Topic: Reverse Engineering world from game mechanics
Started by: simon_hibbs
Started on: 6/3/2004
Board: RPG Theory
On 6/3/2004 at 12:17pm, simon_hibbs wrote:
Reverse Engineering world from game mechanics
This comment got me thinking about something I've considered before and discussed with my gaming friends, but not discussed here. This is a seperate issue to the topic te quote is taken from.
Noon wrote: While art that is made, but then systematically evaluated by a rules system (so as to have a solid system effect), will reward the artist to skew his art toward what that system wants. The artist will start to become scientific in their method, because thats what the rules reward, because of the ways rules would evalutate it. The system rewards a behaviour that isn't art...and 'system does matter' is perhaps so simple one might miss it.
Theoreticaly player characters in game worlds should be able to deduce many of the game mechanics that we as roleplayers use to model their activities. A good example is spell ranges. In RuneQuest many spells had very specific ranges and areas of effect, typicaly a range of 100m, or an area of effect of 10m radius. In theory, the player characters could experiment to determine these parameters very precisely, perhaps even deducing the metric system used in RQ as fundamental physical constants in the game world in a way that they aren't even in our world.
Suppose I run a agme in the same game world, but decide to use GURPS. All of a sudden the physical laws of the universe have changed. GURPS uses imperial measures, so now the same spell has a range of 300ft instead of 100m. Surely the player characters will eventualy notice?
Silly? Well I think so, but I've heard sensible people take the view that the game system realy does define reality for the characters in a game world in this way. my possition is that the game mechanics are merely a convenient model that we use for gaming, but that the moment you start analysing it in this way, you have to chuck the model and say that the reality for the characters is that these ranges and measures aren't as clear cut - we only assume they are clearly definable for our own gaming convenience.
So, what's your possition? Do game mecanics realy define the physical reality of your game world? Or rather, are they a convenient simplification that we use in gaming, but that are not meaningfuly deducible by the characters in the game world?
Do you approach this question diffeently in fantasy games set in imaginary worlds, form how you would cope with this question for games ostensibly set in the real world?
Best regards,
Simon Hibbs
On 6/3/2004 at 1:22pm, xiombarg wrote:
RE: Reverse Engineering world from game mechanics
I will note, in passing, that in Time of Thin Blood, a supplement for Vampire: the Masquerade, one of the NPCs, Doctor Netchurch, manages to scientifically deduce the existence of what in the rules are called Blood Points. Netchurch does a lot of this stuff -- one of his "research assistants" is featured in Ghouls: Fatal Addiction, which answers alot of "stupid PC questions" about ghouling (i.e. "What happens when you ghoul a pregnant woman?") in an in-character document.
So, it seems that some of the people at White Wolf feel that characters can figure out such underlying "rules of reality", as it were.
Also worth mentioning is the "Self Referantial" plot in Over the Edge, where the PCs can figure out that they're characters in a roleplaying game, and this can posisbly grant them special powers.
On 6/3/2004 at 1:39pm, Loki wrote:
RE: Reverse Engineering world from game mechanics
These are two extremely interesting questions: can player characters deduce system mechanics? and if I switch to another system, can the player characters deduce that change?
In my opinion, running a game where the player characters are concerned with deducing the underlying mechanics of their reality (ie the system) is something so weirdly self-referential that prior to xiombarg's post I'd never even heard of it actually happening in a game. It seems to me that if that were the point of the game, then such a deduction would be the Point of the Game--in other words, the game is designed around the pcs determining the rules of the game. This could manifest itself as some kind of "pcs are physicists" or "pcs finding god" game... or it could manifest itself as the kind of game where the pcs become aware that they are characters in our fiction--perhaps they begin a dialogue with the players themselves.
Let me throw this back at you then... if player characters can deduce the system mechanics, how would that change your game?
Certainly my players often run their d20 player characters as though they know all their capabilities to a tee, and are nearly omniscient when it comes to estimating their chances for success within the rules. If the player characters were to discuss ranges of spells IC, it really wouldn't change much as they already behave as though they have conducted many experiements to determine the exact ranges of their magical abilities. Meanwhile, d20 is an extremely tactical game, so perhaps in another system this would be different.
