Training vs. Experience

[Mock]

I've been mulling an idea for the past couple days, based on something Jared posted in the "attribute + skill" thread. It seems to be a fairly common axiom that experience counts for quite a bit in the successful use of skills--training only goes so far.

Suppose you allow skills to be represented by 2 numbers in the 1-6 range. The first one indicates training in the skill, the other indicates experience using the skill (for example, Unarmed Combat 2/5).

If you let the training number indicate the type of dice rolled (1 corresponding to d4, and 6 corresponding to d20, and the others as appropriate in between), and the experience number indicate how many dice are rolled (this is from Jared's post in the other thread), the individual above would be a somewhat trained, but quite experienced fighter, and would roll 5d6.

A highly trained fighter with little actual experience--someone with, say, Unarmed Combat 5/2--would roll 2d12 instead.

The way I see it, training gives the character a higher potential - for example, the highest training level lets the player roll a d20 - but lack of experience means the character may not be able to capitalize on the training easily (i.e., a total rookie might only have the one die to roll).

Finally, in this hypothetical system, conflict resolution would likely be a straight cumulative roll against a difficulty number.

Any thoughts? It feels right, but some questions spring to mind for me that others in this forum might be able to answer:

1) Is this worth the effort of doing? Has it been done before?
2) Do the probabilities bear out the idea that this "feels" right?
3) Should it be harder to advance training or experience? My gut feeling is experience--real experience--should be harder to come by than straight training.
4) Do you feel this to be applicable to most skills, not just combat skills as I used in my example?
5) From a more theoretical stance, does this fit your idea of the relationship between training and experience?

I'm interested in your feedback.

-Mock

[Andrew Martin]

Surprisingly enough, this topic on training versus experience was brought up on the Ground Zero Games email list, in connection with military skills and experience. The soldiers on the list, who fight for the US Army, Marines, Navy, and special forces, and in several other nationalities, rejected this approach. They pointed out that modern military training is designed to match experience. Where it doesn't, you have disfunctional military. Examples were the North vs South in the American Civil War, where inexperienced green troops would quickly advance into musket fire when ordered, whereas highly experienced troops wouldn't advance at at all. This is an example of training not matching experience. Modern military training matches training and skill development to combat experience. The value of this is shown in the past ten years of American military involvement in the Gulf War and, recently, Afghanistan.

So if one's setting is Civil war or similar disfunctional systems (eg corporates in cyber punk setting, perhaps amberites and so on), your system could be useful. Otherwise, it's not. Sorry.

[Mock]

Guess that wraps that up. Back to the salt mines...

[Judas]

Originally posted by Andrew Martin
Surprisingly enough, this topic on training versus experience was brought up on the Ground Zero Games email list, in connection with military skills and experience. The soldiers on the list, who fight for the US Army, Marines, Navy, and special forces, and in several other nationalities, rejected this approach. They pointed out that modern military training is designed to match experience. Where it doesn't, you have disfunctional military. Examples were the North vs South in the American Civil War, where inexperienced green troops would quickly advance into musket fire when ordered, whereas highly experienced troops wouldn't advance at at all. This is an example of training not matching experience. Modern military training matches training and skill development to combat experience. The value of this is shown in the past ten years of American military involvement in the Gulf War and, recently, Afghanistan.

So if one's setting is Civil war or similar disfunctional systems (eg corporates in cyber punk setting, perhaps amberites and so on), your system could be useful. Otherwise, it's not. Sorry.

When I was in the US Army Infantry, my experience was that yes, training is now excellent for preparing for combat, but an actual combat veteran still has an edge. Many of the things they train you, the vets then give the disclaimer that you won't actually have time to use it, you will be too exhausted to think this clearly, here is a simpler way, throw away this equipment because it is more trouble than it's worth, etc.

I am definately not saying that US or other modern western forces aren't extraordinary, because they are. But the first month in combat improves a soldier more than the first year or two of training.

