Would this idea work? Has anyone tried it?

[ffilz]

I'll toss a couple comments in here:

1. It can be hard to define a character when you don't have a list of skills. On the other hand, your proposed method relies on describing experiences from which one can derrive skills, so if someone says they have climbed Mt. Everest, we don't have to worry that he didn't write down "tent pitching" as a skill (whereas if he had put "excellent climbing" he might not have realized that "tent pitching" was an important skill).

2. The key in any system is communication between the GM as to the capabilities of the character. Now it is possible for the capabilities to be fuzzy, but if I've declared I led an expedition to Mt. Everest, we shouldn't be getting into an argument as to whether I have a good chance of estimating the danger of some climbing task, even if the system says that the GM doesn't need to give me the actual chance of failure (i.e. don't let me assume I probably have a 90+ percent chance of success when you're thinking I have a 20% chance of success).

Also realize the players used to systems with objective ratings will be constantly pressuring you for objective ratings...

Frank

[neelk]

I've done it, and it works well and is a lot of fun. But I'll never do it again. The reason is that it's a HUGE amount of extra work for the GM. Basically all the mechanics stuff that the players do, ends up in the hands of the GM, and then even very light mechanics (like Feng Shui) become excruciatingly tedious.

Now, if I understand you correctly, you want to get people to engage with the game at the level of narrative reality rather than mechanical reality. This is a good goal, but I don't think the specific technique you suggest is the best way to go about it -- at the end of the day you still have someone (the GM) translating actions into and out of a mechanical system. Instead, what you should do is look for ways to eliminate that translation entirely, and enable all the players to play with words and narrative rather than with numeric models.

One game you should look at is the Engle Matrix Game. In it, the players make arguments about what should happen in plain English and a referee rates them based on his or her perception of their credibility. The players don't have to fiddle with numbers and mechanics because the way they play the game is to talk about "the real stuff" directly. (You can read my review of it if it sounds interesting.)

[Sir Privy Toastrack]

One game you should look at is the Engle Matrix Game. In it, the players make arguments about what should happen in plain English and a referee rates them based on his or her perception of their credibility.

That is precisely what I am proposing (although I haven't stated it clearly perhaps).  The player states what he wants to do, gives the rationale for the attempt (I climbed Mt Everest, so I should have a darn good shot at hoofing it up Mt Washington) and the GM rates it on a difficulty scale.  Player rolls dice.  Result interpreted by GM.  Room for player debate.

The players don't have to fiddle with numbers and mechanics because the way they play the game is to talk about "the real stuff" directly.

That's what I want to try with this game.

[Sir Privy Toastrack]

Also realize the players used to systems with objective ratings will be constantly pressuring you for objective ratings...

Frank

That's true, but players can always adapt to change.  I remember when the concept of an RPG without level-based progression was heresy.  I remember joking with a friend once about playing an RPG without hit points.  Impossible!  Won't be accepted!  I'm guilty of basing most of my opinions on my gaming group, the most trusting group o' guys you could find, who trust my judgement and fairness and are open to literally any type of new style.

[Sir Privy Toastrack]

2) It doesn't just facilitate the idea of GM-fiat, it removes any access the players have to an exra-GM authority (the 'rules') that they can use to lend credence to their version of events, creating a huge disparity in control.

There is no question that these rules could only work with a mature, trusting and fair GM & players.  They are who this game would be aimed at.  Definitely not for rules-lawyers and players who see RPGs as a competition of GM vs player, but for people who view the GM & players as collaborators in an unfolding story.

3) With players--by far I would guess--more used to having a specific idea of their skills, they may be very unhappy not being able to accurately predict how well their character can perform.

Their accuracy in predicting things would be the same as in real life.  We don't look at our "skill levels" when attempting to jump over a hole... why should PCs?

The underlying problem is that you've taken away the players' ability to cite an external authority and, in so doing, you've only increased the power of GM-as-god.

True.  But again, I'm not aiming this at people who would be attracted to that type of reaction.  I don't view the GM as the enemy of the players, but as an arbitrator and guide.  There would always be room for players to state their case if they feel the GM is wrong.  I just hate rules lawyers (either GM or players) and hate the "wait, you can't do that -- it says so here in the rulebook!" mentality.  By stating that, I realize that I've made this idea appeal to about 1% of the RPG populace, but... oh well.

A lot of 'ifs' have to go in the right direction to make for good play, and that doesn't strike me as good design.  Good design shouldn't mean that you have to get lucky in order to have good play.

No luck involved.  Logic, experience, role-playing.

The fact that some type of negotiation is almost always required means that many experienced gamers, I would surmise, have found it easier to dispense with the idea of setting a statistical baseline in the first place.

Correct.  That is my hope and logic with this idea.

Good luck with the system though!

