Plausibility, Realism and game design goals [an essay]

[Valamir]

Well, here you're citing that reality is different than what typical gamers find plausible.  That's exactly my point!!!  Reality can be different than what the players find plausible.  I find it a bit nonsensical that you say that the term "realistic" shouldn't be used, but also cite "realistic" firearms and how what they find plausible conflicts with "reality".  

Well, no.  Lets be more precise.  Its exactly that casual use of realism that I'm objecting to.  Its not that "reality" can be different than what the players find plausible, its that what experts in police firefight data find plausible is different from what gamers find plausible.

Its what's plausible to me vs. what's plausible to you.  As soon as you start ascribing "realism" to one of our plausibility sets you open up an entirely different can of worms.  Its that can of worms I want to avoid by being very specific that we're NOT actually talking about realism.

Realism is the 800lb gorilla that really gums up game design discussions and distracts from the real issues.  Instead of trying to make a game "more realistic" you should instead identify your target audience and the reason that that target audience would play your game.  Then you make your rules such that they are plausible for that group for the purpose their designed for.  This MAY mean you wind up writing the exact same rules that you would have if you just went for "realism" or it may mean that certain aspects of the rules don't seem as important as they once did.

I'm a bit stuck.  How can I express the distinction between the made-up Japan game and the real Japan game?  As I see it, they're both plausible, but they differ in realism.

Well, the difference is who they're plausible for.  Say you have 8 year old kids, college age fans of samurai flicks and anime, and graduate students in Japanese studies.  Who will find the game plausible.  Maybe the 8 year old kids find them both equally plausible because they don't know any better.  Maybe the college fans find the made-up game plausible because it matches their movie based knowledge but actually find some of the "real" game implausible because it conflicts with their modern day social sensibilities about how people behave.  Maybe the scholar finds both of them implausible but the "real" game is "more plausible" than the "made up" one.

But I strongly recommend calling niether of them "real" because thats the whole can of worms thing again.  Marco hits on a potentially good turn of phrase when he says C-13 was "informed" by his research.  

"Informed by research" seems vastly more accurate an expression than "realistic"

Look at your own reply to Nicolas.  Do you see how it was useful for you to use the term "realistic"?  That's exactly what I feel is useful in discussion.  Avoiding it will only hamper discussion, IMO.

That "realistic firearms" bit was supposed to have been in "quotes" as an example of silly game marketing text.  It was meant to be sardonic not useful.  Sorry if the lack of punctuation confused that.

[lev_lafayette]

I think a fundamental error here is to assume that there are polar two binary opposites based on a thesis "Realistic" and an antithesis "NOT realistic".

Of course, if one takes the view that word "realistic" equates with a 100.00% accurate model of reality then noone will ever be able to achieve it. By definition the model would have to be bigger than the world itself. I don't think anyone is suggesting this, so that straw man can be dispensed with.

Realistic does not mean reality itself. It is an adjective, "a realistic game", not a noun ("reality"). Game X can be more realistic than Game Y.

This does not necessarily mean that Game X is only more realistic than Game Y in certain areas as well. There is such a thing as adaptability or scope which, along with realism (a continuum, not an absolute) and playability (ditto) are also, imho, good game design goals.

Perhaps the essence of the debate then is not between realism and its negation, but between universalism and particularism. Perhaps one is claiming that a game that attempts universalism is axiomatically doomed to failure compared to one that attempts to narrow its scope to particular objectives ("what the game is for").

Again, I would take issue with this as well. But I'll wait until this point is clarified.

[M. J. Young]

I'm siding with Ralph concerning a very specific and narrow use of the words "realism" and "realistic" within the gaming world that I, too, find egregious. Let me approach it in a different way.

Back in the 1980 presidential election, Omni magazine commissioned a simulation. It seems that one think tank had a computer program that modeled the economy fairly well, and Omni asked them to run 1) a projection of the economy over the next four years if Carter's current policies continued into the future and 2) a projection of the economy over the next four years if Reagan's program was enacted in full. It was run. It showed that Reagan's projections were correct, that tax cuts would increase investment, employment, and government revenue. In fact, Reagan was elected, but not all of his plan was enacted; still, over the years that followed there was the projected shift in the economy. The model had correctly predicted what would happen. One could say that it very accurately modeled reality.