I would say that the training allows a warrior to get that improvement. The training keeps the soldier alive long enough to benefit from the combat experience, to be less-stressed and therefore better at picking things up, will know the common pitfalls to avoid, etc.

So I would say that an untrained character will gain very little from combat experience, whereas a trained individual will gain greatly. Perhaps the best way to model it would be untrained characters earning experience at 1/3 the rate of trained characters. Does that make any sense?

[kabael]

I think this is a very interesting idea indeed, but might need a bit of reworking.

For one, I don't usually like systems that use multiple die types, especially in scales like this. There are a few different reasons, but one of them is the big jump between d12 and d20.

Maybe if you tried to make the training or experience reflect a different mechanic?

What is experience determined the number of dice rolled or training determined the target number, or the other way around? In that kind of system, you wouldn't be able to start either one at 0 (except for skills that require training to perform at all, like brain surgery), but you could have the default be 1 or something.

This is definitely worth investigating, but it might turn out to be more complicated than it's worth :p

[Mock]

Originally posted by Judas

So I would say that an untrained character will gain very little from combat experience, whereas a trained individual will gain greatly. Perhaps the best way to model it would be untrained characters earning experience at 1/3 the rate of trained characters. Does that make any sense?



Not having any military experience myself, I find the input from people who've been there very valuable. The relationship you mention was one I had not considered, but it is very interesting: while experience can help you capitalize on your training, without the training, it's extremely hard to capitalize on your experience.

Originally posted by kabael
I think this is a very interesting idea indeed, but might need a bit of reworking.

For one, I don't usually like systems that use multiple die types, especially in scales like this. There are a few different reasons, but one of them is the big jump between d12 and d20.



I also balk somewhat at the d12 to d20 gap--it's a bit too discontinuous for my tastes. It was merely a suggestion. However, I am intrigued by your idea. Consider this system:

Basic resolution is a non-cumulative roll; each die to meet or exceed the target number is a success. Training sets (or more likely modifies) the target number, dictating how hard a task would be for you to do. Experience sets the number of dice to roll, increasing the margin of success.

Thus, a person well trained in computer programming would likely find writing a minor program trivially easy, and the target number for this task would reflect that. However, they might roll only a few dice, so their success margin potential is not as high.

By my interpretation, then, the highly trained yet inexperienced computer programmer would have little trouble creating the program, but the program might not be as elegant as that created by a highly trained and experienced programmer (who would create the program either elegantly, or quickly, or both, depending on success margin).

If we apply this to a combat skill--say riflery--a well trained, experienced soldier would roll lots of dice at a low target number, or in other words, does very well at shooting targets on the battlefield. A well trained, but green soldier, would still roll at low target numbers, but not necessarily have the excellent successes that the vet has. An untrained person with experience in combat zones rolls lots of dice, but the target numbers are quite high (this might mean, for instance, that they don't freeze up in combat, but they might have trouble operating equipment). Finally, an untrained, inexperienced person in combat ends up rolling very few dice at very high target numbers.

Whew. Furthermore, to reflect what Andrew said, there would be certain minimum levels of experience and training required for certain skills. A machine gunner, for instance, will finish training with a certain amount of trained skill, and a certain amount of experience too (this avoids belief-stretching problems such as totally inexperienced Navy SEALS).

And finally, to tie in Judas' comment about earning experience more slowly, the system would make it much more expensive for a person to earn experience at a skill, without having training in the skill.

Again, this is rough--it requires an idea of what "success margin" will represent in the game (something that the GM and the player may decide on the fly), and maybe there's another idea, like experience indicates the number of actions you can take, or the number of activities you can do simultaneously. The goal here is to make a believable model of interrelation, without making it overly complex. Let me know your thoughts.

-Chris

[threegee]

I like the idea very much. The idea that skill and experience are two different things is something that has been needing to be addressed in a game, and this is a very nice solution. Much nicer than the generic 'you have a bucket of dice--two, if you're a god.'

Naturally, this type of system will require target number modifiers for difficulty, rather than straight TNs.