On 6/3/2004 at 1:54pm, iambenlehman wrote:
RE: Reverse Engineering world from game mechanics
I actually heard this exact justification used in a simulationist-stemming argument for gamism (!)
The argument was this -- in a game system where, say, using all the same "type" of weapon (a big axe for hacking, a little axe for throwing backup) is better than using multiple "types" (a big axe for hacking and a dagger for throwing/backup) that experienced warriors will quickly noticed that people who use all the same "type" of weapon tend to survive, and it will survive in military culture as a "legend."
I thought this was cute.
I think it depends on the type of system -- if the rules are overwhelmingly Authority rules (this happens in the gameworld) then yes, it makes some sense that the rules are, in fact, more-or-less physics. To me, at least. If the rules are mainly credibility rules (rules that tell you who gets to talk about the system) then, well, that doesn't make a lot of sense. For instance, it makes no sense for an Over The Bar character to realize that he only succeeds in doing things when some non-existant entity downs a glass... But measuring spell effects in RuneQuest does make a lot of sense, to me, at least.
It, ultimately, depends on the CA. If we're talking Sim/Actor stance, this is feasible, if a little strange. If we are talking any other combination, it does not matter in the slightest.
yrs--
--Ben
P.S. A friend of mine made a MSH character once whose only power was that he actually had a copy of his character sheet. He didn't dare to change it (he'd have to erase first, and was worried about the implications) but he had an exact sense of his capabilities at any given time. It was amusing.
On 6/3/2004 at 2:07pm, ethan_greer wrote:
RE: Reverse Engineering world from game mechanics
It's amusing to contemplate the subjective reality of fictional entities. I like the concept from a theoretical standpoint, but I agree with Ben that it's only relevant in-game when certain styles of play are employed.
On 6/3/2004 at 2:12pm, xiombarg wrote:
RE: Reverse Engineering world from game mechanics
ethan_greer wrote: It's amusing to contemplate the subjective reality of fictional entities. I like the concept from a theoretical standpoint, but I agree with Ben that it's only relevant in-game when certain styles of play are employed.
Yeah, I tend to agree as well. The thing is, in some of the examples Ben cites, it wouldn't be possible for the character to figure it out. From the perspective of the character, the Over the Bar system would look random, and there's no way for the the characters to figure out otherwise.
It's notable that despite its rhetoric, that Vampire is High Concept Simulationism, which is why Netchurch can do what he did, and that characters in Over the Edge figure out their "I am a RPG character" status not by experimenting with the "physics" of the game world, but because a plot allows them to get a copy of the Over the Edge rulebook and allows them to meet their creators, i.e. more or less by fiat.
On 6/3/2004 at 2:51pm, beingfrank wrote:
Re: Reverse Engineering world from game mechanics
simon_hibbs wrote: So, what's your possition? Do game mecanics realy define the physical reality of your game world? Or rather, are they a convenient simplification that we use in gaming, but that are not meaningfuly deducible by the characters in the game world?
Do you approach this question diffeently in fantasy games set in imaginary worlds, form how you would cope with this question for games ostensibly set in the real world?
Well, I think there's a couple of ways of looking at it, even once System Typing and stance are taken into consideration. One is that the mechanics define the physical reality of the game world. Another is that the mechanics describe the physical reality of the game world.
The former could be really whacky, and I can see me in a silly mood running away with the idea of playing a character who discovered that the fundamental rules of reality broke down to Just Because much faster than we would expect, or bringing in ideas from Permutation City and trying to talk the GM into changing the definitions of the physical reality if the PC was able to come up with a more elegant and sensible theory than the game mechanics. But that's mostly me being silly and messing with a GM. I'll do that after too much caffeine or sugar.
Otherwise the answer is always 'as is convenient and suits my/our aims.' Though I argue this with people I play with who feel that the rules should always be deducible and I still haven't convinced them to my way of thinking.
Forge Reference Links:
Topic 10856
On 6/3/2004 at 4:14pm, HMT wrote:
RE: Re: Reverse Engineering world from game mechanics
simon_hibbs wrote: ...Theoreticaly player characters in game worlds should be able to deduce many of the game mechanics that we as roleplayers use to model their activities. A good example is spell ranges. In RuneQuest many spells had very specific ranges and areas of effect, typicaly a range of 100m, or an area of effect of 10m radius. In theory, the player characters could experiment to determine these parameters very precisely, perhaps even deducing the metric system used in RQ as fundamental physical constants in the game world in a way that they aren't even in our world.