To do this, the model required the input of several hundred individual economic factors, each of which was tracked in its interaction with all the others. Using the computers available at that time, it probably took several days to run each projection. While we would say that was realistic, it was certainly far more detail than any of us could include in a role playing game, particularly for modeling something so minor in play as the overall shifts in the economies of nations. We would be happy with something a lot less detailed. That of course means something a lot less realistic. Our compromise is based, however, on a notion that there is a limit to what is playable, a reduced return on investment as it were, so accuracy beyond a certain point while not exactly irrelevant is not worth the added effort to achieve.

The problem is that this "certain point" beyond which the added effort is excessive is entirely subjective. Yet any statement that realism is the objective must inherently include the unspoken caveat that it is realism within the bounds of what we subjectively feel is a reasonable effort to achieve such realism.

And since that boundary is subjective, any statement that we are shooting for "realism" means nothing more than "I want to make it as realistic as I think it's worth spending the effort to achieve."

For some gamers, Fudge achieves that. It is sufficiently realistic to meet their needs, and they are not willing to invest any additional effort in learning or using rules or mechanics that would make the outcomes more realistic, because such additional investment is not worth the return. It is realistic enough as it is.

For some gamers, GURPS and Rolemaster are the models they prefer. They feel that these games are making a serious effort at creating realistic outcomes based on complex inputs, and that the additional investment of effort, the drag this places on game play, the cost of books and materials, and all the rest, are vindicated in the increased realism produced.

Yet there are those who claim these games are not realistic enough; they want something that achieves greater levels of realism. They're willing to invest more into getting that greater level of realism.

I doubt, though, that they're willing to enter eight hundred variables into a computer program and wait fifteen minutes for the program to process one attack. That's more complexity than it's worth--to almost anyone. It might be worth it to someone attempting to model whether a lone gunman in the book depository could have killed President Kennedy, or whether there had to be a second gunman; but it's not going to be worth it to find out whether Grungar the Barbarian manages to cleave his fifteenth orc in this combat.

Thus whenever anyone says that "realism" is the design goal, they mean "modeling reality as accurately as I personally feel is worth the effort", and nothing more than that.

The problem arises when they assert that any decision they make is correct if it more accurately models reality, without recognizing that any decision they make which reduces complexity but less accurately models reality is working from a conflicting principle. That conflicting principle is not recognized as such. The designer doesn't say, "I want it to be as realistic as I think is worth the effort." That indeed would be patently the same is "I want it to be what I want," a design agendum that is tautologous and essentially useless to anyone, perhaps including the designer. Yet that is what any designer is inherently saying when he says that "realism" is the objective of the design.

The questions that could reasonably be asked are how realistic do we want this, how much effort are we willing to put into obtaining more realistic results, and what is the purpose which this particular level of realism serves.

It is that last point that is missed. Since realism cannot be absolute, it is accepted that it is relative, tempered by the cost in complexity. To know what is the proper level of realism we need to know what objective that realism serves. Otherwise, the pursuit of realism is meaningless.

--M. J. Young

[Valamir]

Lev, no one is making realism into a binary condition.  This entire thread has included references to the fact that its a continuum.  And niether is it universalism vs particularism (or "focus" as we more commonly call it around here).  That topic shares some similarities but is not what this is about.

You say that realism and playability are good game design goals...and that is the fundamental untruth that I've been trying to shake you out of since your first post. THEY ARE NOT GOALS.  Period.  Fact.  NOT GOALS.

They are techniques you can use to achieve your design goals.  They are the MEANS to the end.  They are not and cannot be the end themselves because they have no end.  

You said yourself that realism is a continuum (I think that's a poor word to describe the concept, but I understand what you mean by it).  You also agree that you cannot ever get to 100% realism...another seemingly obvious truth.

Therefor it should be abundantly and obviously clear that you must stop somewhere along that continuum.  Somewhere you have to say...that's it.  The rules are now good enough.  But the question is "good enough for what?"

What told you to stop at that point in the continuum.  Why there and no further.  Why that far and not earlier?  Why there?