I'm not sure that different rates of experience are necessary or desirable. With the higher TNs, the untrained are already being penalized. It's true that we almost always frontload our training, but that doesn't make it a good idea. An ideal education would be continuous, matching the experience a person has gained. Without experience, a lot of advanced training goes completely past a person, and must be relearned later on. However, for simplicity's sake in a game, I would treat training and experience as being unrelated.

[Mock]

Originally posted by threegee

Naturally, this type of system will require target number modifiers for difficulty, rather than straight TNs.

...I would treat training and experience as being unrelated.

I assume you mean that the amount of training would dictate a modifier to the target number, rather than a target number itself? That's what I had in mind--high training would subtract some value X from the basic target number.

As for the relationship between training & experience as far as earning them, I think you have a point, but I'm not sure where to draw the line. I think that there is some truth in the concept that better training makes you more likely to recognize and assimilate experience, but perhaps it's too nebulous a relationship to quantify in any believable way. Hell, it could be as simple as letting a player decide on one or the other as they advance.

To move slightly away from the central topic, in such a system, would you allow the player to decide what success margin means based on the situation, or would you set a particular interpretation of success margin? I like the former--for instance, suppose the computer programmer faces two challenges: the first is to impress someone with the elegance of his or her code. In that case, more successes would confer cleaner, more impressive code. The second challenge, at some other time, is to create a virus program on the mainframe before security breaks down the door. This time, successes might correspond to speed, while elegance is out the window.

My concern is that high margin of success only comes with experience. However, I suppose that doesn't seem unreasonable. :)

[Judas]

I like the incorporation of my concept (of course). :D

It is much like trying to learn physics. I could slowly teach it to someone without much math experience, but it would be a long, slow road. Someone with a background that included a good amount of mathematics would pick it up much more quickly. Also, without training it is easy to pick up bad habits, and the training may not truly prepare someone for real combat, but it will sure let him know the common pitfalls and misconceptions to avoid. I know in training that almost everything I knew about firefights from TV, movies, books, etc. was DEAD WRONG... there is no such thing as knockback, automatic fire does not make things much easier to hit in most cases and empties a 30-round magazine in about 3 seconds, you DO NOT attempt to pull grenade pins with your teeth if you value them, explosions do not have any kind of fire associated with them, tracers point BOTH ways, noise and muzzle flashes are as important in a night ambush as accuracy, etc.

As a matter of fact, the more I think about it, the more sense it makes. When you go to college, you learn an aweful lot of useless stuff. But as you are often told, you are being taught how to think critically, etc. When you get military training, you get fundamentals and a real introduction into thinking tactically. All those concepts gel, you learn the specifics for the combat situations your are actually in, and then they are combined into an effective overall skill set.

Sorry, ramblin.

[Mock]

Not rambling at all--this is the concept I'm after. The physics example is good, too--without the background, the fundamentals, you end up having to learn not only physics, but arithmetic, algebra, trig, calculus, all of it on the way. That's going to take a LOT longer.

What I find attractive about this method, too, is that although it may attempt to model skill & experience slightly...I hate to say realistically...let's just say differently, and gives a different better emphasis to experience, it's useful across gaming styles and genres, from ultra-realist to cinematic.

That is, despite the fact that TV combat is nothing like the real thing, this system wouldn't preclude you from playing a Feng Shui-inspired scene. It has an element of realism, without attempting to be a simulation.

[kabael]

Mock said that:
My concern is that high margin of success only comes with experience. However, I suppose that doesn't seem unreasonable.

Indeed, I think that is very much a good thing.

As for target number modifiers, I'm tempted to say that it's not the TN that varies with difficulty, but the degree of success, not unlike the version of the Storyteller system that White Wolf used in Trinity.

Of course, if experience is the only way to get a high number of dice, it's also the only way to accomplish the very difficult tasks in the real world.

That, I think, may be worthwhile to incoporate into a game, but certainly not intp every game, or even for every player.

[Mock]

I'm not familiar with Trinity (I have the freebie demo they gave out, but I haven't read it in ages). Can you give me an example on how they do it?

