TooManySecrets

2011-07-04, 04:39 PM

So, I've been thinking a lot about making an RPG system recently. Its mostly accumulation of thoughts and half-formed homebrew, so nothing really concrete. Because of that, I'm going to be posting my thoughts and in-progress stuff here for discussion instead of posting it in Homebrew (I really dislike when people post WIP in Homebrew).

My philosophy towards roleplaying games are summed up below:

The goal of any game is to have fun.

In order to have fun, a player must be able to make meaningful decisions.

The GM and system should treat the players maturely.

I don't know where the RPG itself is going to end up going, but I still have some general goals for it:

Playable on forums. It's where I do most of my gaming so any system I can't use of a forum is pretty much useless to me.

Marginal return on effects e.g. increasing a skill from +1 to +2 doesn't increase your chance of success as much as increasing from +0 to +1

Classless. Classes have a lot going for them in terms of making sure the player keeps to a strong archetype, but that very strength is a weakness when it comes to players trying to do something the original game designer never imagined.

No or minimal experience points or equivalent. I really dislike experience points and leveling and all that jazz.

Some sort of fatigue system.

Optimization of statistical power is channeled into fluff-appropriate things. For instance, many GMs have a problem with the concept of pacts in Fiendish Codex II since it allows characters to be more powerful in a rather easy way. I, on the other hand, think they're great since the character is getting power by making pacts with devils. Stuff like that has tons of storytelling potential and is, in my opinion, far better optimization than picking a number of disparate prestige classes in the correct order to maximize base attack bonus and saves.

Enough meandering! The first topic I'm going to talk about is

DICE

Choosing what dice to use is probably the most important decision to make when designing a game. Actions in a game are all about determining success or failure and the way you determine that is by using dice.1 Some games try to get around this by using things like cards or spinners or other such paraphernalia but when you get down to it all of those things are functionally identical to dice (plus, remember Goal 1). There are a lot of different ways of using dice and I've listed just some below.

Linear - You roll a die, perhaps you add or subtract something, and then you compare it to a number. D&D has a d20 that you add modifiers to and compares it against a d20. Call of Cthulhu has a d100 that you try to roll under a certain number. Unknown Armies is similar to Call of Cthulhu, with the added complication that, for instance, a 54 can produce a different result than a 45 even if both are under the skill of 60. I won't go into deep details, but basically no matter whether you're adding or subtracting or rolling under or rolling above, all of these style of rolling have the same probability curve i.e. a linear one. Increasing your skill by 1 always increases your chance of success by 1% (or 5% or whatever). Given Goal 2, this style is right out.

Dice Pools - You roll a certain number of dice. Any dice above a certain number - usually called a threshold - is a "success" and any number below a certain number is a "failure". Shadowrun uses d6s. The most recent edition of Shadowrun has a die be a success on a 5 or 6 (failure otherwise) though earlier editions had the number be set at pretty much anything (die rolls of 6 could be rerolled and added to the result, allowing for thresholds above 6). The World of Darkness uses d10s with a threshold which is set by the task. Both of those systems introduce critical success or failure using semi-complicated rules.

Assuming that you are trying to hit a certain number of successes and as many as possible, e.g. "Roll at least 2 successes" vs. "If you roll more successes, you get better results", than there is a marginal return on adding more dice. That's what I'm looking for. The problem is with opposed rolls, which are almost always "roll as many successes as you can" as that turns it right back into linear.

Dice Poker - Alright, this is a weird one. In a dice poker system, you roll a bunch of dice. You then search for sets, e.g. "I got a pair of 5s", and then select one. The "height", i.e. what the number of the die shows, determines whether you succeed or not and the "width", i.e. how many of that number, determines the margin of success. Godlike (which I haven't played) is the only game I know which does this, but I'm sure there is more. The advantage of this system is that it's non-linear and that one roll shows success and an independent margin of success. Interesting, but having the player pick out the set they want is something I find vaguely wrong for reasons I can't properly articulate and I have a feeling the style might scare away some players.