The answer should be because that's the point where your design goals were achieved.  That is the ONLY valid reason for picking one point on the continuum over another.  If you don't have those goals to tell you where to stop...then you don't have a good design.  Form follows function.  Before you can settle on a form...you've GOT to understand the function.

[lev_lafayette]

You say that realism and playability are good game design goals...and that is the fundamental untruth that I've been trying to shake you out of since your first post. THEY ARE NOT GOALS.  Period.  Fact.

.... Assertion.

Maybe you have different concept of design goals. Maybe they are "ends" to you, but what applies to you may not apply to me.

They are techniques you can use to achieve your design goals.  They are the MEANS to the end.  They are not and cannot be the end themselves because they have no end.  

A goal does not necessarily have to have an end. It too can be an orientation towards an ideal.

Playability is a goal. The rules are a technique used to achieve that.
Realism is a goal. The rules are a technique used to achieve that.
Roleplaying is a goal. The rules are a technique used to achieve that.

For an example which certain games have been justly criticised for;

Readibility is a goal. Style is a technique used to achieve it.

... and so on.

Therefor it should be abundantly and obviously clear that you must stop somewhere along that continuum.  Somewhere you have to say...that's it.  The rules are now good enough.  But the question is "good enough for what?"

An interesting, but moot question and very much individually subjective. Some people prefer more realism, some people prefer more playability. A gain in one however, that does not result in a loss in another, is an improvement. It makes the game more enjoyable for more people.

If you wish to remain stubborn over semantics, consider this an improvement in technique, rather than a goal. Heck, it doesn't matter to me. The effect is the same; a better game - and that's the important point, not whether it agrees to some theoretical ediface about what good design should be.

[lev_lafayette]

I hope this isn't off topic, but what if my goal is to educate people about the way that comics portrayed Japan?  Which game is more realistic?  I would say neither one, that they both seek to attain different goals.

An accurate potrayal of Japan from the literary medium comics is not realistic potrayal of Japan. Excuse the semantic quibbling.

[Nicolas Crost]

I would suggest that whenever you use the term "realism", especially in terms of researched material, what you really mean is "more plausible to people who have studied this extensively

I think Thomas nailed it here.
Because here is basically what you do to create a "realistic" game: You read studies and models published by a bunch of experts. You familiarize yourself with their way of seeing things, with their model of reality. You find out what is plausible to them! So your "realistic" game is basically a game that is "plausible by what some experts say". So if your design goal is to teach people about what historian X said about Japan, that is perfectly allright. But this is not "realism" since I bet you quite a few bucks that I can find some expert historian Y who disagrees strongly with the position of historian X...

Which brings me to the point of how plausibility is constructed. Looking at the psychology of attitudes and how attitudes are formed, (at least) two things may contribute to the feeling of plausibility in a player of a roleplaying game.
1. The actual content of the game. That is how plausible do the outcomes of the resolution of the game per se seem to me. Do they fit my model of reality or not?
2. The credibility of the sender. That is who is backing up the claims made in the game? How much and which type of research was done?

This means that the highest plausibility (what John and Marco might call "realism") will be achieved when the output of the game fits my model of reality and the game is backed up by credible experts, that is it does not only fit my own model but also the model of some credible experts.

How does this help us in game design?
Well, first, it might explain why games with a lot of research and a granular resolution system seem more "realistic" (read "plausible" here) to some people. Those two factors contribute to making the sender (the game text) more credible. Basically game texts with lots of detailed rules (best based on real research) seem like experts (they remind you of science class or something). As such people feel much more inclined to accept the outcomes of the (credible) game as plausible ("realistic").
And second it might help in designing games that teach people. When you want to teach someone, you have the following basis: They have model A of reality but you want them to adopt model B. The problem is that model B does not seem plausible to the guys, they already have model A! So what you have to do is you have to compensate this inherent implausibility of the content (model B) by increasing the plausibility of the sender. That is you have to stress that you have done a lot of research an the topic of the game and you probably have to include a lot of detail (setting information and/or resolution detail).

[Callan S.]

Quoting from the sim essay:

The GM problem, only partly solved by GM-only sections, is that it makes it very hard to write a coherent how-to explanation for scenario preparation and implementation. Putting this sort of information right out "in front of God and everybody" is counter-intuitive for some Simulationist-design authors, because it's getting behind the curtain at the metagame level. The experience of play, according to the basic goal, is supposed to minimize metagame, but preparation for play, from the GM's perspective, is necessarily metagame-heavy

This is a quote on session preparation, but I think it applies to designing too.