Thanks, I'll need it, based on reactions so far!

[SlurpeeMoney]

Ah Gawd... The paperwork...

Once, in an effort to remove my players from their character's stats and have them focus instead on their character's personalities and deeds, I took away their Vampire character sheets. I kept them in a nice neat little folder and told them I would make all of the adjustments neccessary to their characters over the course of the game.

Holy Mother of Blessed Gawd did that suck. Dice rolling became a moot point, taking away half the Vampire system; either I rolled the dice, or told the player to roll X-dice, after looking it up on their sheet. Neither of those was particularly interesting. Advancement, injury, all of the standard book keeping you expect your players to do on a regular basis... Impossible in this kind of game. It all has to be done by the Game Master.

Which brings us to two rather important points:

1) Keep the rules as simple as humanly possible, to limit the huge amount of book keeping involved.

2) Why are we using numbers at all?

Keep the Rules Simple
As I recall, the Pool involves a dice-pool betting mechanic in which any action can use as many dice as the player likes, but once those dice are used, they're gone. There was, as I recall, a way to regain dice, but I cannot be arsed currently to look up what that is. Anyway, it works, but the system is quite removed from the Game Mastery games we're talking about right now.

Here is what I suggest:
Character creation consists of
I Am: A description in three points (objective or subjective, it matters not) of who or what the character is.
I Am Not: A description in three points of what the character is not.
I Can: Three things that the character does well (objectively or subjectively, or add more things and make a bit of both; whatever)
I Can Not: Three things that the character either does poorly, or cannot do at all.

All of the points must be in the realm of possibility and relativity to the I Am statement. To say "I am a human, I cannot fly," is rather redundant. Saying "I am a human, I cannot eat," is more possible, and presents an interesting challenge for the character to overcome. It is a point that would need a great deal of expounding upon, and would be abused without end ("I am a warrior, I cannot sleep"), so obviously rules would need to be made to limit this kind of thing; I am only presenting the barest of bare bones here.

All rolls are made according to these statements. If a character Can do something, he or she gets a bonus that the Game Master keeps to him or herself (let's say +3 for now). If a character Can Not do something, the character is disallowed or penalized in performing that action. If something is completely against the character's I Am or I Am Not statements, they are also penalized. Everything else gets a fair shake.

Who's Line is it Anyway? Option
I'm a big fan of Game Master fiat. I realize that makes me incredibly unpopular here, but I feel that Conditional Authorship makes for a much more enjoyable story experience.
My favorite option, therefore, is the Who's Line is it Anyway? Option. In this option, the Game Master has his or her players roll based on numbers he or she keeps on a piece of paper (really just a doodle or something), and dictates the results. The points don't mean anything.
But that's just my opinion on the matter.

Kris
"8000 points to Mr. Edmunds for the hair."

[Sir Privy Toastrack]

Here is what I suggest:
Character creation consists of
I Am: A description in three points (objective or subjective, it matters not) of who or what the character is.
I Am Not: A description in three points of what the character is not.
I Can: Three things that the character does well (objectively or subjectively, or add more things and make a bit of both; whatever)
I Can Not: Three things that the character either does poorly, or cannot do at all.

I like this concept.

[M. J. Young]

Bill, we do something like this in the On-the-Fly character generation for Multiverser that we introduced in demo play and try to get out to our players through other means. (It's a shame we didn't have it for the rules, but at some point you have to publish, and after that you just provide support.)

It's helped by the facts that 1) new players don't really understand the number system we use when they start and 2) it's rarely necessary to create a second character, as your first one will last forever and remain interesting. But even so, it's worth consideration.

For attributes, we ask players to pick out any for which they are among the best of the best--ninety-eighth percentile, one in fifty or better. Those we focus on getting as accurately as we can in the system. The others we figure are somewhere between below average to considerably above average, and we can place them as we need them. Similarly, with skills, we pretty much allow amateur skills to be added on the player's statement that they are things the character can do. At the start, we look for professional and expert skills, where we clearly define "professional" as "could be paid to do this" and expert as "professionals seek advice from you". There are other nuances between the below average, average, or above average professional or expert, but these are easily addressed.

However, everything gets turned into numbers and handed back to the player; and I'm a fan of this for a very specific reason that I think has not yet been mentioned.

If I'm looking at a musical score (something at which I have some skill, which is why I picked this example), I have a pretty good idea subjectively how difficult that will be for me to play or sing. I can't tell you that there's a 70% chance that I'll be able to perform it in three days; I can tell you that I think I can do it.

Presumably if I'm playing a character who is a musician, he can do that, too; but there are some insurmountable obstacles to getting that subjective information to me. First, there probably isn't a piece of music for me to examine; he's examining the music, and making the assessment. Second, I can tell you whether I think I can play a particular piece of music, but I'd have a lot of hubris if I decided to tell you how much practice James Galway would need before he could perform it, even though I've heard him play. Even my assessment of the abilities of musicians with whom I perform is one step removed, and could be way off. There is really no way I can adequately give the character's assessment of the difficulty of a piece of music to him that doesn't really exist in any tangible form.