-Mock

EDIT: I agree--this isn't for everyone, but that's okay, 'cause I don't usually work on universal systems. This one has a military (or at least martial) feel to it, so it's likely to wind up in a game with a military flavor.

[kabael]

Well it was the same basic set-up as for other Storyteller games, but instead of raising or lowering the target number and/or requiring multiple successes to accomplish a task, all modifications to difficulty increased the minimum number of "successes" necessary to succeed and target numbers were always 7.

An easy task needed only one success, a difficult one maybe 3 or 4.

The only major problem I have with that system is that you can only make things so easy, unless you want to provide bonus dice. I wasn't really bothered by the fact that you had to be of at least a minimum skill level to get certain levels of success.

[André]

I've been thinking along similar lines. However, my system is different: Training and experience gives a number of dice. You throw all of them, then pick out (training) dice, and sum them. Experience gain is proportional to skill.

How does that compare?

[KlausGreen]

In my oppinion these two games use a fairly good and fast task resolution system (the system has been given a name but I can't remember it now, sorry!)

The basic concept is that you roll a number of dice equal to your Attribute rating and a number of dice equal to your skill rating. You then keep as many dice as you have Attribute rating and total the sum of those dice. All rolls are made with d10 and 10's are (as far as I remember) rerolled.

Target numbers are based around the complexity of the action.

If you want to achieve a higher degree of success than "Standard" you have to state your intention BEFORE rolling the dice and thereby raising your Target number by a set step of 5 each time. This is called a "raise" and you can only make as many raises as you rating in the appropriate Attribute.

This might be a little imprecise but is as exact as I can recall it.

There is no reason why you couldn't substitute this duo-rating system with a Training/Experience system, making Training essential for proper utilisation of your combined Training and Experience, but enabling Experience to make your likelihood of succeeding much higher (by rolling more dice, and therefore giving you a statistically higher chance of rolling more 10's).

Maximum amount of raises should be based on Training.

I also think that if you use Attributes these could be a sort of maximum for how high your training could get you. For instance if you have a Mental Capacity rating of 3 you could only accumulate theoretical knowledge (training) to a degree of 3 in skills such as Computer Programming, Speed Reading and similar. To achieve more than this initial maximum you'd have to gain experiences yourself through experimentations and adaptions based on your personal potential.

For beginning characters I would think Experience should be set at one or two points less than their Training unless they take a sort of advantage/feat called "Field Experience" or something similar. This could neatly be turned off if the GM wanted the characters to start out as hardened veterans.

How does this sound?

KR KGA

[kabael]

That is the system used in Legend of the Five Rings and (presumably, I'm not sure if it works exactly the same) 7th Sea. It works well enough, but does have a number of drawbacks, among them the fact that you can never succeed better than you intend to (there is no way to get lucky).

I really liked the Legend of the Five Rings at first, but the more I used it, the less I liked it. I think most of my appreciation was initially built on the elemental association of the rings and attributes. I'm a goob for fun elements.

[KlausGreen]

I'm not a very big fan of the incredible lucky break so favored by certain games. If all you're doing is giving a problem your most average attention the chance of you overachieving in any meaningful way is extremely limited. If you want to do your best, you must really put your back into and TRY. And I think the L5R system emulates this smoothly and with a reasonable amount of success.

If you try to do your best you also run the risk of ruining everything or placing yourself in a very bad position, but if you ran the more "safe" route of fighting/working to the average of your ability, you're more likely to get a positive result although you won't get a phenomenal result.

KR KGA

[Shawn Conard]

The L5R / 7thSea System does also have one other problem you should keep in mind. Now, I haven't played the newest L5R game, only the older one, so I'm not sure if they fixed it there, but...

You used to roll [Attribute + Skill] dice and add up the highest [Attribute] dice. It's a nice system. I like the feel.

