I wanted to run this idea for using a karma based system for resolving conflicts and I was hoping for some criticism

At the beginning of a scene the GM secretly selects a numerical value from 1 - 10 as a representation of the scenes Dramatic Significance (abbreviated as DS). The more integral a scene happens to be the lower its DS. Thus a scene in which the PCs are in a climatic battle would probably range in value from 1-3 while a scene in which the PCs are gazing at their navels is likely to have a DS of around 8 or higher.

Next the GM secretly selects a Luck Value for the scene (abbreviated as LV). This is again a numerical value from 1-10. The higher the LV the more luck is available within that particular scene. Who will be able to use that luck is dependent upon factors which I will explain later.

Anytime during a scene a player has the opporturnity to use their PC's Defiance to improve their chances of resolving conflicts in their favour. Defiance is a measure of a characters health, experience, determination, will etc. Players can gamble their character's Defiance on the chance that it will make their character more effective during a scene. If the Defiance gambled by a player exceeds the DS value of a scene their character accesses some of the luck attributed to the scene through the LV assigned to it by the GM.

The more luck a player can access for his character the better their chances of successfully resolving a conflict. However any luck that they fail to access for their character will go against their character. Moreover players who gamble too much Defiance may lose out on accessing any luck at all for their character. Any luck which is not used by the character becomes a modifier for the difficulty of any conflicts in the scene. Thus luck can be good or bad depending on how much of a scenes luck a player accesses for his or her character and how much luck is left over.

There are two ways of avoiding luck altogether. Don't game enough Defiance to access the luck in any given scene or gamble to much Defiance.

For example... A scene has been secretly assigned a Dramatic Significance of 5 and a Luck Value of 5.

A player gambling 4 points of Defiance would not receive any modifiers to their attempts to resolve conflicts as they gambled less than the scenes DS of 5 and all other considerations would be exactly as if they hadn't gambles their 4 points of defiance.

If the player gambles 7 points of Defiance they will have accessed some of the scenes Luck for their character. In this case it would translate into a +2 modifier. Unfortunately as there are three points of luck still unassigned these points go to the difficultty of any conflict within the scene, e.g. a conflict with a difficulty number of 10 would now be 13 (higher numbers translating to increased difficulty).

If the player gambles 10 points of Defiance they will have accessed all of the scenes Luck for their character. In this case it would translate into a +5 modifier. Fortunately as there is no luck left to assign all difficulties remain as they were orginally, e.g a difficulty number of 10 would remain at 10.

If the player gambles 12 points of Defiance they will have lost all luck for that scene resulting in no modifiers for their character and no modifiers to the difficulties of conflicts within the scene.

On to the actual resolution of conflicts of which I'll give a quick example...

A character is attempting to persuade an official that documents that have been handed over by the character are genuine and definitely not forgeries. The GM ask which attribute the player will be using to resolve this conflict. To which the player responds that she will be using her characters confidence attribute. The GM agrees to this but decides to penalize the chances of the character succeeding because of his unusually high sincerity attribute.

The characters confidence attribute is 9. Taking in no other considerations, the character will always pass any conflicts that depend upon confidence as long as the conflicts difficulty is 9 or less.  Confidence is a specific kind of attribute called a quality. Qualities include such things as Awareness, Acumen, Ingenuity, Strength etc. Qualites are ranged from 5 to 15. The higher a characters quality the better they are at using it.

The characters sincerity attribute is +3. This means that whenever the character is attempting to persuade someone of a truth, he gets a +3 bonus. However anytime the character lies that +3 bonus becomes a -3 penalty. The character just doesn't have it in him to be a liar. Sincerity is a specific kind of attribute called a virtue. Virtues include such things as Honour, Loyalty, Compassion, Patience, etc. Virtues range from -5 to +5. The more virtuous and principled a character the more positive virtues they will possess and the higher those bonuses will be.

(Note that in the game virtues and qualities are not called attributes. I'm just using the word attribute in here because that's what these things are usually called in most RPG's I've ever played and it easier to use common terminology)

Thus if the character attempts to use confidence tell a lie he will succeed as long as the difficulty is 6 or less (9-3) and if they try to use confidence to convince another person about a truth he will succeed as long as the difficulty is at 12 or lower  (9+3).

The GM has set the difficulty for persuading the official at 8. If the character was telling the truth he would have no problem at with resolving this conflict in his favour, but his damned sincerity is playing against him. There's no way that he'll be able to make this official believe that the document is not a forgery. Unless that is if he can get lucky.

To get lucky the player will need to gamble some of her character's Defiance. She will need to gamble more than Dramatic Significance of the scene but less than the scene's Dramatic Significance and Luck Value. For ease of explanation lets assume that the DS and DR of the scene is at a 5/5 as it was in the previous example.

In order to get the enough luck to raise her characters ability to successfully lie from 6 to an 8 the player will need to gamble 7 points of Defiance. 7 minus the scenes DS equals two. But this leaves excess three points luck which will be assigned to difficulty of the conflict raising it from 8 to 11 and that means even with the extra luck the character will still fail.