Suppose I run a agme in the same game world, but decide to use GURPS. All of a sudden the physical laws of the universe have changed. GURPS uses imperial measures, so now the same spell has a range of 300ft instead of 100m. Surely the player characters will eventualy notice?...
If I may be overly literal for a moment, many cultures do not have this notion of the scientific method. Absent a somewhat refined scientific technique, it might be difficult to tell the difference between a range of 300ft and a range of 100m. However, surely the characters would know the spell could not reach someone 200m away. As a child, I had a sense of how far I could throw a rock. If spells are as reliable as that sort of physical task, spellcasters will have a good sense of their limits.
On 6/3/2004 at 7:33pm, Doctor Xero wrote:
RE: Reverse Engineering world from game mechanics
Well, one game-mastering technique I've used to give players a feel for the "uniqueness" of magic or superpowers has been to use one game mechanics system for the characters' ordinary lives and a rather different game mechanics system for magic or for superpowers.
For example, in one game I ran, the game mechanics for regular skills was one of those "1d10 + skill + relevant attribute to surpass the task difficulty number" systems ; however, the game mechanics for superpowers was one of those "roll relevant attribute or less on 1d20" systems. This even applied to senses -- players found they wanted to roll a high number when using normal sight to look for something yet wanted to roll a low number when using super-senses. It gave a marvelous sense of the difference between the human world and the superhuman world in that game.
Another time, I ran a game in which activities in the Astral Realm worked off different ruling attributes than did activities in the ordinary world (in a system wherein one could temporarily "increase" an attribute if one knew which attribute to increase). Characters made a concerted effort to discover which attribute ruled which effort, such as realizing that concentration mattered more than muscle memory in combat (an in-game way of realizing that Brawling now worked off Willpower not Strength).
Doctor Xero
On 6/4/2004 at 1:20am, Noon wrote:
RE: Reverse Engineering world from game mechanics
As the owner of the quote, I can positively say rules as physics has nothing to do with that quote.
I was refering to player behavior. For example, if the rule is that I get to hit Jim the player with a stick every time his PC goes near a cave of riches, his PC wont go there (lets say being hit is enough to stop Jim).
Notice that his PC has absolutely no game world disensentive to go to the cave. Indeed, it is one of ritches and he should be drawn there.
In a gamist CA, its the same deal...the player engages the system or misses out on a reward.
I'll argue its even the same in sim...the player engages the art evaluation system when making the art, or he misses out on exploration of what that system would otherwise provide. He does get to explore not exploring (which is valid), but I think the former is more rewarding and that in the end will tilt the balance.
On the actual topic of rules and game world physics being discovered, I'd like to discuss it. But I think it'd get incoherant to do so in this thread.
On 6/4/2004 at 4:15pm, simon_hibbs wrote:
RE: Reverse Engineering world from game mechanics
Noon wrote: As the owner of the quote, I can positively say rules as physics has nothing to do with that quote.
Oh quite, I hope I made that clear. The idea of characters altrering their behaviour based on the rules structure, which lets face it happens a lot in RPG, got me thinking about the wider issues.
On the general point, my feeling is that game rules are simply a model of the game world we use for convenience. In tabletop battles game we are not personaly realy personaly taking the role of a specific character, even a general. In Monopoly we aren't relay playign the part of a capitalist, the disconnection is there because the game is so clearly about trading cards and moving metal shoes. The theme of property trading is just a theme.
Roleplaying games are different in my mind because we are trying to actualy play a (fictional) person. They are a step beyond other kinds of games in this respect due to the personal identification between ourselves and the character we play. I'm not saying we become the character, but we identify with the character in a way that we geenraly don't in other kinds of games.
For example, only in roleplayign games is it possible to transfer a character from one game system to another in any meaningful way. For example I've translated characters from Traveller into GURPS because the referee switched game systems in mid campaign. For that to be possible, we must consider that the character exists independently of the game system, therefore the game system cannot be truly definitive of the character or the game world.
