not in character or setting terms. totally vital. The question a game designer should be asking is: what are the decisions the player is making—what tradeoffs need they consider? This is what distinguishes a meaningful subsystem from tedious bookkeeping.

Concretely, writing a computer program involves lots of choice, but rolling on a Computer Programming skill doesn't require any. If you want hacking to be central to the game, you can't just require lots and lots of Computer Programming rolls; you need the player to be able to make decisions (which need not correspond to character decisions) that can affect the way things turn out.


16. On 2005-05-12, Michael S. Miller said:

Hi, Neel.

You're looking in the wrong end of the telescope. In Nobilis, the whole Hollyhock God = GM bit was done for artistic effect, to *acknowledge* that there are real people—people who have a spark of divinity within them—making these stories and build upon that fact to make better stories. (whether it succeeded in this goal is another matter) IIRC, doesn't the full-chapter example of play in Nobilis talk about players and GMs and give them names?

What's being discussed here is not blurring the line between player and character, but saying there is ONLY character and ignoring the character—"Pay no attention to the man behind the curtain." THAT is what is dysfunctional.


17. On 2005-05-12, Michael S. Miller said:

Cross-posted. At least I said what Vincent said, so the discussion kinda goes in order.

If you squint...


18. On 2005-05-12, Neel said:

What do you mean by cross-posted? Is there somewhere else this discussion is going on? If so, I'd prefer to continue there, because I'd rather talk about Vincent's actual post in this thread.


19. On 2005-05-12, Vincent said:

Crossposted like "crossed in the mail."

But yeah. Somebody (like Steve or you, Neel), post your concerns up in the open house. I'd prefer to talk about my actual post here too.


20. On 2005-05-12, xenopulse said:

"Simultaneously posted with another user" is what's meant by crossposting, not the frequent LJ usage of posting in several forums at once (which is a misuse of the original term, IMHO).

- Christian


21. On 2005-05-12, Neel said:

Wilco, Vincent.

More thoughts about what you said: the flip side of ensuring that the players get meaningful decisions is that when you have something in your game that isn't really about decision-making, you can often cut to the chase and get a better game.

An example of this shows up in combat systems. In most non-tactical games, the only number that can change are the hit points or equivalent. This means that each round the player basically has one binary decision—"Fight some more" or "Run away"—usually with a clearly superior choice. This is not a very interesting decision (and is parodied to hilarious effect in Progress Quest).

So, in my game Leftover Dudes I ended up with a different kind of system for action scenes. It was a Buffy-esque game by genre, so I wanted action scenes in which the PCs showed off how cool they were, but I didn't want to make them tactically deep.

So the way they worked was like this: the opposition shows up in game, and I set a difficulty level for thwarting them. Then, each player described in turn their actions in the usual stunty style of Feng Shui/Wushu/Exalted, and rolled dice and counted successes. Usually, the difficulty was much higher than they could achieve alone. However, they could use teamwork. Teamwork meant that if a player's description built off of and supported another player's description, they could take their successes and give them as bonus dice to the final roll. If the final total exceeded the difficulty, they succeeded, and otherwise, they failed.

So there weren't any really serious decisions to be made in an action sequence. The important thing was that the system extremely strongly rewarded taking other players' verbal contributions and building on them, and this was wildly successful. So, the problem with stunt mechanics is that no one can reliably improvise one impressive bit after another. It's hit or miss. But, it turns out that if you require players to build on each others' ideas, you can chain together five average ideas and get a stellar result in terms of player response.

Emily is probably nodding her head and going "Of course it works; you've got the players validating each others' contributions by actively incorporating them into their own stuff", but this was a major revelation to me. And I figured this out because I saw that there wasn't any decision-making going on in a particular subsystem, and then thinking about the purpose it really served.


22. On 2005-05-12, Vincent said:

Emily probably is nodding along, it's true.

Not me though - I'm like, dang, yeah! That's striking. I wish I'd thought of it.


23. On 2005-05-13, Em said:

But, it turns out that if you require players to build on each others' ideas, you can chain together five average ideas and get a stellar result in terms of player response.