I think your going to have a hard time getting this message across, Ralph. Because most often these designers see immersion as the way to design. To set up this grand, meta game goal, they'll have to get out of immersion to do so. To them, this is like saying you have to stop designing, to design. It sounds crazy to them and I can even see a point where in the other thread it's tried by a poster, very briefly, and with a visible, distasteful shudder (which shocked me...I wash thinking "Dude, you should not be throwing such a wonderful, project defining moment away, like that")

I mean, if you want to immerse and explore so as to design, how is not immersing going to help you do that? That's not my confused question, it's theirs.

The fact is, the very act of thinking about these designs is (a type of) play and it's very fun. It's so rewarding that your accurate message is most likely just going to nose dive.

It's a very here and now reward. I think, IMO, you need to illustrate the here and now rewards of what your talking about. Even though what your saying is deadly accurate, it still doesn't beat the play I think these designers are experiencing. Your not going to get through unless you manage to compete with the game they are very much enjoying (unless you can somehow grab them by the throat).

[John Kim]

I think Thomas nailed it here.  Because here is basically what you do to create a "realistic" game: You read studies and models published by a bunch of experts. You familiarize yourself with their way of seeing things, with their model of reality. You find out what is plausible to them! So your "realistic" game is basically a game that is "plausible by what some experts say". So if your design goal is to teach people about what historian X said about Japan, that is perfectly allright. But this is not "realism" since I bet you quite a few bucks that I can find some expert historian Y who disagrees strongly with the position of historian X...

Er, so what?  Sure, experts in any field may differ on details or lesser-known subjects.  But the same thing is true in any field outside of RPGs.  For example, I built computer simulations for high-energy physics experiments.  Yet, you could find an 'expert' who objected to how my simulations were run.  Does this mean that they weren't realistic and I shouldn't use that term for them?  

In short, you're not just arguing against "realism" in RPGs -- you're arguing against use of the concept anywhere.  Yet I feel this is rather misplaced.  While certain philosophers can argue endlessly that there is no such thing as objective reality, in practice most of us buy into it -- and at least understand what someone else is talking about when they refer to it.  

Which brings me to the point of how plausibility is constructed. Looking at the psychology of attitudes and how attitudes are formed, (at least) two things may contribute to the feeling of plausibility in a player of a roleplaying game.
1. The actual content of the game. That is how plausible do the outcomes of the resolution of the game per se seem to me. Do they fit my model of reality or not?
2. The credibility of the sender. That is who is backing up the claims made in the game? How much and which type of research was done?

This means that the highest plausibility (what John and Marco might call "realism") will be achieved when the output of the game fits my model of reality and the game is backed up by credible experts, that is it does not only fit my own model but also the model of some credible experts.

Well, again, this fails to distinguish between simply saying what sounds good and actually attempting to match reality.  "Plausibility" is defined by how the players feel about the game.  It may be helped by lying about your credentials and research, or feeding players' misconceptions, or many other techniques.  However, there are people who want to do more than just achieve the feeling of plausibility.  i.e. I want to look at more than just how the player feels about the text at the time of reading it.  I want to look at how that affects the player's relationship with the rest of the world (i.e. reality).  

How does this help us in game design?
Well, first, it might explain why games with a lot of research and a granular resolution system seem more "realistic" (read "plausible" here) to some people. Those two factors contribute to making the sender (the game text) more credible. Basically game texts with lots of detailed rules (best based on real research) seem like experts (they remind you of science class or something). As such people feel much more inclined to accept the outcomes of the (credible) game as plausible ("realistic").
And second it might help in designing games that teach people. When you want to teach someone, you have the following basis: They have model A of reality but you want them to adopt model B. The problem is that model B does not seem plausible to the guys, they already have model A! So what you have to do is you have to compensate this inherent implausibility of the content (model B) by increasing the plausibility of the sender. That is you have to stress that you have done a lot of research an the topic of the game and you probably have to include a lot of detail (setting information and/or resolution detail).

