Game mechanics which attribute to govern firearms

[M. J. Young]

I am now moving away from a attribute + skill system.  and have it just an attribute system with skills used to modify rolls.  So to fire a weapon you would roll your precision attribute and it would be modified by a skill score.

O.K., I'm not getting it.

That is, I'm not certain what the functional difference between the two as described is.

Attribute+Skill: you add the two together, and get a number below which you have to roll to succeed.

Attribute with Skill modifier: you roll the die, subtract the skill, and see if you've rolled below the Attribute.

Or, to do a bit of algebraic analysis:
A+S=D
A=D-S
Same thing.
*****
To quote

I don't want to sound like I'm knocking your ideas, because they sound like a good start too me, but I think you're missing the point of some of the attribute + skill naysayers. At least some of them mean get rid of one or the other completely.

I'm the first to have mentioned the problem; let me restate that I do not mean that A+S is a bad design (or that D-S is a bad design, either). What I mean is that you need a reason to use both Attributes and Skills in your game system besides "That's how these things are done."

In most A+S systems, the problem comes up in character creation and character improvement. Let's say that we're using it to fire a gun. Our perception plus our gun skill is what gives us the chance to hit. Now, Perception 9 plus Gun 1 gives us a chance of 10; so does Perception 1 plus Gun 9. But if we have Perception 9, it's going to be used for unnumbered other things. So if we've got a choice between raising gun skill to 2 or raising perception to 10, which makes more sense? Obviously our ability with the gun improves the same amount either way; but if we raise our Perception to 10 and leave Gun at 1, we're going to impact a lot of situations in which guns don't matter. Most attribute+skill systems are plagued with this problem: it's better to have lousy skills and great attributes than the other way around. A guy with a gun skill of 10 and a perception of 1 isn't as good a guy with a perception of 12 and no gun skill at all.

How do you fix this?

First, it shouldn't necessarily be a matter of wanting to fix it. You should come at it from the other direction: do I have a reason to do it this way? If you do, then you want to think about how to make it work. If you want both attributes and skills in your system, here are a few fixes.

[*]Attributes are scaled below skills, and are the defaults for unskilled characters. That is, you never roll against "attribute+skill". If you have a skill, it ranges from six to twenty; if you don't have a skill, you default to an attribute, which ranges from one to five. This gives you the "use attribute in lieu of skill" catch-all without ever creating the situation in which an unskilled character is better than a skilled character. If you want attributes to impact skill in this situation, you could allow a character with a higher attribute to increase his skill faster, or start it higher.[*]Make attributes more expensive than skills. If for the same ten points you can buy one point of attribute or twenty of skills, the choice is a lot tougher. This creates a generalist/specialist situation in play, in which generalists are going to buy up attributes slowly so that they'll be adequate at everything and eventually be good at everything, while specialists are going to push individual skills high so that they can do those things very well quickly and incredibly well ultimately.[*]Give skill numbers more advantages. As long as Attribute+Skill determines chance of success and there's nothing else to determine, there's no reason for skill to matter as much as attribute. But if skill means you can do the job faster (e.g., more repetitions, faster fire rate, completed sooner), or if skill means better quality of performance (e.g., greater damage, better product, higher yield), and attribute doesn't impact these aspects at all, then having a high skill matters in a way that having a high attribute does not, and the value of skills as an investment makes them worth the cost.[*]Have skills, but have them function merely as statements of ability. Thus, you'll use your perception or hand/eye or whatever for shooting the gun, but if you don't have "shoot gun" on your paper, you can't shoot one at all. Be certain that the ability to acquire skills is reasonable--if it's not possible for a character to learn to fire a gun even when someone is showing him how to do it, you're going to have to be able to explain it.[/list:u]
There may be other fixes. The point is, I'm all in favor of attributes and skill existing in a game, but the traditional approach to using them is broken and you need to give some thought to how to do it right.

I'm going to blatantly recommend you look at Multiverser, and how it handles skills. It makes skill ability level matter in a lot of ways that attributes don't, and makes them easier to improve than attributes in several ways, thus avoiding much of the problems inherent in systems of that sort.

I hope this helps.

--M. J. Young

[ZeOtter]

I think one of the reasons why I have been stumbling through this is I have hang ups about how to accurately games model human learning.  I originally went to college to get a degree in teaching (which I didn't get), so I did pickup a lot on how people learn.