Numbers overcome this. They give us an objective basis for making a subjective assessment. Now I can say that my character has a 2@5 instrumental music ability, which combined with his other numbers and subtracting the difficulty rating of this piece gives him a 60% chance to play it right on the first reading, and a 93% chance to play it right given three days of practice. In other words, because I can look at those numbers, I know it's as reasonable for him to say, "I think I can play it if you give me a few days to learn it" as it is for me to do so relative to a real piece of music for myself.

You seem to be suggesting making the entire thing subjective. "I've played Flight of the Bumblebee; this is more difficult than that, but not by much, so I think I can play it." Now the referee makes a subjective assessment of how much more difficult this is than that, has the player roll the dice, and determines from that whether he successfully plays this piece.

Either you're going to have to dump the dice and go with a karma/drama blend on the mechanics, or you're going to have to find a way to define the translation from the subjective to the objective pretty clearly. A great deal depends on the referee's ability to make those assessments of how difficult climbing K-4 is compared to climbing Everest, which while not apples and oranges is also not something which is easily quantified on the fly. It's much simpler to say that this character has a stat of X in mountain climbing and has successfully climbed Everest as part of building up that stat, so he has a chance of climbing K-4 that is determined by subtracting the difficulty rating of K-4 from the mountain climbing stat.

Objectivity is rather important in this process, and that's what the mechanics give us. If all I know is that I swam the English Channel once, and now I'm faced with trying to cross the St. Lawrence Seaway, I have no basis for deciding whether the risk is reasonable. How does crossing the one compare with the other? It's an entirely subjective assessment that I would make by looking at what was known about them, and remembering what I felt like after crossing the one (thus would I think I could have gone farther if necessary). I can't get that from my character, so I need to use the numbers instead, which tell me what my chance of successfully making the crossing would be--or at least which point me in the right direction (Multiverser includes at least one number unknown to the player in its calculation of chance of success in skills).

That's my thinking. I hope it helps.

--M. J. Young

[dredd_funk]

dredd_funk wrote:
A lot of 'ifs' have to go in the right direction to make for good play, and that doesn't strike me as good design. Good design shouldn't mean that you have to get lucky in order to have good play.


No luck involved. Logic, experience, role-playing.

Sir Privy - I understand what you're saying, though I think there is luck involved from a player perspective: you have to be 'lucky' to find a GM that is a logical, consistent and experienced role-player.  From some comments around here, it seems like that takes more luck than one would think.  I was really just highlighting the fact that a system like this won't encourage the GM to be these things; i.e. there are no checks and balances against GM authority.

In terms of the accuracy of prediction (my problem #3) I absolutely agree--that's why I like this system in principle!!  I was merely pointing out the habits of players that would need to be overcome (as I mentioned in the paragraph which immediately follows: "#3 can be overcome through a good social contract and repitition of play."

I do like the idea, however, and would love to see it work.  Good luck!

If I'm looking at a musical score (something at which I have some skill, which is why I picked this example), I have a pretty good idea subjectively how difficult that will be for me to play or sing. I can't tell you that there's a 70% chance that I'll be able to perform it in three days; I can tell you that I think I can do it.

MJ - I agree with this.  Like I said, I think statistics provide good guidelines for negotiation, though I do think some form of negotiation almost always happens.  Even in the music example--a good one for me btw, because I've played guitar for a number of years, just for fun--while you can be pretty sure you can play it in three days, how many hours of practice are required for the 93% chance of success?  Three days doesn't mean 72 hours of straight practice--and no GM or player would argue that---nor does it mean 30 minutes of practice each day.  This doesn't make much difference if the GM and players have about the same amount of time in mind.  Problems arise when the GM and player don't have a similar amount of time in mind.  Problems also occur when qualitative issues arise, or 'how well can I do it'.  Practicing enough to play well for friends and family is different than practicing well enough to play for a television performance beamed out to an audience of 10 million.

To sum up:  both the 'can I do something' and 'how well can I do something' questions almost always have conditions attached, even if those conditions aren't explicity discussed by the rules (perhaps because a common shared idea of what is 'reasonable' is assumed).  Again, I'm not against statistics in any way!  In fact, I view them to be a very workable method to help 'center' the negotiation, both in the sense of providing a point of reference for negotiating what a character can do and also in the sense of providing a point of reference for player/GM control of the negotiation.  Without statistics a designer would have to supply a mechanic to fulfill both these needs.  Both ways can work out well.