Wow, I am so seriously opposed to this.  I consider it lousy teaching.  Good teaching should not stem from asserted authority and heavily shoveled detail.  It encourages people to think for themselves.  Further, realism in general benefits from simplicity, not detail.  The more detail you have, the more difficult it is to be consistent and the harder it is for players to see the underlying concepts.  You should emphasize the basics and encourage players to think for themselves.  

You can see this in the contrast between original Traveller, say, and some of its 80's imitators like Universe and Space Opera.  Traveller was published in little 32-page digest-sized books.  So, for example, the Scouts book had a relatively simple system which nicely explained the basics of planetary conditions -- which was superior in both clarity and realism to the later imitators.  

I'm not saying that Traveller was ideal, but it had some good qualities (this example among them).  Basically no games have been made with a real pedagogical focus, even though I think RPGs are in many ways ideal for that goal.  I think the problem is that later games bought into exactly what you suggest here -- shovelled detail as a method for plausibility.  And it probably works in the sense that it gives a false impression of realism.  But that's why I consider good teaching to be a different goal than plausibility.

[LordSmerf]

Er, so what?  Sure, experts in any field may differ on details or lesser-known subjects.  But the same thing is true in any field outside of RPGs.  For example, I built computer simulations for high-energy physics experiments.  Yet, you could find an 'expert' who objected to how my simulations were run.  Does this mean that they weren't realistic and I shouldn't use that term for them? <snip>

John, you have a point here, but I believe it's misdirected.  Physics, which can be considered a "natural" science, has an absolute that we can check it against.  I can duplicate your experiments which test things against the natural laws of the universe, which are observable.

You were the one that brought up history, but history is not a natural science.  Without a time machine or some other method to actually observe the events of a specific time we have to take on faith the reports we receive third party.  It turns out that this is perfectly fine, but it's important to acknowledge that this is what we're doing.

But, yeah, John you've nailed it with your point that we're arguing that "realism" isn't attainable in the real world in these subjects.  But if that's true, then as a matter of course you can not have "realism" regarding these subjects in an RPG.

Anyway, I think that we're dangerously close to hijacking the thread, so if anyone is interested in continuing a discussion of "the nature of reality" or some such I'd be willing to do so in a new thread.  At the moment I want to address this:

Playability is a goal. The rules are a technique used to achieve that.
Realism is a goal. The rules are a technique used to achieve that.
Roleplaying is a goal. The rules are a technique used to achieve that.

For an example which certain games have been justly criticised for;

Readibility is a goal. Style is a technique used to achieve it.

These statements are all terribly problematic...  "Playability" means what, exactly?  It's not a universal sliding scale.  Each person is going to have a different definition of it.  The same is true of "roleplaying".  The problem with stating these as goals as you have is that it implies that there is some sort of absolute right answer.

Let's open up "readability" to illustrate my point.  Some people I know prefer dense text.  Density is one of their primary diagnostics for "readability".  Some other people I know (like myself) prefer less dense text.  I can read dense text, by my preference, the more "readable" text, is not that dense.  To make a text "readable" to me, you must be making it less "readable" to someone else.  It may turn out that people prefer low density to high density in some insane ratio (say 100:1) and writing low density text has more people believing that something is "readable" than writing high density text, but you haven't achieved some sort of absolute sense of "readability".  You have only achieved "readability" for the majority (so to speak).

And finally, I want to address John's point regarding teaching.

It turns out that this is still a plausibility issue.  Let's say that you, using your understanding of physics, write a game with dead-on accurate physics in the rules.  Somehow you write rule so that they perfectly mirror reality as we observe it.  I read your game.  I have two possible responses (well, it's a continuum really, but...): 1) Accept that you are a more credible source than whatever source provided me with conflicting infomration, in which case I consider your rules to be plausible.  2) Consider some other source to have higher authority than you do (say my buddy Joe, who read a book once) and thus consider your position implausible.

It doesn't matter which one is backed up by research, or even observable reality.  If I don't consider you to be a credible source then you haven't actually taught me anything except (maybe) what your personal viewpoint is.

It's all about plausibility.

Thomas

[Callan S.]

I further my thoughts on the rewards involved here, Vs user perception.