My idea is a skill tree where you have a general skill heading such as computers (I am a network admin, I picked something easy for me as an example).  Under computers when you first get the skill there will be sub-skills like; basic operation, e-mail, using the Internet, etc...  As you gain more levels in it you will have access to more sub-skills, to follow the computer example, you could get; networking, programming, hacking, etc...  Some sub skills could have pre-requisites like you can't get hacking until you have networking and programming.

The idea is that your skill (computer) will be your target number for your attribute roll.  Right now I want my attributes to be assigned as die types, like you have 2d10 strength, etc...  So if my "Intelligence" is 2d6 when you buy computer as a skill you start at lets say 8 you need to roll an 8 or lower to accomplish what you want.  But you can only use the skill if you have the sub skill the governs what you want to accomplish.  So when you first are learning how to use computers you can't hack into a secure system you have study for quite a bit of time before you have the knowledge to do that, and before you can learn how to hack you have to know how networks work, and how computer code is used to make the computer do what you want it to.

I have not "crunched" the numbers on this yet to see if the probabilities of the dice mechanics are reasonable but I am comfortable with the overall concept and think that once I play with the numbers I can get it to work in this form.

Learning a skill is a growing process that can take some people a life time to master, I don't think that a player with a high intelligence score, a computer skill of 1, and a lucky roll should be able to hack into Microsoft's network.  You need to first learn the basics of computers, turning it on, knowing the steps to take to create a word processing document, etc... Then you can start studying up on how networks are put together, how security works, etc... After that you want to understand programming and how to alter code to get software to do something is was not intended to do.  After you are proficient in all of that I would say you had a reasonable chance to hack a secure system.  Even after all of that your first couple of try's are going to be a lot of trail and error but you have a shot at it.

I was kicking around the idea of giving the sub skill a score from 1 to 3 to reflect experience, 1 being novice, 2 being proficient, and 3 being master.  To illustrate this idea following the hacking example, if you try to hack Microsoft your first time you have a dismal chance of success but you at least have a chance to know when you are in too deep and get out before you get caught.  With hacking 1 you are a novice that could probably get into someone else's personal computer at home and go through their hard drive when you find them on the Internet, but that is about it.  To get Hacking 1 you would need to have your pre-requisites (this case programming and networking) at certain levels, I would say one needs to be level 3 and the rest can be level 2.

The setting is the modern world where magic is discovered by science to be real, and the world is trying to come to grips with the fact that people can cause earthquakes, summon demons (of a sort), and fly without mechanical aid.  The game focus's on what will society do to come to grips with this fundamental change, and what else is out there they haven't discovered yet.  

I apologize for the long winded post, but the ideas just starting percolating out and I know better than to try and stop them once they start flowing.

I would love to hear more of your thoughts on this, your posts on this thread are what have gotten me this far with the system and I really appreciate your help.

[Ron Edwards]

Hi 'otter,

My question is: why should a role-playing game be in any way concerned with modelling how people learn?

If its topic is such learning (say, a drama in which an adult struggles to learn to read), then fine. But an adventure story? Even if it concerns the character growing up to be a warrior, which only a few do, the actual process of learning generally concerns ethics and maturity, not the moment-to-moment skill acquisition.

What I'm asking is, are you interested in creating/playing a game about shooting people or a game about learning how to shoot?

Best,
Ron

[Rob MacDougall]

The first edition of Paranoia had a skill tree something like what ZeOtter describes. They dropped it in the second edition, no doubt realizing that, of all games, Paranoia should be (as Ron says)  "a game about shooting people, rather than a game about learning how to shoot".

[JMendes]

Hey, all :)

To be fair, I don't think this kind of tree models 'how do people learn to shoot', but rather, 'ok, if you know how to shoot, what else can I assume you also know'.

Also, this whole discussion seems to be deeply rooted in the 'characters must improve over time' paradigm. Correct me if I'm wrong.

Cheers,

J.

[Mark Johnson]

Both theives a warriors need to be dexterous, but I wouldn't trust the first to win a fight, or the second to be able to steal anything. Yet most game systems will tell us that there is a correllation. What you're arguing against is stat/skill systems in general, or the extent to which one relies on the other. Or am I missing your point?

Suffice it to say that many games with stats(attributes)/skills would have been better served with a single system of attributes, skills or descriptors.  Often the split creates nasty currency issues or characters not in keeping with the point of the game.  

