How does one achieve balance in a game between the characters and their opposition? Here are a few of my rambling thoughts on the subject, which I've been thinking about heavily as I go through playtesting of various games.

First of all, this isn't a question that everyone considers relevant. In particular, the people that I work with on my games are far less concerned than myself on this issue. For me, if I am going to play against an opponent (and roll dice with them), I generally want a certain type of experience in play. I want the outcomes to be close enough that neither victory nor defeat is entirely certain, so that when dice are thrown their results are relevant. (Incidentally, just writing down that statement unfolded an entire twinkling world of alternate possibilities, which I will have to track down later; this discussion is very much "in the box" of traditional gaming conventions). The dice create tension; you roll and hope, you cheer or cry out in pain. But dice can only provide this experience if the results matter; ie if there is a subjectively substantial chance that both failure and success are on the table.

This is not to say that all opposition needs to be equal or balanced. Mooks, extras, thugs, and what-have-you are appropriate for some genres. Similarly, some opposition figures will be obviously superior. The mechanics of interacting with them are different, however, and vary wildly in purpose and depth.

But supposing that you are trying to design a systematic method for having a conflict with subjectively meaningful dice rolls affecting the outcome, how should such a thing be designed?

The simplest method (in design, not execution) is perfect symmetry. Player characters and their opposition share the same rules and the same creation techniques. I use this frequently for balanced fights in Exalted. And it works, more or less, but player creation systems serve poorly for creating opposing forces when the PC chargen takes more than a trivial amount of time (as it does in Exalted). The time required becomes prohibitive. Also, the outcome is not guaranteed to be "balanced", since players create characters that have many more concerns than combat-effectiveness, whereas the prospective opponent need only allocate resources toward those attributes that will be relevant in the conflict.

One could opt for a streamlined process for creating opponents, while still having essentially the same characteristics as the players (parity in structure, but not process). This is the case in Anima Prime, a game I enjoy partly because it creates a good feel of balance in fights and has a simple system for creating opposition. In DnD 4ed, the process is simplified even further; monsters are primarily selected rather than created. This works, but it begins to raise the question of "how does one determine how difficult a given conflict will be?" If the resulting opposing force is of equal size and relatively similar relevant stats, then one can be reasonably assured of a balanced conflict, but scaling can be tricky. DnD handles this by giving experience values for each monster as well as levels, tables that determine how fights should be scaled to party size, and guidelines for monster party composition.

But sometimes opposition is partially or entirely asymmetrical. It possesses different attributes, or uses them in different ways. In that case one must resort to mathematical guesstimates (or statistical analysis) to determine what will make for a challenging conflict (supplemented by testing to prove out results).

But how can you be sure that dice rolls are subjectively meaningful? This is where things get tricky. In most games, the odds of success should be on the side of the player. I've still got a game on the back burner that is designed to work the other way around. But how often should players "fail" if conflict is to be reasonably interesting? Should it be half the time, or one in ten?

Naturally it can't be simplified to a simple ratio. A pyrrhic victory can be analogous to defeat; for example, winning your first conflict in a given day of DnD adventuring after using most of your healing surges and daily powers isn't what most players would consider an unqualified success. In games of resource management, like DnD and Exalted, success is often measured by the ratio of resources as compared to the outcome achieved.

Also, defeat means different things in different conflicts. Defeat in exalted can mean just getting beat up; in DnD it means a TPK or full retreat, but in many story games it might be as simple as losing narration. The measure of acceptable and interesting chance of failure must also take into account the cost of failure.

Clearly I don't have an answer on the subject, but these are my thoughts and the issues I'm gnawing at. How have you folks solved these problems in your designs?

-Drake

This is David again, hence the atrocious picture.  Here are some thoughts on the relationship between game design and coding, and an idea for an ambitious game design method.

When I dabble in the story-games forums (a practice akin to drilling a tree to find the number of rings), I see more and more small games and games tailored to more and more specific situations. This comment is unremarkable; the posters there are perfectly aware of this.

What slowly presents itself is the idea that a game is a series of reusable mechanics. I've come at game design thus far with the idea that I should make a "new" thing, although it's acceptable to take heavy inspiration from other systems and occasionally use certain common tropes (like "characteristics" of a character that have numerical values). It only hit me today that I'm designing in a fashion that is very very similar to how I code; I do a search for similar solutions to the problem, then adapt one of them to my needs. Searching is a bit easier with code (google is merciful), but I don't mind the social aspect of searching for game mechanics; just ask Lukas, Andy, or Ben how it's done, or post on livejournal for a wider net.