Simon Hibbs
On 6/4/2004 at 5:02pm, lumpley wrote:
RE: Reverse Engineering world from game mechanics
Simon, can I ask you to consider this:
Game rules coordinate three things: 1) the shared imaginary stuff, the imagined and described in-game; 2) the input of the real-world players; and 3) "representations" like numbers and words on a character sheet, dice, maps, life stones, whatever.
A game's rules draw on and manipulate the representations, whatever they are, to constrain the players' input into the imagined in-game. Neither the rules nor the representations can touch the in-game directly - how could they? Instead, they act on the real-world players. They contribute to the players' communicated imagination of what happens in the game.
So of course a particular character is portable from one game to another! The game rules don't have anything to do with the character. They don't determine what a character can do or how effectively a'tall - they determine how the players interact. Which then in turn determines what the characters do and what comes of it.
The players follow the rules, not the characters or the game world.
...So when I port my character from Ars Magica to the short-lived ill-advised godawful Pendragon hack I wrote, does she notice the changes in her capabilities? The question's nonsensical. Do I notice the changes in her capabilities? That question's nonsensical too: she's imaginary, she doesn't have capabilities. What I notice is: I'm allowed by my friends to say things about what she does that I didn't used to could say.
Do I pretend that she notices the difference? Well that's up to me, isn't it?
-Vincent
On 6/5/2004 at 4:25am, Noon wrote:
RE: Reverse Engineering world from game mechanics
Ah, now I know where I am.
Okay, Vincents dead on...characters aren't independant of rules, its the players who are. As said, rules are like a guide on how to imagine something/a world...rather exact guidelines.
That said though, most characters are imagined with some amount of curiosity. They investigate stuff, which includes the effects of game mechanics.
This means a bizarre loop, where the player, though following his own guidelines on what his character does (in this case, investigation of stuff), can end up investigating the effects of the rule system/shared method of imagining.
An extreme example might be a character who ends up falling off a 100 foot cliff and actually surviving okay, when the systems guide to shared imagining suggests such a thing is fatal.
Now, on to Vincents point. Does the player ignore this? Now, remember, although the character is imaginary, it does have some substance to it. Just like a program on your computer isn't an physical object, but it is a configuration and has the substance of such, so is a character in a human mind a configuration and has some substance.
Once you've established that it has some substance, you'll realise to get around the cliff problem, you'll have to abandon part of that substance. The part that would, as per its guideline on how the PC thinks, think about it. Your going to have to skip using those.
The problem with this is that it can be like removing cards from the middle of a house of cards. Other parts of the guideline to playing this character depend on them.
So the options are:
* Abandon and rewrite the guidelines somehow to avoid this (how?)
* Simplify the characters guideline so they gloss over such things, then find the PC is more shallow to play.
* Abandon the problem parts of the guideline and have the rest of the guideline slowly collapse as the interdependant other parts of the guideline show up.
* Make no changes, operate from the guidelines and find the character notices system effects and acts on them as per the guidelines.
Note: A clarification on what I mean by guideline is whatever the player decided for the character while making it. Usually its some personality details and then the bulk of the guide is extrapolation using common sense. The common sense part giving you more than you bargained for.
On 6/5/2004 at 8:51pm, Doctor Xero wrote:
RE: Reverse Engineering world from game mechanics
All right, after all this, I simply have to bring up three cross-overs that I think epitomize the wonder that can come from exploiting these differences.
In the recent JLA/Avengers cross-over, the authors had a great deal of fun having the Marvel Universe heroes end up in the DC Universe and puzzle over social and physical differences between universes, with the DC Universe heroes doing the same. Quicksilver was astonished by the Flash Museum and the fact that in the DC Universe, superheroes are celebrities (while Captain America, his mind addled by outside forces, wondered if this were enforced worship), while Superman discovered that, in the Marvel Universe, the area which would be Metropolis was only overgrown woodlands. The Scarlet Witch found her magical abilities going out of control because magic is so much more common in the DC Universe, while The Flash found his powers ebbing away in the Marvel Universe because it lacks the Speed Force which is the source of his superspeed. Et cetera.