However, it does take a while to roll anything. It doesn't seem like much, but when you also get to reroll 10s... Well, you can sometimes be stuck adding 10+3 + 9 + 8 + 6. Not impossible, but when this happens every roll, it does slow things down.

Second, and perhaps more importantly, Attribute is the uber stat. In fact, it is so useful that a higher attribute will win contests nearly every time unless there is a HUGE difference in skill (as in 5 or 6 points, or more). Also, if a character ever takes an Attribute of 1, they will have to get used to failing at easy tasks over and over again. This is a bit frusterating in a game.

~Shawn Conard

[KlausGreen]

= QUOTE =======================================
However, it does take a while to roll anything. It doesn't seem like much, but when you also get to reroll 10s... Well, you can sometimes be stuck adding 10+3 + 9 + 8 + 6. Not impossible, but when this happens every roll, it does slow things down.
==============================================

You're right, the system can be bogged down by a lot of dice rolling. I would like to state that this is quickly overcome when the players (and GM) become more accustomed to the system.

= QUOTE =======================================
Second, and perhaps more importantly, Attribute is the uber stat. In fact, it is so useful that a higher attribute will win contests nearly every time unless there is a HUGE difference in skill (as in 5 or 6 points, or more). Also, if a character ever takes an Attribute of 1, they will have to get used to failing at easy tasks over and over again. This is a bit frusterating in a game.
==============================================

[COLOR=red]If you want to have an attribute rating of one you want a character that is pathetically poor at something. It is not a flaw in the game system if this character never succeeds at anything related to this attribute, he just doesn't have the potential (at all!)

Furthermore my point was to come up with an alternate suggestion to the Training/Experience topic of this thread and I still think my suggestion is valid. This would also mean that as time passes the character will increase in Experience unless he receives some form of formal training (and thereby raise the Training part of their skill rating). Attributes would only be used in reference to a character's maximum learning potential or capacity. (See my original post)[COLOR=red]

KR KGA

[kabael]

Originally posted by KlausGreen
I'm not a very big fan of the incredible lucky break so favored by certain games. If all you're doing is giving a problem your most average attention the chance of you overachieving in any meaningful way is extremely limited. If you want to do your best, you must really put your back into and TRY. And I think the L5R system emulates this smoothly and with a reasonable amount of success.

I certainly agree that luck should not be central or even a commonly significant factor, but it should be a factor. Sometimes you do stumble into great success, especially in dramatic stories where the highs are higher and the lows lower.

Additionally, a big problem I've grown to have with the Legend of the Five Rings system is the overwhelming importance of the attribute over the skill. Your suggestions of limiting the number of raises to skill level can help that, but I'm not sure it goes far enough. My old gaming group actually divided the attribute and skill pools, using different colored dice for each, and only allowed open-ended rolls for skill dice that produced 10s. Again, some help, but not enough.

Attributes are still given center stage over training and experience, unfortunately. That's not something I really value in a game system anymore, especially in this context ;)

[Shawn Conard]

KlausGreen:

If you want to have an attribute rating of one you want a character that is pathetically poor at something. It is not a flaw in the game system if this character never succeeds at anything related to this attribute, he just doesn't have the potential (at all!)

---------------------------------

After a while, you realize what the attribute levels mean. But this isn't something which the rulebook helps you with, so people will probably create rather crappy first characters. Me, for instance. My first character was a German swordsman (trained in a well known swordsman's school and everything) who -- get this -- couldn't fight. He could hardly hit anything.

Little did I know... Ah well, I've since figured out what the attribute levels mean.


So... I do consider this a flaw. Given that the game is all about heroic swashbuckling, and that there was no warning given (or implied) about low attributes, I'd expected more out of them. Sure, the system still worked. But it didn't match the setting.

I guess I'd just wanted some warning that a trait of 1 meant your character was nothing but comic relief, no matter how many points you put into skill.

~Shawn Conard

[KlausGreen]

I don't think you are really getting my point in all this writing. I'm not suggesting that you use the L5R system as it is but I'm wondering whether or not it could simulate the connection between Training and Experience.