Attribute Determines Die - This system is similar to linear - you roll a die and compare the result to a number. The difference is that the die that you roll is different depending on your skill e.g. a novice might roll a 1d4 while an expert might roll a 1d20 (or vice versa if it's a roll-under instead of roll-over). Serenity (which I also haven't played) does something like this. I find this more interesting than strictly linear and by fiddling around with dice size you can make it non-linear.

Median of Three - This is sort of similar to attribute determines die. Basically, you have three die. The size of the first die is set by ability score2, the size of the second die is set by skill rank 2, and the third die is just always the same (or perhaps set by the task). Roll the die and get the average result e.g. 6, 7, and 1 gives 4.66. This gives a lot of interesting probability qualities, not to mention separating ability and skill.

Highest Die - In this system, you roll a bunch of die and the highest dice result is your score. If it's above the threshold, it's a success. Failure otherwise. This is what I'm currently leaning towards. It's nonlinear and it's easy to understand (since most people have experience with dice pools and this is a simple modification of that). The main problem with this system is that results are bound by the selection of the die, e.g. if I choose a d20 nobody is getting a result higher than 20.

One of the modifications of this is 2nd-Highest Die (or 3rd-Highest, etc). This slows down the rate of return on increasing an attribute as well as decreasing the minimum chance. For instance, on a Highest Die system with a d20, a novice has a 5% chance of getting the best result (rolls a 20). On a 2nd-Highest Die system with 2d20, a novice has only a 0.25% chance of getting the best result (rolls 2 20s). Of course, this also means that I'll have much higher attributes, e.g. with Highest Die an attribute of 5 or so isn't bad but with 2nd-Highest Die I would need an attribute of ~14 or so instead.

Finally, Highest Die (and variants) have the same problem that all non-linear dice rolling has, namely that the probabilities are non-obvious to the average player. In D&D, I know that I have a 50% chance of succeeding on a DC of 14 on a 1d20+4. It takes longer to calculate what my chance of success on a 2nd-Highest Die with 5 dice on a threshold of 11 is.

1 Technically, there are diceless games such as Amber Diceless. While I do want to do something like that eventually, let's just pretend that it doesn't exist for right now.

2 To use D&D parlance.

Thoughts? Comments? What system of dice rolling have you used and what sort do you like?

My philosophy towards roleplaying games are summed up below:

The goal of any game is to have fun.

In order to have fun, a player must be able to make meaningful decisions.

The GM and system should treat the players maturely.

I don't know where the RPG itself is going to end up going, but I still have some general goals for it:

Playable on forums. It's where I do most of my gaming so any system I can't use of a forum is pretty much useless to me.

Marginal return on effects e.g. increasing a skill from +1 to +2 doesn't increase your chance of success as much as increasing from +0 to +1

Classless. Classes have a lot going for them in terms of making sure the player keeps to a strong archetype, but that very strength is a weakness when it comes to players trying to do something the original game designer never imagined.

No or minimal experience points or equivalent. I really dislike experience points and leveling and all that jazz.

Some sort of fatigue system.

Optimization of statistical power is channeled into fluff-appropriate things. For instance, many GMs have a problem with the concept of pacts in Fiendish Codex II since it allows characters to be more powerful in a rather easy way. I, on the other hand, think they're great since the character is getting power by making pacts with devils. Stuff like that has tons of storytelling potential and is, in my opinion, far better optimization than picking a number of disparate prestige classes in the correct order to maximize base attack bonus and saves.

Enough meandering! The first topic I'm going to talk about is

DICE

Choosing what dice to use is probably the most important decision to make when designing a game. Actions in a game are all about determining success or failure and the way you determine that is by using dice.1 Some games try to get around this by using things like cards or spinners or other such paraphernalia but when you get down to it all of those things are functionally identical to dice (plus, remember Goal 1). There are a lot of different ways of using dice and I've listed just some below.