I really get the impression that this urge for realism is a desire to learn. Now, you don't want to learn silly crap. You want to learn the real deal (indeed, it could help you in real life). I think the phrase "Informed by research" indicates how much the user wants a manual filled with the best research currently available. 'Realism' is a missplaced term for that.

However, there is a bit of a twist here. Imagine were back in the day where the world is thought to be flat. And this crazy guy who thinks it's round, brings out an RPG with that in it. If you want 'realism', do you buy his RPG?

The problem with wanting 'realism' is in the inherant denial of the need for trust. Desiring 'realism' is pretending that you don't need to trust any RL person (in their findings), and that realism can just be had, sans the involvement of any human (and their perception).

While 'informed by research' definately includes that idea of trust. Someone who seeks an RPG which is 'informed by research' can actually consider buying the crazy 'round world' RPG. Someone who seeks 'realism' is going to ignore that RPG, and in their delusion never even have a chance at what they really seek.

[John Kim]

The problem with wanting 'realism' is in the inherant denial of the need for trust. Desiring 'realism' is pretending that you don't need to trust any RL person (in their findings), and that realism can just be had, sans the involvement of any human (and their perception).

While 'informed by research' definately includes that idea of trust. Someone who seeks an RPG which is 'informed by research' can actually consider buying the crazy 'round world' RPG. Someone who seeks 'realism' is going to ignore that RPG, and in their delusion never even have a chance at what they really seek.

Somehow we have different understandings of the word "realism".  To me, it is a common everyday word.  If someone at my work comes up to me and says "I'd like a realistic estimate of how much this is going to cost" -- I don't turn around and blast him for being a deluded fool who imagines that he can get that estimate without trusting any real life person.  I assume he just means "informed by research".  

If it's just semantics over the word, I guess I'm fine with substituting "informed by research" if consensus is that "realism" means something preposterous.  I find it a little odd, but I can live with it.  Are we agreed on issues other than the implications of the word?

[Vaxalon]

I'm surprised noone has brought up the word "Versimilitude" yet.

[LordSmerf]

If it's just semantics over the word, I guess I'm fine with substituting "informed by research" if consensus is that "realism" means something preposterous.  I find it a little odd, but I can live with it.  Are we agreed on issues other than the implications of the word?

John,

I think I find myself in agreement with you as long as you and I are on the same page regarding the fact that just because you do extensive research and all the experts agree with your analysis you are not necessarily close to modelling "reality".  The reason I have problems with the term "realism" is that it implies that you are modelling objective reality.  Anyway, yes, in fact I think Callan may have something solid there about the desire to learn.

Now, I still disagree with Lev, mainly because he's stating that all other things being equal, a more "realistic" (i.e. more heavily researched) game is better.  I'm not sure that I buy that.

Thomas

[John Kim]

If it's just semantics over the word, I guess I'm fine with substituting "informed by research" if consensus is that "realism" means something preposterous.  I find it a little odd, but I can live with it.  Are we agreed on issues other than the implications of the word?

I think I find myself in agreement with you as long as you and I are on the same page regarding the fact that just because you do extensive research and all the experts agree with your analysis you are not necessarily close to modelling "reality".  The reason I have problems with the term "realism" is that it implies that you are modelling objective reality.

OK, well, we're not in agreement then.  If I create rules informed by research, then I am modeling reality.  

I suspect there's some kind of miscommunication here, though.  My impression is that you consider a "model" to be something by definition flawless.  i.e. Only God can actually model objective reality, because it apparently requires more than mere human knowledge.  In contrast, I consider a model to be a normal human endeavor.  A child may make a model as part of his elementary school science project; a businessman may make a model to improve his marketing; or a scientist may make a computerized model to study his subject.  All of these are models of objective reality.  They all will have flaws in them, and some may be more flawed than others.  

They may be have known and deliberate flaws -- i.e. the businessman's model may ignore parts of the population who aren't his customers.  The scientist's computerized model may have simplifications to allow it to run faster.  These is fundamentally the same thing as a role-playing model which differs from researched results for the purposes of playability.  

So when I say that some rules model objective reality, I don't mean that they are 100% perfect -- any more than a school science project or business model or scientific simulation is 100% perfect.  There is rather a sliding scale of how accurate the model is.