In a wonderfully odd cross-over, Archie Meets the Punisher, the Punisher was astonished by the isolated innocence of Riverdale and vowed to protect it from the nastiness of the rest of the world, even deciding not to use firearms because it just didn't feel right to use gunfire in the world of Archie, while Archie and the gang were confused by the ill-fitting hoodlum who entered their idyllic polder (a place somehow protected from real world corruption, like Pleasantville or Themiscyra).
The oddest cross-over which included some level of recognition of variant laws of reality took place in a two-parter between the old Happy Days t.v. series and Laverne and Shirley. The first half involved Richie in the Happy Days series reality, with its wry verbal humor; in the second half, Richie ended up in the Laverne and Shirley series reality, and suddenly found himself continually the victim of slapstick of the sort which never occurs to him in his home series. What made this amusing as a cross-over is that Richie himself noticed that these sorts of things never happened to him except when he was around those two : "You live in a violent world, girls! "
I once ran a summer-long mega-cross-over celebrating my then-gaming group's anniversary, and one of the things we played off were the differing laws of reality between the many different campaigns.
(P.S. Don't forget Alan Moore's occasional cross-overs involving Supreme and Tom Strong, such as the funny animal universe in which Tom Strong discovered the Laws of Comedy trumped the Laws of Physics.)
Doctor Xero
On 6/6/2004 at 10:02pm, simon_hibbs wrote:
RE: Reverse Engineering world from game mechanics
lumpley wrote: So of course a particular character is portable from one game to another! The game rules don't have anything to do with the character. They don't determine what a character can do or how effectively a'tall - they determine how the players interact. Which then in turn determines what the characters do and what comes of it.
We disagree on some points of semantics, but I think we basicaly agree that we as players follow the rules, not the characters.
This is why I say that the characters exist independently of the rules - it's in the same way that Frodo exists independently of the English language. My wife has read the chinese verson of LotR and if all the english editions were burned, Frodo would still exist as a character in a book. The game rules are simply the language we use to describe our game world and it's characters.
Simon Hibbs
On 6/8/2004 at 12:36am, Doctor Xero wrote:
RE: Reverse Engineering world from game mechanics
simon_hibbs wrote: I think we basicaly agree that we as players follow the rules, not the characters.
I would add that that usually characters are like modern Americans and Europeans in that we all genuinely believe that we are following rules and work hard to discover what the game mechanics of our universes may be -- we call this work to discover our game mechanics Science (and our efforts to explore/evoke our responses to said game mechanics Art).
That's why I suggest that player-characters would notice differences if placed in a different rules system only if the differences between rules systems were intended to replicate differences between the Laws of Reality of two different universes. And even then, only the scientifically-inclined would notice unless the differences were radical (e.g. if a player-character in the Champions/Hero System universe suddenly discovered how far more lethal a fall from a great height is when she has plane-walked into the A-D-&-D universe or suddenly discovered that her maximum running speed had suddenly shiftered radically downward when she plane-walked into the Call of Cthulhu universe).
Doctor Xero
On 6/8/2004 at 2:48pm, simon_hibbs wrote:
RE: Reverse Engineering world from game mechanics
Doctor Xero wrote: That's why I suggest that player-characters would notice differences if placed in a different rules system only if the differences between rules systems were intended to replicate differences between the Laws of Reality of two different universes.
I think that's right, if your game rules say things about the game world that you want to be said, it's probably reasonable for the characters to behave apropriately. In a heroic game characters can do heroic things, and therefore you expect them to behave in a heroic manner.
I think the 'problem' can come when there's a marginal missmatch between what you'd realy expect to happen in the game world and the game rules, which after all aren't an exact simulation but can only ever be 'good enough'. Precise magical effects expressed in arbitrary real world units of measure are an example of this. Hex based movement in GURPS might be another. Won't the characters notice that their movement is always in 'quantum' units of so many metres per second? Well no, because I think we are only using these as a convenience for the purposes of game play. As you say they're rules for us, not for the characters. I as a player might say 'my character advances 5 hexes' while playing GURPS, but in a narative fiction account of the action you'd never see any mention of hexes or metres in this context.
I'd be interested in considering the difference between narativist and gamist game systems in this area, but time is pressing. I'll think on it and maybe post later.