This isn't how everyone codes, of course. Some people are happy taking entire modules or libraries off of CPAN and using someone else's code with abandon. I don't use this method for the same reason that I don't borrow mechanics without futzing them: the delicate issue of ownership. Well, not exactly; at work I don't have admin privileges so installing module is a ridiculous pain. In design, you have to wonder whether you will have enough ownership of the final product to sell it. In addition, I don't know if there is a big library of free game mechanics anywhere; I suspect it probably exists. Design Patterns of Successful Role-Playing Games, by Whitson John Kirk III (sorry for the messy link), is a start, though the ideal would have to be a wiki.

And if it does exist, it could be turned into a game. What if the first session of any game was a combination of character generation and system generation? "I want to play a highly skilled gunslinger, and I would like a technically detailed system to model my fights". "I want to be in a downward spiral as I get closer to the truth; I want to insert a conflicted mechanic that both rewards and punishes knowledge".  Ideally, to be truly and utterly beautiful, this "game" would have a website to help  you procedurally define what the game is, and at the end would create a tight little pdf of the rules, complete with links to have it published and sent to your front door for play (although I see the rules in such a game as being malleable, so a print copy might become obsolete before it arrived).

But one would first need a database of mechanics to work with.  Assuming that all the mechanics could be linked in some sort of intelligible fashion, I wonder if you couldn't do quick automated tests of the system, before starting your alpha... doing things like looking for buffer overruns that powergamers might exploit, or excessive processing time where the play gets boggy.  Sadly I don't have the coding skills to reach for so lofty a goal today.  But maybe someday...

I don't want to talk about labs yet, so instead I'll throw around maps.

Three years ago I was teaching Math, which means that four years ago I was back in school doing postgraduate (not masters-level, *sniff*) work to earn a certificate to teach Math and English at a secondary level in the state of Washington. Those classes were some kind of dull. That's where I learned to knit! But I also learned about how people learn to do things. Part of that is having models of the activity that they wish to perform.

So to jump to gaming: how do people learn to run a game?

Everyone really starts with a model, a "map" for how to do it. Maybe you saw your big brother do it once. Maybe you had someone else act as your first GM. Maybe you pieced it together just by reading the rules and considering how they work together. And no one has just a single model that they work off of... we're also bringing in our ideas of what makes a good story by looking to models in literature, cinema, graphic novels, etc. There are models to be found in the imaginary friend play of childhood, games of "let's pretend", and games of cooperative sports and competitive sports. All of these things together form an idea of what a game should look like, and what it means to run a game.

Having good models of the end result is a start, but I think it's something that could be supported better in most gamebooks. Spirit of the Century is one of the best examples of this; they give models galore in the back of that book which are appropriate to the genre. They also give many different kinds of models, which is even better, to support stories of different complexity and prep time. I almost forgot that I ran that system for a little while... it's too bad I'm not that hip to pulp these days. When I think back on it, that was pretty good.

But you need more than a good model of the end result. You need a sort of map of how to get there.

In this most game books fail. There is a basic assumption that the process of creating a story that is suitable for roleplaying is an intuitive process if you know the end result that you want. I'm making broad and unfair generalizations, so let me quality: this is my impression without having performed an exhaustive analysis. And the process is somewhat intuitive, because we have lots of models already of the end result; we know what it needs to roughly look like in the end (I'm skirting the issue of player interaction and how it fouls up and enhances things because it's not the focus of this rant). But one way that the process of creating sessions can become better (easier, more productive, more reliable, higher quality) is by exposing these models, so that we can correct for faulty intuition and understand the elements that are in play.

When you write an essay, most of us have been trained in a process that might look something like this: you pick a topic sentence, brainstorm for ideas that support it, organize them, write a rough draft, revise it one or more times, edit it, publish it. You learn the process not because it's set in stone, because it's the only way to write an essay; it's a map for how to get from "I need to write an essay" to turning in an essay. Once you understand the model, it doesn't keep you from trying different methods, skipping steps, or what-have-you, but it gives you a certain space to think about the process from an objective perspective. It's a map of the territory between start and finish in essay writing. Most of all, it helps the user overcome the paralysis of infinite freedom. (As an aside, maps can also constrain creative thought, but I honestly think that's more of a concern for people who have mastered the process than for people who are having trouble learning to do it effectively in the first place).

But what kind of maps are there for making a game session? What sort of models are there of the end result?

In that vein, I wish that every roleplaying book I bought had a sheet in the back called a "session guide" or something like that; a character sheet for a session. At the top you could write the name of the session, your name, the date, etc. It would have places for entering the names and descriptions of places players might go, people they might meet, challenges they'd encounter. There'd be a place to write the beginning and a place to write the ending(s), and any rewards the players might receive. Anything that externalized the process of thinking about what a session is. And there should be worksheets for making sessions, not because I think anyone would always use them, but just so that people have help in thinking about how to build things.