Training would act as the kept dice.
Experience would act as the modifier dice.

To put it in L5R lingo the formula would be:

(Training+Experience)K (Training)

In some of the previous posts it has been mentioned that experience means little or nothing if you hadn't had some training to base the experience on, hence if you have a very low Training score (kept dice) and moderate Experience (modification dice) you would probably not succeed, although LUCK would now play a factor.

Furthermore a person with some Training would gain greatly from at least a little Experience, thereby making his skill more dependable and LUCK not playing such a big part any more.

In this system Attributes would only define the maximum value of Training value and therefore your immediate potential. Attributes should have a maximum value of 5 with 2 being the starting value. To keep things simple I suggest four distinct Attributes: Physical Resilience (feats associated with strength or stamina), Physical Capacity (feats associated with speed or agility), Mental Resilience (feats associated with static knowledge such as history and feats associated with maintaining focus or concentration), and last Mental Capacity (feats associated with thinking abstract or using your mind to deduct something. Could be used in situations such as creating a computer program or understanding the grief associated with loss in a fellow human being).

The Attribute names are on purpose very generic.

[Shawn Conard]

Well, you are right I was getting a bit off subject...

I do think that the use of a pseudo L5R system would model Experience and Training fairly well. I just wanted to point out what I considered to be a few issues with that system. Now, maybe this is acceptable, or even something you want to encourage in a game system. In which case, no problem at all. However, just be aware that these issues exist:

1) Having a low Training means that you will almost never succede, even at easy taks, regardless of Experience.

2) There will be very distinct levels of Training. For example, a sharpshooter will be able to easily hit targets which others in the party (using the same rifle) can't hit.

3) Experience means very little, compared to Training. If you want to get better at things, go back to school.

Now, in a military setting, some (all?) of these might be considered good things. But again, just know that these things exist and either plan around them, or welcome them.

~Shawn Conard

[[Ego]Wumpus]

Ok, my major hesitation with breaking up training and experience like this is that unless the two are inter-related, I might build a gun-bunny with a single d20 when I start the game. Sure, this might be fine; I'm really talented, but have no practical experience. Except that suddenly, when I dump all the xp I ever get into experience in shooting things, I've got 4d20 or 5d20.

Which, I think, is totally out of hand.

What I do like is the 'core mechanic' of having two attributes of the character interact to determine the dice pool. What it needs is some constraints. I suggest:

- Use only d4, d6, d8 and d10. If you're really into it, add d12. There is an even distance between these.
- Make the costs for buying them exponential. If it's 1 XP for the d4, make it 4xp for d6, 9xp for d8, and 16xp for d10. Repeat for the other attribute. So to get 1d4 you spend 2xp. If there is one thing Sillouette did right, it's utilizing exponential mechanics.
- Which makes me think of another solution; suppose 'training' gives you skill 'tracks'. When you open a 'track' in a particular skill, you get 1d4 in that skill. If you gain experience using that skill, then you increase the size of the die. So, if you train for 2 years you might have 5d4, but once you get into combat, suddenly you have enough experience to get 3d8 and 2d6. More training means more things you can utilize your experience on cheaply, more experience means bigger potential.

Either way, I highly suggest whipping out Excel and looking at the 'dynamics' of the situation; comparing the percentage likelihood of all possible results to each other. Find a system you like already and compare it, see if they have similar attributes. As usual, I suggest Sillouette. Play around with specific dice and options and see how things change. As long as you aren't using open-ended systems these analyses ought to be simplistic enough.

[Lunamancer]

Both training and experience assume some form of improvement in a skill due to practice and correct repitition. The difference between the two is that experience encompasses more. Training is an isolated sort of experience, where only the correct way is emphasized, and typically under a series of common scenerios (depending on the nature of the skill).

What experience covers that training does not are those things outside the "theory" of the skill--those miscellaneous things that can often be a factor in success or failure but is not truly a part of the skill. The best way to handle these miscellaneous things varies greatly depending on the rest of the game system, but for a majority of the systems, I'd say "experience" would best be reflected through the non-skill attributes.