Simon Hibbs
On 6/8/2004 at 3:43pm, lumpley wrote:
RE: Reverse Engineering world from game mechanics
Simon, I'm with you all the way, except: there's no real reason why your game rules have to be a simulation of the game world, "good enough" or a'tall. We've come to expect simulations because that's what we've mostly got, but that's sort of an accident of history, not essential to roleplaying.
Put another way: the problems you're identifying are problems only with games that purport to simulate the game world. Games that regulate the players' input more directly - take my Nighttime Animals game for instance - don't have any such potential mismatches.
-Vincent
On 6/8/2004 at 5:09pm, simon_hibbs wrote:
RE: Reverse Engineering world from game mechanics
lumpley wrote: Simon, I'm with you all the way, except: there's no real reason why your game rules have to be a simulation of the game world, "good enough" or a'tall. We've come to expect simulations because that's what we've mostly got, but that's sort of an accident of history, not essential to roleplaying.
This is why I sometimes used the term 'model' in my posts rather than simulation. The game rules are a mechanical means of generating results. Some games try to generate results that 'realisticaly' simulate game world activities. Other systems use game rules intended to produce results similar to what you get in various forms of fiction. These narativist mechanics simulate, or model narative fiction conventions and tropes rather than physical actions, but they are still simulations.
Note I never actualy used the phrase 'simulations of the game world' because that's not always what we're simulating.
Simon Hibbs
On 6/9/2004 at 1:33am, M. J. Young wrote:
RE: Reverse Engineering world from game mechanics
Doctor Xero wrote: That's why I suggest that player-characters would notice differences if placed in a different rules system only if the differences between rules systems were intended to replicate differences between the Laws of Reality of two different universes. And even then, only the scientifically-inclined would notice unless the differences were radical (e.g. if a player-character in the Champions/Hero System universe suddenly discovered how far more lethal a fall from a great height is when she has plane-walked into the A-D-&-D universe or suddenly discovered that her maximum running speed had suddenly shiftered radically downward when she plane-walked into the Call of Cthulhu universe).If I understand your caveats aright, I would agree.
Player characters do notice changes between universes in Multiverser, largely because the bias rules can make a significant impact not merely on the difficulty of doing certain things but on what is possible. Technological devices can go from easy to use to unreliable or even non-functional; magic similarly can be stronger, weaker, impossible. Although it's rare, even the ease with which characters walk can be impaired by very low bias in the body area.
So if you'll allow that significant changes in the difficulty of doing various tasks constitutes a difference in the laws of reality between two universes, I'd agree.
--M. J. Young
On 6/9/2004 at 7:17am, contracycle wrote:
RE: Reverse Engineering world from game mechanics
simon_hibbs wrote:
Note I never actualy used the phrase 'simulations of the game world' because that's not always what we're simulating.
Oh, I think there might be some use to that approach. I expect the systematic model to be as accurate a model of the game world as a real model of the real world would be. That is, I expect the model to abstract those things above or below its level of detail, but I also expect it to be a reliable guide to the actually possible within that scope.
I don't think that characters are likely to actually notice the measurement units of weapon ranges and the like; but I would expect them to be able to determine what the max range is - whether metric or Imperial - and clear undergrowth from a perimeter that wide, if they needed to.
On 6/9/2004 at 3:49pm, Doctor Xero wrote:
RE: Reverse Engineering world from game mechanics
M. J. Young wrote: Player characters do notice changes between universes in Multiverser, largely because the bias rules can make a significant impact not merely on the difficulty of doing certain things but on what is possible. Technological devices can go from easy to use to unreliable or even non-functional; magic similarly can be stronger, weaker, impossible. Although it's rare, even the ease with which characters walk can be impaired by very low bias in the body area.
So if you'll allow that significant changes in the difficulty of doing various tasks constitutes a difference in the laws of reality between two universes, I'd agree.