That said, stat/skill systems do have some things going for them.  First, often the, for lack of a better term, inefficient nature of such systems makes character creation fun in the min/maxing sense.  Second, there is a familiarity since at one time most games did things this way (even Theatrix did not escape it despite its obviously drama focus).  Third, it does make it easier to create standardized stat blocks for characters and monsters, sometimes reducing handling time.  And finally, any game that deals overtly with scaling issues (FUDGE) or different species may find it easier to handle these through attributes rather than trying to figure out how those issues affect skills on an individual basis.

Most of the games that I have loved have split stats and skills.  I just don't think it is necessary.  And a poorly instituted skill system is often times superior to a poorly instituted split stat/skill system because of the currency effects.

[Mark Johnson]

The setting is the modern world where magic is discovered by science to be real, and the world is trying to come to grips with the fact that people can cause earthquakes, summon demons (of a sort), and fly without mechanical aid.  The game focus’s on what will society do to come to grips with this fundamental change, and what else is out there they haven’t discovered yet.  

Fun!  Man vs Society.  A possibility for loads of intense dramatic conflict.  What do those with the power do?  Are they feared, hunted?  How do they survive?  What does it mean to be human and have this kind of power?  How does one uses one's power?  What does it mean to be different?  Awesome stuff.

I think that trying to capture this drama is probably more worth your while than a cool superhero setting.  Given that, you should probably pursue mechanics that capture the drama and conflict rather than focusing on learning curves, character improvement and complicated interactions between attributes and skills.

[ZeOtter]

I gave a very brief description of the setting because I strongly feel that mechanics of the system should enhance and give flavor to the setting of the game.  I really do not support generic systems, I have played a few of them and tried to give them several chances but I honestly can't stand them.  

While I am sure there are a lot of people who would disagree with me games like Deadlands, and In Nomine really worked for me because their mechanics supported the atmosphere they were trying to convey.  Deadlands with their poker chips and decks of cards (not the d20 remake), and In Nomine with their 666 dice mechanics I think really added something to the game, and really let the players immerse them selves in the setting of the game.

To answer Mark Johnson's post, the background for my game hasn't established how society has dealt with the fact that magic and the occult exist, I want to have the characters be a part of how society comes to grips with this paradigm shift.  I really want to explore the themes that come with drastic change in society as a whole, the fear, the hatred, the hope, etc...  

The setting is that science has discovered magic exists and how to use it.  The information has recently leaked to the public and who can figure out the "code" are starting to pop up all over the world.  The arcane practitioners of magic that have been around for centuries are more than  a little upset that all of their secrecy is for naught and now upstart wiz kids are using shortcuts to power that took them years to learn.  The worlds governments are exploring the possibilities of this "new science", people with no idea the power they are wielding are unleashing things on this Earth that should not be and social and religious groups are trying to come to grips with the ramifications of what magic means and how it effects their beliefs.

I really think to help immerse the players into the atmosphere of the game I need to give them a firm grounding in the mundane to really hammer my themes home.  The mechanics I have mentioned are my first attempt at this.  I don't want to make things overly complex for the people reading my game but I don't want to swing the other way like White Wolf did and end up with a system where skills are so generic as to make them almost meaning less.  In our society learning skills are a little more complex than that, and as we are becoming more, and more specialized in our roles.  I want the feel of everyday life to clash with the setting and form the basis of conflict in the game.

The other reason I pursuing the system of have posted is to emphasize more individuality between players.  Like in 7th Sea 2 players can have the same skill that has different sub skills or emphasis for what they know.

I am wondering if I should start a new thread for this discussion since it seems a few are confused that I have moved beyond the original topic of what governs shoot a gun to the basic structure of my character creation system.

[Solacus]

I would have to wonder whether an attribute would be the best way to go for firearms in particular. I am on a junior shooting team at the moment and can say that the majority of shooting is based on one's experience/training and "situational modifiers." Experience is the most important factor (i.e. skill rating) as an experienced shooter is so much better than a novice that there is little comparison. Also, minute differances in body position can make a fair differance, especially at a range.

Equipment comes into it as well, as a sling is always nice. Also, the weapon itself matters. Pistols are much less accurate than rifles (kind of common sense, but still) but differant models can also have their own accuracies within a group. A good shooter taking his time with a good rifle and good ammo can get shots into the same bullet hole with pistols or rifles.