Note: Most of this is written from the perspective of "how one writes prep games when you are the only person prepping", the whole superplayer perspective. More thoughts on multiple player input later... after playtesting.

Double note: this is a spawn of the thread started on my personal lj here: yurodivuie .

5 - "Codex John Harper"
4 - "Snakes in my Garage"
3 - "Kiss a Dude: A game of homophobic paranoia by D. Vincent Baker"
2 - "Bronson or What"
1 - "Dogs in the Vineyard"

( I cannot tell you how happy i am to be the scribe of this legend. )

a tale of Storming the Wizard's Tower

This is a scan of the actual character sheet that made StWT so rad for me. Note the excellent 'gear' list.

It happened...again. I think, as time goes on, my ability to actually run a good game is slowly being overshadowed by the constant fear of how consistently i can ( botch it. )

A tale of Storming the Wizard's Tower.

Having failed by Three days to live up to our New Years Resolution i offer you a tardy essay on the nature of roles in media and those invented by the sub-culture of Role Playing Games ( and Gamers. )

In what is quite possibly the worst game i've ever run, ever in my life, David and I sat down to play a game of Storming the Wizard's Tower that resulted in a field that stretched longer than Rainbow Road merged with Galactus. ( And don't think i didn't see every lap of it. )

a tale of Storming the Wizards Tower

In an attempt to further explain myself, much the same way DRDrake's Sandstone will explain himself, i'm going to tell you the story ( Of the One Good Game I Ever Ran. )

Hi, this is Drake/yurodivuie  again. I'm spending a lot my spare time working on another game idea, chewing it over and over in my head. I've dragged amnesiack  and graypawn  (more or less willingly, respectively), into it, and the results have been incredibly satisfying.  At the moment it's only called project Sandstone, a game built rather selfishly around the idea that I would like for there to exist a game that I can play given a half-hour's notice that's a mashup of the Magical Land of Yeld, Bliss Stage, Burning Wheel, Valkyria Chronicles, Final Fantasy III, Fallout III, Haibane Renmei, Lord of the Rings, and a Miyazaki Film.  Of course, anyone that's really wanted to play a game knows that the only way you play what you want the most is to run the game, and that's double true if you write it... at least from what I've heard... so it may be a self-defeating enterprise.  I understand now why jake_richmond  and benlehman  have both designed games that require shared ownership of the traditional game leadership role (Magical Land of Yeld and Polaris, respectively, again, although they may have others that qualify.  Do buy them.)

Working with another creative person and friend to design a game is, thus far, a thousand times better than trying to do it on my own.  Frankly, I don't know how it gets done properly any other way.

I'm incredibly excited, but I have nothing to show... yet!  Hopefully later this month I can start putting out excerpts.  First up: setting.

I've spent one year in this place, immersed in the Seattle-driven "Indie Scene" for gaming. And i've gone a bit skitzo in the way i design games, as a result. But after some time in thought i began to see the common threads between what i'm learning now, and what i'd been taught ( before. )

( Lucky Pip 4.0 )

Authored by Mac and graypawn

Someone i know and respect recently asked an interesting question about Gaming. This led me to pondering, which lead me to reflection. And in reviewing my history, i had to laugh. ( Again. )

Dear Readers!

Behold: A new Year arrises! And we, the madmen here at Calculated Error are beginning with it a quest! Our goal: To write one Anecdote, one Essay, or publish one Lucky Adventure per month!

It's a small goal, yeah. So sue me. NO, ..wait...don't do that. We are very poor.

( Details: )

Horton Crow asks by Txt message:

"What happens if the Fortune says Remember the date, something good will happen to you in one week. Does that mean Lucky Pip is tortured for a week?"

The answer is emphatically yes. Many of our players don't realize that Lucky Pip is near to invincible, and can withstand all but a Kryptonite tipped yellow bullet, dipped in silver and shot at the end of a wooden stake with a Force Push. So if you get the chance to place Pip in a precarious position that lasts several hours, days, or even epochs, do not fail to do so. Pip will easily survive to mete out his comrades incomparable advice.

(Please leave all of your qualifying questions about the significant details to The Adventures of Lucky Pip and the Incomparable Ithamar Conchie in a response to this entry, or send them via advanced web mails to: CalculatedError.games@gmail.com)

The Epic Level Awesome of the Un-Store is now powered by one more raw dream. The Adventures of Lucky Pip and the Incomparable Ithamar Conchie is now available on Mr. Vincent Baker's most recent gift to Gamers.

Behold! And be wary...