I've always had a certain fondness for the befuddlement of hero and villain alike in Last Action Hero when they cross over from their native reality to that film's version of the real world : " Think of villains, Jack. You want Dracula? I'll fetch him. We'll have a nightmare with Freddy Krueger, hold a surprise party for Hitler, Hannibal Lecter can do the catering, and then we'll have a christening for Rosemary's Baby! They're lining up to get here, and do you know why, Jack? Because in this world, the bad guys can win! "
^_^
Doctor Xero
On 6/18/2004 at 4:14pm, Dauntless wrote:
RE: Reverse Engineering world from game mechanics
I personally view system rules as an abstraction for what the game world reality is like. It is not game reality, but a model or representation of it. If it is not seen this way, then there really shouldn't be a need for having a GM. While I dislike rules-lite systems and like more detailed and complex rules, I also realize that rules should only be there to help the GM arbitrate things or to make thigns more consistent. But the final word is the GM's and not the system-rules.
If you think about it, we as real living breathing people in this world do the same thing. We try to find our limits and chances of success at things. So really, we are trying to deduce the nature and essence of the reality that we live in here in the real world. So I don't see it as a problem that characters would want to do the same in a game.
The problem is two-fold however. The first is that our real reality is vastly complex. There are so many unknown variables and so many hidden variables that we'll never know that it is impossible (probably) for us to ever figure out certain things with precise certainty. And this is where games don't have that benefit. Because game rules are simplified models it is easy to extrapolate from game events the precise system mechanics involved. And it's far easier to do this than to come up with Physics or Calculus to explain things in the here and now. There's no easy way to solve this problem, because no matter how complex we make the game rules, it will be orders of magnitude more simple than the rules of the real world we live in.
The 2nd problem is that because we are playing a game, everything is quantified. Because everything is quantified (numerically speaking) there is no mystery. A player knows exactly what his character's stats are, and he knows that Spell A can go exactly X amount of meters. The random element of games usually dictate success or failure, and not the limits of the system itself. For example, let's say I have a Strength of 10. Most games define this as being able to lift a certain amount of weight before being encumbered and doing a certain range or bonus to hand to hand damage. Few games see the number 10 not as static value, but rather an average probablity. So for example, on some days the character may feel stronger and lift more weight, and on other days he may be weaker and have less force behind his punch. The randomness is seen in rolling for damage, but not in the attribute itself.
Moreover, we in the realworld often only have vague ideas about how good we are at things. Strength is a bad example here, because it is easily quantified. In general most physical tasks can be figured out fairly well (we generally speaking now whether we have a lot of endurance compared to other people, or are more dextrous or agile....but we probably won't know...."yeah, "I'm in the 90th percentile of the most agile people in the world"). What about charisma, or attractiveness, or intelligence? These are all things which in the real world are dangerous for us to try to quantify. And yet in our games we know exactly how good we are compared to other people. And this makes it very easy for the players to deduce things about the world around them.
So a solution to this problem is to make many of the game attributes transparent and unknown to the player other than through qualitative modifiers. Through trial and error a player may figure things out, and realize that he gets better with time (through experience). But then again a character may not. It's entirely possible that a character has very high stats, but unfortunately manages to fail more often than he succeeds through bad luck. As a consequence, he'll become less confident in his abilities despite having in reality, good attributes.
When you hide the variables from the players, it becomes much harder for them to deduce how the game world works. But some system mechanics must be revealed to the players. For example, I use a Pool points in Focus and Drive that players can use to increase their chances. However, they don't know their limits necessarily. They may only have 2 points to spend and will say they want to spend 4. They will get all the bad stuff for spending 4 points, but will only get the positive benefits for spending 2. This in conjunction with not knowing the exact levels of their attributes or even their skills means that you don't run into that dreaded, "let's see, I have a Bow skill of 7, with an Dexterity of 10, and by spending 2 points of concentration and 10 Action Counts of aiming, I should be able to hit that man-sized target at 80m as long as I roll a -1 or better!".
By hiding the variables (or at the least obfuscating them) then it's like trying to fit a jigsaw puzzle together without a picture to help you. The underlying pieces are there, but you won't have any help. Another way to help avoid the absolute deduction problem is to varying variables or varying system mechanics. For example, my probability function changes according to the type of task. There are low variance tasks, average variance tasks, and high variance tasks. Some tasks are more dependent on luck and chance than on skill, and others are more reliant on skill and ability than on chance. If the player doesn't know which task type he's performing, then he can't readily calculate his odds of success (though trial and error will help).