The problem I have with assigning an attribute to firearms marksmanship is that none of them really fit. Strength is marginally applicable, but if you have a sling or hold it correctly, the weight of even a fairly heavy rifle will not be significant (and I have a relatively low muscle strength). Agility isn't a factor in non-combat shooting at least (I've never shot in combat, so I cannot comment on that. Maybe an ex-army person could talk about that). The more you move, the more you miss. Even rocking slightly on your feet in a standing position can throw a shot off by several inches at 50 feet.

Perception is arguable as well, skill is more important. My father is blind in one eye, and has to wear glasses to see properly at all in the other (a low perception stat if I ever heard of one). And yet, he is the best shooter I have personally seen. He can, with an accurized AR15, get bullets into the same hole with fair consistancy at 200 yards. He wasn't terribly in practice at the time either. When he was younger (still blind in one eye due to an early childer accident, but the other worked fine), he once won money on a bet to see whether he could pound a nail into a tree with a bullet at about 100 yards. He won that bet. Thats some of my personal experience with firearms, I hope it might help.

From a design perspective, if you are going to have attributes affect shooting, I would use Perception for Long Guns and Agility for Handguns/SMGs. I say this for two reasons. One, it makes more sense, as rifles are better used at longer ranges, while pistols aren't good beyond a fairly close range. Also, if it pleases you, you might want to seperate Long Guns from Small Arms. They require the same general skills but nevertheless, they dont overlap too much. I am a good rifle marksman, but I am horrible poor at pistols. I would keep them together if you do not put much emphasis on combat, but if combat is a significant factor in your game, you may want to seperate them. I hope I helped.

[Mike Holmes]

Ze,

So you want to provide a game in which players really feel like the system describes a world consistent with ours, right?

Well, look at Solacus' experience. In his example, he points out how skill is key, and attributes add only in certain less than optimum circumstances. What you could do, is to just go off of skill, and then, if the appropriate conditions occur, allow attributes to cancel penalties. So, strength makes it so that you don't acquire a firing penalty until fatigue sets in. Agility allows you to cancel some penalties for moving.

Do this on a case-by-case basis.

That would be very realistic, IMO.

Mike

[JMendes]

Ahoy, :)

I suddenly felt the need to point out my own experience with small arms.

I did basic officer training in the army (3 months), and that included very rudimentary shooting practice with a 9mm pistol and an automatic assault rifle. Amazingly enough, having never handled a firearm before, I was the second best shooter in the company. (A 'company' in the portuguese army is somewhere between 100 and 200 men. Since we were in an instruction unit, we were closer to the 200 mark, I forget how many exactly.) Now, some of the people in the company had alledgedly shot guns before. In fact, two of them were on loan from some portuguese-speaking african country (which one, again, I forget) and already had extensive military training. One of those two was the guy that shot better than me, the other one came in fifth.

My conclusion: some sort of attribute was definitely playing a part, here. And I'd venture it wasn't agility. At close to 200 lbs, I'm no epitome for cat-like grace. Thus, some sort of hand-eye coordination would be indicated. Note that I'm not exactly going against Solacus, here, as none of those guys had ever done any competitive shooting, that I know of.

Finally, to really complicate matters, I couldn't throw a rock in a straight line to save my life. I'm not half-bad with detailed mechanical work, though.

Cheers,

J.

[Bob McNamee]

I would say some level of "perception" has a bearing on skill in shooting, as well as a mental "relaxation focus" and "hand/eye co-ordination".
This may very well be an "attribute" working.

While perception isn't very trainable, the "relaxation focus" level and "hand/eye co-ordination" is.

It is common too for people to hear the words of the trainer, then not actually follow them when excited or stressed.

It is quite possible that you actually followed the trainer's instructions, where those who've shot before may not be or are following poor habits learned previously.

any firing range teachers out there?

[Mike Holmes]

And around and around we go.

See, everyone in real life has a different opinion on what sort of attribute backs what skill. And this is problematic because in the end all that attributes are are groupings of abilities.

That is to say that the only thing that can completely and accurately portray the ability to use firearms would be a rating of your abioity to use firearms.

Because in the end, for each individual it's going to be multiple and unique qualities that make that person good at what they're good at. Any attempt whatsoever to divide these underpinning abilities into just a few categorizations that supposedly overlap is doomed to failure.