That's a great insight, Neel. Thanks for your vote of confidence—I don't know if I'd have nailed it like you just did!

I remember having that selfsame problem when playing Feng Shui back in the day. It's hard to be clever all the time. But for some reason it's easy & fun to brainstorm with other people and come up with amazing stuff. It's that creative synergy thing. PtA has it in spades. Collab games thrive on the support this kind of dynamic gives to whoever's on the spot. Heck, what game wouldn't?


24. On 2005-05-13, Tobias said:

Neel - your example of cooperative combat descriptions mirrors what I wrote down on a piece of paper yesterday frighteningly. (And some old thoughts as well - for instance, giving orders to each other, which translates to a mechanical bonus - thus encouraging both a hierarchy in a 'party' as well as talk/cooperation during combat)).

If you could mail whatever system you have for Leftover Dudes (could find nothing on the internet other than a post on 20/20), I'd greatly appreciate it.

Some other thoughts about RPG's i'm thinking about designing (probably stupid to put this out just before the Chef thing, but if someone steals it well, I'll also be the better for it!).

The 'party' mentality - other than something to rail at, how about exploiting this concept? If roleplaying is meant to share stories with your friends and draw them closer, why not have a game in which a party also is drawn closer (through mechanics). I know Mountain Witch is hot in this regard as well.

Other things I've thought about:

-setting party opposition (potential) as equal (in d6) to the player power potential (in d6). This means an 'average encounter' will sometimes be easily met, sometimes suprisingly hard. When it's hard (opposition rolls higher than players as a group), players can support each other by giving one of their d6's away, which a fellow player can then use as 2d6. The only way people are allowed to do this - is if they describe how their character feels about the other charcter (in the current situation).

-giving a player a reward for both good and bad/weak aspects to his character by giving him some reward for both showcasing the good/bad sides. Success at the good thing means you character gets what he's attempting - a regular 'reward' for the player. A high result on the bad thing means the character does not get what he wants (but 'resolve' instead, which is like fallout)- which is also a reward to the player, considering that achievement without adversity, or light without shadow, is unexciting.

-instead of pre-scripted Screen presence like in PTA, instead I have a number of themes all *players* find interesting in a Bag (here I go again, bringing in the Bag concept). All characters have a (distributable during group chargen) differing stake in the these theme's (some may find it equally important (but differ in opinion), some may not care, etc. At the beginning of a story cycle, all these theme's are put into the bag. There should be about 5-7. Then, 2 themes are drawn - the upcoming 'scene' is about those 2 themes. The themes are not replaced until the bag is empty - ensuring that during a cycle of play, all theme's the players have put in are 'hit' (and each theme should be considered interesting by as many players as possible), and all characters have had the full mechanical impact of their stakes/attributes. You just don't know in advance how it's going to play out, but you're guaranteed some play you like.

As you can see, I try to steal from the best. :)

And my apologies for all the "()".


25. On 2005-05-14, Green said:

I've been (trying to) follow the conversation, and I'd like to share some real-world experience that may illustrate what I think is going on.  You may use it to interpret it however you like.  I only hope it's relevant.

About a year and a half ago, I Narrated a Matrix RPG using an earlier incarnation of Kathanaksaya.  Some of the most enjoyable scenes with that game came about through the players' individual contributions to the narration of events, supported mechanically through the Story Point concept.  A few examples of in-game things:

1.  Doing car stunts with a van while the rest of the crew shoots and the police chasing it.
2.  The entire crew taking on the albino twins and pooling resources to defeat them.  After one crew member (Arachne) simply pulled a "Dodge this," the rest of the crew members quickly overcame their shock and pooled their resources in order to kill the other albino twin.
3.  The players using Story Points to alter the environment in particularly Matrix-like ways.
4.  The Narrator pool mechanic helped to emulate the system inertia of the Matrix itself, making it easier to show why Agents were such a threat.
5.  I had an important NPC portrayed by Orlando Bloom.  He died.  This point has no relation to the other four.  I just wanted you to know I killed Orlando Bloom.


RSS feed: new comments to this thread