Luck versus skill: Let us say we set a master up against a novice. Should the novice have any chance at all, however remote, of beating the master? In other words, shall an unskilled character have a small chance of succeeding at an extremely difficult task, and shall a master have a small chance of failing at a routine task?
The luck versus skill ratio may depend on the task in question -- some tasks are inherently more random than others, such as Poker versus Chess.
I think this is a key question to ask whenever designing a system. And no answer is perfect. Gamers typically like their novice PCs to have a chance against better opponents, but dislike it when their better PCs lose to novice NPCs.
Rather than say that poker has a higher ratio of luck to skill than chess, think of it in terms of the odds of success. In poker, a +1 in skill shifts the odds of success less than it does in chess. But it may also matter what the skill levels are in the first place, what the scale is, and exactly which two games we are talking about.
Quote:
On a related note, the amount of random variability in a task may depend on ability. In the "real world", you would expect highly skilled persons to be more consistent (and, hence, less random) than unskilled dabblers, but, as we shall see, this is not true in all systems.
The amount of a "random variability" in a task is often an illusion of the dice and doesn't really translate into actual game outcomes. For example, many people feel that a d20 is "more random" than a 3d6. Statistically, this is correct. But for any given task in a game when you are trying to roll a specific value or better (or worse), neither is more random. For example, you are more likely to roll an 11 on 3d6 than 1d20. But rolling an 11 or better is exactly 50% for both 3d6 and 1d20. How modifiers affect the roll needed is different, but once you know what you need to roll, you can't say that 3d6 is any less random than 1d20. If you have a 25% of succeeding, it doesn't matter what dice you use.
However, the choice of dice does go back to the first point. Skill matters more when you use 3d6 than when you use 1d20. This is not really a function of the "randomness" but how modifiers affect the roll. A +1 is always +5% on 1d20, but can be anywhere from 0.463% to 12.5% on 3d6. All else being equal, having an edge of +1 matters much more on 3d6 than it does on 1d20. Of course, how the modifiers is a function of the probability distribution curve of 3d6 vs. 1d20 and this curve also determines the "randomness."
I really do not know what this article is trying to achieve. It starts out talking about the dice systems and statistics but moves into role playing circumstances and the desired effects-which appears to be realism-of creating a rule system. I have found that the use of a random system is just an aspect of trying to achieve some action within a collective game. Therefore, what really is being address is how do you determine an outcome. You can not look at random statistics until you couple it with action. By combining both; then analyzing all aspects you can determine what type of statistics you want within a game. Here is my analysis of combining random solution with action:
Using fortune- there are only the following aspects that can change when using fortune
1. Chance of success-this is the straight up percentage and all systems with a random element has this. Naturally, all of this turns into statistical data and some sort of percentage. If you have to roll 7 out of 10 on three dice from your pool there still is a mathematical chance or percentage. One point on a 20D is just 5%. Furthermore, all combination of values would be included to calculate the odds. This covers the question within various encounters such as; do you harm an opponent? Can you repair the robot? Can you detect a trap? Can you persuade the captain?
2. Degree of success-So you succeeded and accomplished something. But how well did you do or are doing it becomes the next question. Many times this is just a yes/no answer and any measure or degree of success is not important. Finding a secret door is a yes/no success, you can not find more doors from a very successful roll nor does the door become any more of a door. However, if something can be measured within a system, then the issue of degree of success comes into factor. Healing and Damage all have some sort of point system so therefore if you harm an opponent you have to roll damage which measures how well you harmed them. If you are healing you roll how much health has been returned. It is interesting because within most games the real system mechanics start off having chance and degree completely separate from each other. Only through some additional rules is there some influence of chance on degree. Critical hits are an example. Additionally, there are some actions where there are degrees that can not be measured within the system but can exist within a story. How much information does a NPC know and how much does he tell you? Would each bit of information require a roll or is there a degree to how personal will he tell. These subjective results are very important. Actually, the degree of success is much more important to most players than the chance success roll.
3. Range of success-Quite often the range of success is overlooked within many systems. Think of range of success as multiple successes or “freebie”. Degree measures how much while range covers what else. They bring in additional benefits to successful rolls. A simple example is while repairing a robot you find a hidden message from the princess or while searching for the passage in the dark you find a ring. While looking for weapons of mass destruction you find a civil war, okay not that. For many subjective skill rolls this is a treasured effect for it helps moves the plot along while players gain more benefits to their choices and actions. However this is not to say that there are no game systems where additional effects are not written into the rules. Anything that Knocks outs, stuns, and tripping are ranged of success examples. Any system that increases skills based on successes is including range.
4. Resources expended- Anything that cost to implement the role. Actions in combat, bullets, loss of dice, etc are all apart of resources. However, the cost of failure if included in the system is also resource. So such things as fumbles or botches are also apart of resources. If a roll of one on a D20 causes a fumble, then 5% of the time potential actions will cause more problems. Overall, resources can fall into two categories, implemented cost & effect cost. Implemented cost, are things which you need to expended to start with, dice from pools, actions used, equipment and so on. Effect cost, do not happen as often but are expended after the results have been determined.
5. Invested stakes- Most RPG’s do not included this within their game systems, but like poker the player invest additional resources beyond what is expected in the hopes that winning will gain a greater reward. The player is willing to put at risk a certain value demonstrating what they are willing to win.
So what does this mean? Well here are five aspects to any decision or resolution system. As a game master you have to judge and rule on the results and effects of these decisions. When creating items or adding rules, these are the areas in which you will modify. As a player, these are the areas which will help you make your choices. As a designer this is what you have to contend with where your rules will have to go.
