Dice mechanics are the core of any RPG system. This is a subject that has been discussed at incredible length in other places.
What it all boils down to in the end is how much do you want randomness to play a part and to what degree.
One of the things I hate about d20, the big thing really, is that randomness is the defining feature of any action or you have placed enough resources into this particular action that randomness is functionally no longer a factor at all.
The problem is that any given result on any die on any individual roll is as likely as any other. This problem is exacerbated by the size of the die in question, the larger the die the more completely unmanageable the results. d20 is bad, Chaosium’s percentile system is worse.
This can be mitigated by rolling more than one die, such as BASH!’s 2d6 (which is then multiplied) or GURPS’ 3d6, or even White Wolf’s or ShadowRun’s “Fist Full O’ Dice”.
The second group can be divided further into additive systems or success based systems. They both have advantages, they are both superior (in my opinion) to a straight roll.
I’m leaning toward additive but I’ll examine success based to make sure it doesn’t fit my purposes better. One of the biggest problems with a success based game depends on a system of critical hits and misses, I actually can’t think of any success based games I’ve ever played that haven’t had such a system, so it might not be so bad. In most of these games a 1 cancels a success or in some way causes something “bad” to happen this makes being unskilled at anything extremely dangerous as a 1 is a bit less common than whatever a success is (4-6 in Shadorun or 7-10 in WW) but if you roll dice often enough still sure to happen frequently. If we eliminate that we end up rolling our fist full of dice and parsing for successes, we can do that, but if we remove that success canceling mechanic we are going to end up with virtually guaranteed success pretty easily, it briefly looked like a good idea but I’ve returned to my original position that it’s not. We’ll be going with an additive system.
So do I want a fixed dice pool or do I want to have skill add dice? I’m pretty comfortable with both actually. With a fixed Dice pool like GURPS’ 3d6 you will always get a result between 3 and 18 with results being significantly more common around 10.5 and 3 and 18 being memorable events, worthy of song and legend. This is pretty realistic but it’s not very dynamic and I think goes a bit to far in the not random enough direction. It’s a delicate balance. Whereas if skill adds dice you will get a result between 1 and “who knows”, I like more predictability than that, I guess. I also think I may have problems balancing Adding Dice later on.
One of the things I like about BASH! is that it is a fast, light, easy to run, easy to play system. That gets bogged down a fair bit the more dice you roll and if you have to parse the results of the die roll for successes, so I think I’m going to keep the number of dice to two and keep it additive.
Do I want to keep the dice a constant where I always roll 2d##, and add the result or do I want to go with a system more like the wildly popular Savage Worlds with die stepping. with greater levels of skill offering a larger die to roll and add to your result.
Die stepping kind of defeats the advantage of multi die rolling. Typically the better you are at something the bigger the die you get to roll allowing you to roll higher numbers, that’s the theory anyway. In practice what happens is the better you are at something the more unpredictable your results become. That doesn’t make sense, it should be exactly the opposite. If you are going to do a Die Stepping system you should step the dice down from say a rank amateur with a d12 or d20 to an expert with a d4 there are two two ways to handle this one is to make up the difference between the larger die and the smaller die with an add, so you would step, for example, from a d12 to a d10+2, or you could score it like golf where a lower number is better, the add is somewhat clunky the golf score is counter intuitive. I suppose die stepping could be done well, I’m not positive it is worth the effort to try.
So we are going with a pair of fixed dice. Which ones?I like d12s the geekiest of all polyhedrons.
Let’s run the numbers on that. There are 144 possible results on 2d12, that’s an awful lot. 1 possible result each of 2 and 24, 4 results for 3 and 23, and so on, The median result is 13 with 12 possible results for 13 78 possible results are 13 or better. geeky as it is I think we are going to have to give the d12 a pass, sorry big guy.
OK the d12 was a complete bust, let’s rein our theorycrafting back in a bit and shoot for the d6, everybody has some. It has 36 possible results with 7 being the most common having six results that total 7, 1 one result each that totals either 12 or 2. this is starting to look a bit like a more stable d20, it’s familiar, I think people could be pretty comfortable with this, I am. It feels kinda nice.
Now we have some dice to roll and have decided that we are going to add them together now what?
The core mechanic of the d20 system is that you roll a d20 add modifiers and compare it to a difficulty. It’s simple it works on a fundamental level, I think we can easily adapt it to our purposes. I want to create a nice simple easy to play game, but I’m not intending to reinvent the wheel.
Now that we have our coremechanic of roll 2d6 add modifiers and compare to difficulty we can make meaningful progress with the rest of the game.