Linear - You roll a die, perhaps you add or subtract something, and then you compare it to a number. D&D has a d20 that you add modifiers to and compares it against a d20. Call of Cthulhu has a d100 that you try to roll under a certain number. Unknown Armies is similar to Call of Cthulhu, with the added complication that, for instance, a 54 can produce a different result than a 45 even if both are under the skill of 60. I won't go into deep details, but basically no matter whether you're adding or subtracting or rolling under or rolling above, all of these style of rolling have the same probability curve i.e. a linear one. Increasing your skill by 1 always increases your chance of success by 1% (or 5% or whatever). Given Goal 2, this style is right out.

Dice Pools - You roll a certain number of dice. Any dice above a certain number - usually called a threshold - is a "success" and any number below a certain number is a "failure". Shadowrun uses d6s. The most recent edition of Shadowrun has a die be a success on a 5 or 6 (failure otherwise) though earlier editions had the number be set at pretty much anything (die rolls of 6 could be rerolled and added to the result, allowing for thresholds above 6). The World of Darkness uses d10s with a threshold which is set by the task. Both of those systems introduce critical success or failure using semi-complicated rules.

Assuming that you are trying to hit a certain number of successes and as many as possible, e.g. "Roll at least 2 successes" vs. "If you roll more successes, you get better results", than there is a marginal return on adding more dice. That's what I'm looking for. The problem is with opposed rolls, which are almost always "roll as many successes as you can" as that turns it right back into linear.

Dice Poker - Alright, this is a weird one. In a dice poker system, you roll a bunch of dice. You then search for sets, e.g. "I got a pair of 5s", and then select one. The "height", i.e. what the number of the die shows, determines whether you succeed or not and the "width", i.e. how many of that number, determines the margin of success. Godlike (which I haven't played) is the only game I know which does this, but I'm sure there is more. The advantage of this system is that it's non-linear and that one roll shows success and an independent margin of success. Interesting, but having the player pick out the set they want is something I find vaguely wrong for reasons I can't properly articulate and I have a feeling the style might scare away some players.

Attribute Determines Die - This system is similar to linear - you roll a die and compare the result to a number. The difference is that the die that you roll is different depending on your skill e.g. a novice might roll a 1d4 while an expert might roll a 1d20 (or vice versa if it's a roll-under instead of roll-over). Serenity (which I also haven't played) does something like this. I find this more interesting than strictly linear and by fiddling around with dice size you can make it non-linear.

Median of Three - This is sort of similar to attribute determines die. Basically, you have three die. The size of the first die is set by ability score2, the size of the second die is set by skill rank 2, and the third die is just always the same (or perhaps set by the task). Roll the die and get the average result e.g. 6, 7, and 1 gives 4.66. This gives a lot of interesting probability qualities, not to mention separating ability and skill.

Highest Die - In this system, you roll a bunch of die and the highest dice result is your score. If it's above the threshold, it's a success. Failure otherwise. This is what I'm currently leaning towards. It's nonlinear and it's easy to understand (since most people have experience with dice pools and this is a simple modification of that). The main problem with this system is that results are bound by the selection of the die, e.g. if I choose a d20 nobody is getting a result higher than 20.

One of the modifications of this is 2nd-Highest Die (or 3rd-Highest, etc). This slows down the rate of return on increasing an attribute as well as decreasing the minimum chance. For instance, on a Highest Die system with a d20, a novice has a 5% chance of getting the best result (rolls a 20). On a 2nd-Highest Die system with 2d20, a novice has only a 0.25% chance of getting the best result (rolls 2 20s). Of course, this also means that I'll have much higher attributes, e.g. with Highest Die an attribute of 5 or so isn't bad but with 2nd-Highest Die I would need an attribute of ~14 or so instead.

Finally, Highest Die (and variants) have the same problem that all non-linear dice rolling has, namely that the probabilities are non-obvious to the average player. In D&D, I know that I have a 50% chance of succeeding on a DC of 14 on a 1d20+4. It takes longer to calculate what my chance of success on a 2nd-Highest Die with 5 dice on a threshold of 11 is.

1 Technically, there are diceless games such as Amber Diceless. While I do want to do something like that eventually, let's just pretend that it doesn't exist for right now.

2 To use D&D parlance.

Thoughts? Comments? What system of dice rolling have you used and what sort do you like?