Let’s take a look at combat for D20 system (D&D) First, chance of success: Based on a percentile with modifications. You have a large amount of rules governing the basic “to hit someone” covering “realism” rules along with some mega-gaming adaptation. The core odds are based on the character skill level with modifiers based on circumstance. There are few actions or choices that the character can make that will affect the attack. In other words, the action that is declared is “the attack.” The “Chance of Success” results in either a yes or no results. There are only a few areas or optional rules where the chance influences the degree or range of success. Second, is “Degree of Success” which, once again, a random chance with modifiers. This just measures the amount of damage done verse the hit points. The core odds for this statistic are based on the choice of weapon and abilities all relating from the character creation. Once again, there are few mega-game rules or modifiers. Third, is “Range of Success” there is none. Any action that is broaden to include things such as knock outs, disarmed, tripped, etc have to be declared before the attack and then different additional rule systems are used which goes back to the yes/no covered by “Chance of Success.” The final aspect, resources is highly complex and can influence all areas of play. The move/action-full/partial actions, creates a series of choices throughout combat. Additionally, there are many mega-game rules that are tacked on moving some resources back to character generation and advancement. There are also optional rules for fumbling and the like which move into additional cost to resources. There is nothing for the fifth aspect.
Last edited by Phlophouse; 06-17-2007 at 11:12 AM..
I think this is a key question to ask whenever designing a system. And no answer is perfect. Gamers typically like their novice PCs to have a chance against better opponents, but dislike it when their better PCs lose to novice NPCs.
Indeed. And you might use this observation to make a non-symmetric systems: PCs can (in rare circumstances) achieve far above their norm, but most NPCs can not. In the extreme, only PCs roll, while NPCs always get the average result for their ability.
[QUOTE]Rather than say that poker has a higher ratio of luck to skill than chess, think of it in terms of the odds of success. In poker, a +1 in skill shifts the odds of success less than it does in chess. But it may also matter what the skill levels are in the first place, what the scale is, and exactly which two games we are talking about./QUOTE]
Quite so. You could define a +1 increment in skill as something that will give you, say, 5% extra chance of success against the same opponent. But then you would have to consider whether the cost of skill increases (in terms of time spent in training or studying) should be the same for all skills. And you would still have cases where no amount of skill will help against bad luck.
In any case, my aim was not to say that your system needs to handle degree of variability, just that you should consider if you want to include it in your system. Later, I will come with some suggestions of how to do it, but the are just that: suggestions.
I really do not know what this article is trying to achieve. It starts out talking about the dice systems and statistics but moves into role playing circumstances and the desired effects-which appears to be realism-of creating a rule system.
These are not independent. A dice-roll method is an integrated part of a rule system, so it i worthwhile to consider how the choice of dice-rolls may help or hinder what you want to achieve with a rule system.
(The following quoted text is abbreviated a bit)
Quote:
I have found that the use of a random system is just an aspect of trying to achieve some action within a collective game. Therefore, what really is being address is how do you determine an outcome. You can not look at random statistics until you couple it with action. By combining both; then analyzing all aspects you can determine what type of statistics you want within a game. Here is my analysis of combining random solution with action:
1. Chance of success-this is the straight up percentage and all systems with a random element has this.
2. Degree of success-So you succeeded and accomplished something. But how well did you do or are doing it becomes the next question.
3. Range of success-Quite often the range of success is overlooked within many systems. Think of range of success as multiple successes or “freebie”. Degree measures how much while range covers what else.
4. Resources expended- Anything that cost to implement the role.
5. Invested stakes- the player invest additional resources beyond what is expected in the hopes that winning will gain a greater reward. The player is willing to put at risk a certain value demonstrating what they are willing to win.
All these are very good points. I'm not sure I would say that degree and range of success are all that different -- exceptional success or failure (or partial success) often produce side effects, and most side effects result from nonstandard success degrees. The resource and stake issues are quite interesting, though.
The point that I am trying to make, is that the article is only considering one aspect of action resolution, what I would consider the “chance of success.” Skills, circumstances, difficulty are all just modifiers to the overall “chance.” You mention degree of success, but only qualify it with a “yes or no” with two extremes of “critical” and “fumble,” which seems to be stressing the D20 system. As I pointed out, the D20 system has two mechanics mostly separated from each other-the “chance of success” is the “to hit roll” with the “degree of success” being a separate damage roll. There is only a 10% chance that the “chance of success” will affect the “degree of success,” by rolling a natural one or twenty.
If you are going to design a game system, you have to consider all the aspects of action resolution. So you do not get bogged down in only one aspect- why go into a huge mathematical equation where the affects end up only changing the chance of success by 5-10%. As a GM, if you are aware all the aspects, you can modify and reward actions in different ways other than a bonus to getting a “yes” answer. For example, PCs are hiding up on a cliff where all they have to do is club anyone who walks by. Should they get a bonus to attack? Bonus to damage? Bonus affect such as a knockout? Will they have to invest their whole action? Can they do it without being noticed by others? If I was playing the D20, I would not want to argue about getting a better chance to hit, but rather get a bonus of a knockout or damage. Because, I could hit well but still roll a one for damage. However, if I was playing Riddle of Steel, I would argue for a better chance to hit, because for each success I get from hitting affects how much damage I do. You see all these aspect of action resolution; I believe influence each other, unless as a designer you make a point to separate them.
Linear Mode