In the end, I don't mean to quibble, especially since I'm very new here!  Take all of this rambling with a grain of noob salt!  I just think if negotiation isn't explicit, it's usually because the GM and the players are simply making the same, unvocalized assumptions about the conditions of the task, thus preempting the need for vocalized negotiation.

I do think that your point here highlights a very real problem of making this system workable: how to you provide players/characters with a system for predicting their success in a reasonable fashion relative to your design goals (mine would be simulation)?  Certainly it would be unfair to provide them with no mechanism for prediction.

Cheers!
Chris

[Sir Privy Toastrack]

MJ, Dredd and everyone else:

I really appreciate the intelligent feedback given on this topic thus far.  I can't disagree with anything said here, really.  The system has the potential for GM abuse and has a big hill to climb to gain acceptance.  I don't even know if it can work.  What I'm going to do this week is post a version of my game with this system (the game is called Land Without a King ad is a historical RPG set in 1141 England.  It has been posted here a few months ago with a VERY different ruleset).  

I'm also going to be giving away free hardcopies (yes, you read that right) to the first 25 people willing to playtest it and give thorough, honest and objective feedback, as well as suggestions on how to make it better.  I'll mail copies out with a feedback form.  It will be printed from a color laser printer on some linen paper.

[Michael S. Miller]

Hi, Bill.

Writing a role-playing game that is closer in style to improvisational theater is a fine goal. I'm still not convinced that using numbers while hiding them from the players is the best way to get there. One of the strengths of improv is that the players know as much as the audience at the moment that they're doing it--no more, and no less.

You say you've never played a game without numerical ratings, so I'd like to point you to some.

If you're looking for mechanics to support improv, it seems to me that leaving numbers behind completely may yield the most fruit. Have you read Jon Tynes' Puppetland? If not, do so right away! This game uses no numbers, but has firm rules--such as dictating how the GM and the player must describe what's going on in the gameworld--that work toward creating an "improvised storybook" experience.

I'd also recommend reading Paul Czege's wonderful The Valedictorian's Death. Play centers around creating an improvised murder mystery. While I'm suggesting games, I can't fail to mention Ralph Mazza (posts as Valamir) and Mike Holmes' Universalis, and Vincent Baker's (posts as lumpley) Matchmaker.

BTW, could you please refrain from posting line-by-line replies in the future? The Forge's etiquette guidelines frown upon this practice.

[Sir Privy Toastrack]

Michael,
Sorry about the line by line replies.  My understanding was that these were frowned upon because of potential flaming and taking things out of context, which I haven't done.  I just felt it was easier to read a reply if you know specifically what was being addressed. No biggie, I'll just stop doing it.

Anyway, I have changed some of my ideas regarding having the GM "hide" the skill numbers.  This won't be the case any more.  The GM will describe the situation, listen to the player's intent, ask what experience the PC has to address the situation, reflect on the PCs appropriate in-game history, then assess a degree of difficulty.  Most of the time he will say something like "okay, roll a 15 or higher".  The player can accept the ruling or discuss why he thinks he might be better than what the GM has judged.  So there won't be any hidden numbers, the GM won't have to keep a massive amount of stat sheets, etc.  After a session or two, the GM will have a much clearer idea of what each PC is capable of and these exchanges will become shorter.  This also can take into account improving skills ("well, the last time you tried to climb the wall you needed a 16 or better and succeeded.  This time you'll only need a 14 because of your past experience with it.")

[Michael S. Miller]

Hi, Bill.

That sounds much more on-target for your stated goal. Personally, I'm fond of games that explicitly create opportunities for player/GM collaboration. I look forward to the PDF you mentioned.

[ffilz]

So here's a problem. By relying on memory to recall that last time you did a task similar to this one, you needed a 15+ to succeed, you are risking either inconsistency, or assuming everyone remembers the numbers that you don't want to record.

I think it's much simplere for the player to have a number on his sheet and then you modify it as appropriate. Of course even then, you still have the memory inconsistency problem, which is why D&D these days gives examples of DCs, and in many cases even sets a formula for the DC.

Now one might argue that some inconsistency has value, but I think that's what the dice do, they introduce an uncertainty that means you aren't consistent in what you can do (unless it is so easy you can always succed, or the system has something like D&D's Take 10 rule).

My thought is that resolutions that occur on a regular basis are worth codifying so they remain fairly consistent. Then rely on GM subjectivity to deal with the oddball cases.

But that's just my current thinking, perhaps there is a something to be said for getting away from this rigid thinking.

Frank

[Sir Privy Toastrack]

ffilz,

You have a very valid point.  In my experience, however, I feel that gamers don't forget this type of thing even if the GM does.  I know that none of my players would ever forget a successful attempt at just about anything!  Gamers of all types are so possessive of their characters and their accomplishments that I can't imagine them letting a GM forget something important.  I hope.