The only way that the player will have for resolving this conflict in a successful manner will be to gamble 9 to 10 points of Defiance as this will raise their ability to lie from 6 to 10 if she gambles 9 points of Defiance or 11 if she gambles 10 points of Defiance. Meanwhile the difficulty of the conflict will only increase from 8 to 9 if she gambles 9 points of Defiance and it will remain at 8 if she gambles 10 points of Defiance.

Thoughts, suggestions, criticisms?

So, if I understand correctly... using shorthand: G = amount of Defiance gambled; L = Luck available in pool; DS = Dramatic Significance.

G < DS: No effect
G > DS, G <= (L + DS): Luck pool decreases by (G - DS), which becomes positive modifier for rolls. The remainder (effectively L + DS - G) becomes negative modifier for rolls. Net effect: Modifier = (G - DS) - (L + DS - G) = 2G - L. In other words, if G < L/2, a net loss.
G > (L + DS): Luck pool emptied, no modifiers.

For the gamble to be effective, you need to gamble between max(L/2, DS) and (L + DS).

Conclusion: Players should gamble low if the scene is exciting & the GM is stingy. They should gamble high if the scene is boring & the GM is generous.

Questions: How does the GM decide what values to assign? How does he know if a scene is dramatically significant? How does he decide whether to make lots of luck available?

Matthijs,

DS is typically tied into how far into a particular storyline/quest/adventure the characters have gone. By this I mean the amount of sessions that have been roleplayed for the current storyline. Every partipant is made aware that there will be X number of gaming sessions for a storyline. As those sessions pass by the DS for scenes will get lower and lower to signify the characters getting closer and closer to the final climatic scene.

Let's say that a storyline will consist of three sessions. The first session may have DS scenes of 7-10, the second 4-6, and the last session would have DS from 1-3.

Further consideration is given to whether the scene furthers the emerging storyline or delves into the premise being explored. A good rule of thumb for any GM assigning a DS to a scene would be to consider how the scene might written in a novel. Would an author spend a lot of text on a scene of this nature in the context of the story being written. If the answer is no then it's probably a good idea to give the scene a high DS. Conversely if the GM can picture an author devoting a considerable amount of text to the scene it makes sense to give the scene a lower DS.

To go back to our three session game mentioned previously... It's the second session and the characters are in a scene in which they have to buy new clothing. The GM knows that the typical range of DS for a scene in the second session will be from 4-6. But the exact DS will depend upon the scene's context within the emerging storyline and the premise being explored. In this case the scene has very little to do with the premise of whether you should forgive those who have betrayed you. However it does fit in with the emerging storyline of trying to gain entry into palace grounds (the characters will need to be dressed in expensive and fashionable attire if the hope to not be thrown out or arrested). Because the scene is significant to the story,  but not significant to the premise, he opts for a DS of 5, which is half way between the typical range for a scene's DS within the second session of the current storyline.

The actual assigning of DS is somewhat arbitrary, but it not so arbitrary that players shouldn't be able to have a somewhat reasonable notion of what DS a GM will assign to any given scene and its no more arbitrary than any time a GM assigns a difficulty to a certain conflict or task. Plus GMs are more than welcome to drop hints as to whether a scene will have a high or low.

As for the assigning of luck, that itself becomes a guessing game between the players and the GM. The players can guess more successfully if the recognize any habits or patterns the GM follows for the assigning of luck. Moreover players know that the GM must make enough luck available for characters to successfully resolve all plausibly* resolvable conflicts. That doesn't mean that all conflicts can be resolved by relying solely on luck. Players will still need to decide which qualities and virtues (attributes) to use and even then they need to decide which characters will attempt to resolve the conflict. The GM only needs to make it possible for the conflict to be successfully resolved. He or she does not need to make it so that a conflict can be successfully resolved by any character within a scene or by any means or method elected by the player.

To return to the previous example of buying new clothes... the GM has assigned a DS of 5 to the scene. Based upon the kind of conflicts he anticipates for the scene he decides to give the scene enough luck so that the the smartest character will be able to haggle down the price of any purchases. The players don't know whether it would be better to use the smartest or the most confident character to accomplish this particular goal. But they do know it can be done and they know that the more difficult the conflict is to resolve in a manner that favours the PC's the greater amount of luck will be assigned to that particular scene. In the case of haggling down the price of new and expensive clothing, that's difficult enough for the players to assume quite a high amount of luck having been assigned to the scene. In this case the GM gives the scene a luck of 8 as the PC's will need to have quite a high modifier if their going to hope to be able to get new and expensive clothing at less than luxurious prices.

*By plausibly resolvable I mean that the conflict is not outside what the playing group thinks as being possible to accomplish within the game itself. A superhero style game will have different requirements for plausibility than a more prosaic style of game.

You're going for a lot of GM fiat here. The GM controls everything, and the players are left guessing. Have you considered giving players more control?

What do you think would happen if the players actually knew the DS for each scene, and the GM had a set number of Luck that he had to distribute between the scenes? What would happen if the players participated in setting the DS?

In fact, would it be possible in some way to retroactively determine the DS? After a scene, the group votes on how significant the scene was - and then, they get luck points based on what they gambled in that scene?