So you have one of two options. Just make a choice, and accept that it cannot be totally realistic. Or, just list all abilities as single abilities, and have no "default" or "supporting" stats. This is the less used option, and people reject it because it's an "end description" sort of mechanic rather than a "begining description" sort of mechanic. Despite the fact that they both work just as well in play.

You can think about something like this forever, and get a million different opinions on the subject. In the end an educated guess will work as well as anything else.

Mike

[M. J. Young]

That is to say that the only thing that can completely and accurately portray the ability to use firearms would be a rating of your [ability] to use firearms.

There are, however, two aspects of reality that are not generally well modeled by this which have come up already in this thread: what is your skill ability level in a skill which you have never before attempted, and what is the degree to which skill in similar or related areas impact that?

J. has pointed out that the first time he ever fired a weapon, he rated second in the ability, quite near the 99th percentile of a company of military recruits, and outshooting even one person with prior training in firearms. I'm certainly not against the idea that a person could learn a skill at above beginner level (Multiverser includes this as a possibility, although the difference allowed is rather slight); the problem is devising a way to determine when it happens that the character starts at a higher level than most.

The other question of course is whether similar or related skills can impact skill ability in an untried skill. As an interesting example, can video or arcade games using gun-like controlers in a first-person shooter mode train you for firing a real gun at real targets? Most people would say no, such a simulation of combat firing will not make any significant difference--yet we do train our pilots on flight simulators which attempt to provide the look and feel of flying a plane without risking the hazards of actual flight. At some point, it must be the case that having done A gives you skill in B; the question is how to identify A for any given B, and to determine the degree to which that skill is gained.

Creating attributes which support certain skills is certainly a shortcut to this, and it's certainly less than completely adequate--but it captures some of it. There are probably other ways to do it. In the example offered by J., one could almost imagine the dice hitting the table and the referee announcing that he found himself quite comfortable with the guns, as if he were born to them, as he accurately hit more targets than either he or the instructors expected. Such a random method can create surprising talents; but it also creates inconsistencies and a lack of reliability at times. You might do better to provide what we might call "blank skill slots", where we simply say that the character has five skills of specified levels not listed, and can fill them in at the opportune moments in play. This, though, will lead to strategic use of such skills (it's unlikely in most games that the player will suddenly announce that he's discovered he's quite good at playing the flute, particularly after someone else at the table has announced superior skill at the rifle). The referee can create a list of hidden talents for each character, and reveal them as the situation arises--but this might lead to charges of favoritism if it appears to the players that one got better hidden talents than another.

So there are a lot of ways to deal with the problem, and problems that stem from each of them. Attributes as part of success has the merits that it is reliable and familiar, along with many of the disadvantages already observed in this thread (and some by me).

--M. J. Young

[Mike Holmes]

The first problem, that of "starting skill" is one that has to be addressed by the character enumeration system (including generation and advancement) I agree that not all characters will have the same starting default. But how to set that default is another matter. My point has been that if firearms are that important to the game, that they ought to be rated from the start, and rated using whatever method the system uses. This still leaves the question of skills that are not so enumerated. In that case, I see two options. Either start them all at a basic default which is unrealistic but simple, or allow for some sort of on-the-spot calculation that might involve something similar to the chargen system, or even randomization. Lot's of potential ways to do this in play as neccessary. I can envision some system where the player argues his other abilities, and spends points to back it up, possibly combined with a random roll.

Anyhow, as to the idea of "supporting skills" what I'm describing is an End Descriptive system where all attributes and abilities are already considered. So, the skill rating selected already has taken into consideration all possible "supporting abilities". (This, BTW, is the only simple way to achieve actual point balance).

The point is that this is the most accurate way to enumerate an ability. It doesn't provide the sort of support for underlying analysis that you seem to think is needed, MJ, but then I'm not sure this is sought. If it is, I've said that the other alternative is to simply pick some method of assigning defaul support, and go with it, despite the fact that it can't possibly be accurate. Most games do, and it works fine.

There just is no perfect solution. One has to pick between what they see as the lesser of two evils. Either have no underlying rationalizing support and be more accurate, or have such support and fail to be accurate. Which is worse? You decide. I can play either.

Mike

P.S. this is treading very close to territory covered by Mike's Standard Rant #4: Stat/Skill systems.