If the skill in question were Engineering, for example, the "theory" of the skill would pretty much assume required materials are on hand. In practical use, when some of those materials are lacking, compensating for them would not be a test of one's engineering in and of itself, but rather some other stat: Intellect, Wit, Creativity, Pantology, what have you. Only when that applicable stat check succeeds, can the Engineering commence.

The problem with trying to express training and experience separately with funky dice mechanics is that, as in my example, some stuff you'll want to roll for while others you won't. I'd require a roll on Creativity or whatever to find ways to use what you have in the place of what you need. Once that's done, it's simply a matter of whether or not the Engineer is capable of what is being attempted. Only if something never done before but is theoretically within the capabilities of the engineer would I require the engineering skill check.

As for assigning advancement to the various areas.. training vs experience.. assume practical use will award both training points specifically for the advancement of the particular skill, as well as some generic advancement points that can be assigned anywhere. In the above example, the character might recieve an XP in Engineering as well as a generic XP (which the player can assign anywhere... since in the future, it could be any of the other stats that make Engineering posible, depending on the situation). Training, however, would only award specific XPs in the area being trained.

That's what makes the most sense to me for dealing with Training vs Experience issues in RPGs, and to satisfy the differentiation doesn't really require any kind of new or Earth-shattering design.. it's already been done in many RPGs!

[Finmenagon]

D&D, the first of the popular systems, used levels and training to determine a character's ability to strike an openent. Experience = Levels. Level determines THACO. Also, as you gain in levels you train on your weapons(weapon proficiencies) and level proficiency in a weapon grants a bonus or penalty to the roll.
So, training and experience effectively determine how successful the character is at combat. While I don't entirely disagree with those who say D&D is a abstract dinosaur of a system but I think it is interesting that it utilized training and experience and now many of us are trying to find ways of dealing with it.

My real addition to the conversation...
I've been tossing the idea of Formal and Informal Paths of training...this is in no way a completed comprehensive system of dealing with the problem, just a concept to be considered.

If a character is studying a skill under the tutelage of a mentor or in an academy or at a military training facility than they are following a Formal Path.
If a character finds themselves shipwrecked on a desert island and forced to learn how to survive and fend off monsters and all that, than it is an Informal Path.
When a GM or World Builder designs a setting they would have to include the Formal Paths available in the setting. I think that most Formal Paths would be exclusive to other Formal Paths: the Formal Path of a Marine would most likely keep a character from partaking in the Formal Path of a Proffesional Athelete. Formal Paths usually cover many abilities but never all(at least not in normal Campaigns).
Both Formal and Informal Paths would allow for a character to develope his abilities through both training and experience. Formal Paths would have clear deliniated training advancement...for instance: first you get your associates degree, then bachelors, then masters and finally a phd OR first you pass your manhood rites, than you begin your village warrior rites, ect...
Any ideas to add? Critisisms? Considerations? Possible suggestions for developing this concept into a workable system?

[[Ego]Wumpus]

Originally posted by Lunamancer

As for assigning advancement to the various areas.. training vs experience.. assume practical use will award both training points specifically for the advancement of the particular skill, as well as some generic advancement points that can be assigned anywhere. In the above example, the character might recieve an XP in Engineering as well as a generic XP (which the player can assign anywhere... since in the future, it could be any of the other stats that make Engineering posible, depending on the situation). Training, however, would only award specific XPs in the area being trained.

That's what makes the most sense to me for dealing with Training vs Experience issues in RPGs, and to satisfy the differentiation doesn't really require any kind of new or Earth-shattering design.. it's already been done in many RPGs!

I disagree.

First of all, yes training and experience has been done in many RPGs, but I really wonder how well? After all, are you really going to say that some RPG out there has found a mechanic that models well how people learn? To say such a thing is to say that you have done a job the entirity of Educational Theory has not.

