First a question on some general thoughts on system design, do you think it's better to design a system and then build a setting around it? Or build a setting and design a system around it?

That said I was wondering if anyone would be able to give me their thoughts on some general ideas I had for a system.

I was thinking that I kind of dislike single die based systems. The results of a single die roll are a bit to swingy for my tastes. The range of numbers given by a d20 suits me well enough though, enough room is given for a bit of granularity.

So what I was thinking was this, everything works as a roll over check, combat and other opposed checks work on an opposed roll. So the attacker and defender would both roll, highest wins. Nothing remotely difficult for me to work out so far.

However due to my dislike for the up and down tendency of single die rolls I was thinking things would work like this. PC's and NPC's that are skilled in an area roll 2d10 instead of 1d20, whilst unskilled characters roll 1d20. So the baseline level of skill would allow you to roll 2d10, further ranks of skill after that would add a flat numerical bonus to the 2d10 roll.

So a basic scenario would work like this; A attacks B, so A being skilled with his weapon rolls 2d10 adding any relevant attribute or situation modifiers, B being unskilled with any weapon rolls 1d20 adding any relevant modifier. A has rolled higher than B so B is hit.

Are there any potential pitfalls to this approach? I know that it will tend toward skilled characters producing an average roll around 11 or so before modifiers, each successive rank in a skill should push that average up right?

Another thing I wanted to do was account for degree of success, particularly for combat as I was toying with the idea of having damage being tied to it. Off the top of my head it seems like you could count the difference between rolls to find the degree, maybe going be 5's being a level of success? Is there any quick and easy way to do this without having to come up with charts and the like?

I say give them more options on how to handle defense.  Say B is untrained with weapons, but is a skillful dodger so gets 2d20 to defense if player B chooses to dodge the attack while it be 1d20 if the player chooses to parry the attack.

Yeah that could work, I mean if they have a relevant skill or ability to the situation then there is no reason for them not to use it.

I'm not crazy about the 1d20 unskilled vs. 2d10 skilled concept.  The average of both rolls is roughly the same (10.5 for 1d20 vs. 11 for 2d10), but, as you mentioned, 1d20 is a lot swingier.  While this makes the unskilled guy a lot more likely to fumble than the skilled guy, he's also a lot more likely to succeed spectacularly (rolling a 20 is 5 times as likely on 1d20 (5%) as it is on 2d10 (1%)).

This doesn't seem like it provides any significant benefit to being skilled, or at least not until you start getting bonuses for being highly skilled.  (2d10+2 is where the skilled guy becomes more likely to roll 20+ than the unskilled guy is to roll 20.)  Modifiers also affect each of them differently; depending on the typical modifiers to rolls, either one could actually be better overall.

Yeah the more I thought about the idea the more it broke down. I think if I am going to have the difference between skilled and unskilled have anything to do with the number of dice an actual dice pool makes more sense. Besides it seems like degree of success mechanics are easier to pull off with a dice pool.

I don't like the fact that off the bat the unskilled person is somewhat more likely to roll a 20.

I guess it's back to the drawing board I go.

Actually thinking further about the whole 2d10 for skilled rolls vs 1d20 for unskilled rolls, would it work better if skilled rolls were 2d20 keeping the highest and discarding the lowest?

That would fix the problem of the unskilled being slightly more likely to roll a critical, at least I think it would.

Quote from: Arkansan;821982Actually thinking further about the whole 2d10 for skilled rolls vs 1d20 for unskilled rolls, would it work better if skilled rolls were 2d20 keeping the highest and discarding the lowest?

That would fix the problem of the unskilled being slightly more likely to roll a critical, at least I think it would.

it would certainly fix the problem
roll two dice keep highest basically doubles the odds of succeeding

the function I believe is  F= 1 - [(1-x)^2] where F is the chance of overall success rolling two dice and picking best and x the basic chance of success if you just rolled one dice.

there are games that use this system just can't think of them off the top of my head.

I don't care for swingy linear probability so use a 2D10 add together approach myself

Yeah that is what I did with sniper rifles for my custom game known as Struggle of the Moirai.  Sniper rifles are 6d5 instead of 3d10 because the odds are better.  We are speaking about a damage range of 6 to 30 instead of 3 to 30.  Sure not much, but I figure with some thing as accurate as a sniper rifle it should do the most damage as it can do.  Not to mention if you did roll horrible the least you can do is six damage which is double the least amount of damage of 3d10.

I was revising this project last night and had a thought on how to resolve the issue of the unskilled character having a better chance to critical.

Instead of having skilled characters roll 2d20 and keep the highest, which I feel gives them a bit too great an odd of success why not keep the original idea of having them roll 2d10 and change how critical hits work. I was thinking that if an unskilled character rolls 1d20 and hits 20 then instead of automatically hitting the critical they have to roll to hit once again and succeed again to crit. If they fail the roll just acts as it normally would. Skilled characters would just automatically crit on 20.

I would think this would fix the problem of unskilled characters being more prone to critical hits.

Oh well 20 on 2d10 = 1% chance.
Success on 20, rolling to confirm could still be more. Say you're 50% likely to succeed again, then you'd have a 2.5% chance.

You could however just have skilled characters critical on say 18-20 (6%).
J Arcane had an interesting system where he could modify crit chance specifically, by reading the 2d10 as d100 for crit purposes.