Secondly, when you sit in an Engineering class - and I've been in a few in my day - yes, you are learning theory regarding engineering. But between the homework and the labs and the lecture, you are learning things which have nothing to do with Engineering more than you are learning Engineering. On the other hand, on-the-job experience will give you more engineering knowledge than any Ivy League-equivalent education can. But will that on-the-job experience translate to knowing how to draw well? Both fall under the realms of Creativity.

At least in the case of Engineering, in your example, I would say that for training you'd get a general XP and for on-the-job stuff you'd get a specific xp. I think, though, its a poor mechanic.

At the very least, the interesting notion about the dice mechanic I proposed was that with sufficient training you are unlikely to *ever* fail. Likewise, you will be unlikely to ever succeed spectacularly. Only with on-the-job training can your numbers really get up there, but with sufficient training you can gaurentee you will get a minimum number. If you have seven tracks open, with no practical experience, you will get at least a seven. Which is twice as good as a lucky beginner.

However, my preferred mechanic for handing out experience has nothing to do with sessions or specific in-game activities. Instead, I think its optimal to give PCs a certain amount of experience per unit time. This amount can be adjusted for any number of reasons, but stays pretty constant. The player then decides where they put that xp.

'But that doesn't account for high-learning situations!' Except, I think, it does. PCs will place experience in things they think they will need; and that will be based on past experience. And not all stressful or action-filled situations result in a high amount of learning going on. One thing you *can* say, however, is that over time people learn a constant amount, and therefore it will balance out.

[Ardwen]

Finmenagon wrote:
> Both Formal and Informal Paths would allow for a character to develope his abilities through both training and experience. Formal Paths would have clear deliniated training advancement...for instance: first you get your associates degree, then bachelors, then masters and finally a phd OR first you pass your manhood rites, than you begin your village warrior rites, ect...


Man....what a stumper for mechanics! The formal path would be easy to deal with I think. As for the informal......would some of this be left to GM discretion as to when a character is granted Xp or an ability or whatever?

Informal training would have to invovle some sort of intelligence check or inspiration by the character to actually pick up on an idea that works for a given situation. (that was a mouthful) :rolleyes:

[kabael]

This thread may be dead, but my post in the Mechanical Dream rules (http://forum.rpg.net/showthread.php?postid=21828#post21828) thread reminded me of it, and people interested in another possible application of a difference between training and experience may want to check it out, or look at the PDF directly at http://www.steamlogic.com (I think that's the url).

[Godwin Fortesquieu]

There are really two different categories of skills that we have to talk about: those for which training and experience are different, and those for which training and experience are typically of the same nature (craft skills are the best example of this).

One way to handle the difference is the GURPS solution (though I don't know if it's intentional):

A) When designing a list of skills, keep in mind which skills are best learned by experience as opposed to training (particularly combat and outdoor skills), and which are not.

B) When designing categories of skills best learned by experience, make it advantageous for players to take numerous peripheral skills. Non-Skill mechanics to separate the experienced in some of these categories from the trained are also a good idea (GURPS uses Combat Reflexes this way).

C) When designing skills like many craft and academic skills, where training can simulate experience, there would be less overlap.

The problem that GURPS encounters is, I think, a side-effect of the (sensible) distinction between Physical and Mental skills. It's really easy to get Mental skills at really high levels (someone with a Dexterity of 12 would have to invest a prohibitive 56 points in an average physical skill to have it at rank 20 (master), but a character with IQ 12 would have to spend a mere 18 points on a mental/average skill to have a 20). This works fine, but it makes it really tempting to design more mental skills for the 100-point scholar with everything, which leads to even more of the skill-glut which makes GURPS so unwieldly for anyone who doesn't really know it well.

If you can get around that, you can simulate this dynamic relatively easily. Maybe a hierarchical skill system, or a very light class/level system on a fundamentally skill-based game, where attributes are primarily useful to provide defaults and secondary modifiers, in addition to an effect on skills. That would allow you to trade the GURPS problem of an excessively-long skill list for the albatross of class/level hell. Six of one, the square root of about 2.45 of the other.