d16 for ability tests? Talk me out of this idea.

[bholmes4]

I've always used the "roll under d20" system for ability tests but something about the probabilities has always bugged me. To me, a score of 18 should be a 20 on that scale and a score of 3 should be a 1, in my opinion. I've always accepted it as a quirk of the system and a limitation of the dice. Recently I realized that the ability point spread (for rolls of 3-18) add up to a total of 16 points. Then it occurred to me, why not use a d16 to simulate this system?

You could quickly map a base modifier of +2* to the roll such that a score of 18 succeeds on a roll of 15 or less on a d16, a 17 on a roll of 14 or less, 16 on a 13 or less and so on, until you reach a score of 3 which only succeeds on a roll of 1. Under this sytem, no longer will that that 18 strength warrior be shown up in a dwarf-throwing contest by a crippled, 80 year old woman**.

Something in my gut is preventing me from adding it as a house rule though. Talk me out of this idea.

* Ideally you would just have custom d16 dice created with scores from 3 to 18 but that's a pipe dream at this point.
** Ok it could happen in this system too but it would be really, really rare...

[TheNobleDrake]

What would you do in a situation where someone has a 1 or 2 in their ability score? Would that be auto-fail territory?

Also, your numbers quoted are off by 1 unless you were meaning to also have a "natural 16 auto-fail" rule and forgot to mention it (16+2 = 18, which since you say 1+2 = 3 is a success should be a success as well).

Personally, I find the roll under your score method to be a bit limiting - simple tasks become significantly more difficult for a person despite their score still being within the average range... let me expand on that:

If you roll for stats, which I do, then you have to allow a wider range of scores being considered "average" or "good enough" - that's why 9-12 are all lacking in modifiers instead of the trend that only 10 & 11 have no modifier that started with D&D 3rd edition (a system which penalizes rolling for scores as opposed to simply using point buy because of it's narrowed modifier brackets).

In one system - using set DCs and rolling to beat them - a score of 9 (average) and 12 (also still average) have the same chance of success at the same difficulty of task. Example: both have a 55% chance of succeeding a DC 10 check with a d20

In the other system - roll under score - a score of 9 (average) and 12 (also average) have different chances of success. With a d20 for the check that chance is 45% for the 9 and 60% for the 12, giving one supposedly average score 1.33 times greater chance of success than another average score. Changing that to a d16+2 check method gives the 9 a 43.75% chance of success and the 12 a 62.5% chance, exaggerating the difference to a 12 being 1.43 times more likely to succeed.

If that - accentuating the difference between ability scores while also make the already good better and the already low worse - is your goal, then you have a good system in mind for it. Otherwise, I believe you should reconsider.

Now for the devil's advocate portion: One place where I believe that the roll under a score method worked was in later 2nd edition AD&D materials. The difference that set things in a better light was to slightly detach your base capability in an area from your ability score - the skill (non-weapon proficiency, if you prefer) had a base rating that represented how difficult a task using that skill typically was, your related ability score had a modifier (pretty close to the current modifiers in DCC) that would alter that base rating, and you could also invest some of your earned points upon gaining a level to increase the rating - but you never ended up in a situation where "average roll" for ability score didn't also mean "average competency for level of investment" in a skill.

That system, however, might be a little too complex and simultaneously limiting considering it only works if every possible choice of skill has a default base rating (meaning a finite, even if ever-expanding, skill list would be in order).

[bitflipr]

If I were to veer away from DCC's DC-based d20 ability test I'd roll 3d6 instead.

[finarvyn]

What would you do in a situation where someone has a 1 or 2 in their ability score? Would that be auto-fail territory?

I tend to treat any scores lower than a 3 the same as a 3.

I never like to have auto-fail the same that I don't like auto-success.

[bholmes4]

What would you do in a situation where someone has a 1 or 2 in their ability score? Would that be auto-fail territory?

Also, your numbers quoted are off by 1 unless you were meaning to also have a "natural 16 auto-fail" rule and forgot to mention it (16+2 = 18, which since you say 1+2 = 3 is a success should be a success as well).

My numbers may very well be off because I was over-tired and having a (couple) glasses of wine when I wrote that lol.

Anyway you have given me lots to think about here.

Some notes:
1. You are correct in guessing that a roll of natural 1 would be a success and a natural 16 would be a fail, regardless of score.
2. I don't like auto-fail or success either, unless as a DM I simply rule an action impossible for everyone (ie. the player wants to try and leap a 60 foot chasm or something silly). If it's in the range of human capabilities, anyone can try.
3. That said, I probably won't allow scores to go beyond 3 or 18. Yes, if you pick up a belt of Ogre Strength some will argue that you have a strength of 19 or 20 but I don't see it like that. In my mind, at that point you are on a scale that is not covered by the 3-18 range and are judged as a special case. Normal rules no longer apply. Likewise if you are deemed to have a score less than 3, you must be crippled or incapacitated or something. Want to bend those bars with broken arms? Ok roll a d30 and on a 30 you succeed. Roll a 6 or less and pass out from the pain. Whatever I decide I will make up on the spot as needed.

Personally, I find the roll under your score method to be a bit limiting - simple tasks become significantly more difficult for a person despite their score still being within the average range... let me expand on that:

If you roll for stats, which I do, then you have to allow a wider range of scores being considered "average" or "good enough" - that's why 9-12 are all lacking in modifiers instead of the trend that only 10 & 11 have no modifier that started with D&D 3rd edition (a system which penalizes rolling for scores as opposed to simply using point buy because of it's narrowed modifier brackets).

In one system - using set DCs and rolling to beat them - a score of 9 (average) and 12 (also still average) have the same chance of success at the same difficulty of task. Example: both have a 55% chance of succeeding a DC 10 check with a d20

In the other system - roll under score - a score of 9 (average) and 12 (also average) have different chances of success. With a d20 for the check that chance is 45% for the 9 and 60% for the 12, giving one supposedly average score 1.33 times greater chance of success than another average score. Changing that to a d16+2 check method gives the 9 a 43.75% chance of success and the 12 a 62.5% chance, exaggerating the difference to a 12 being 1.43 times more likely to succeed.

If that - accentuating the difference between ability scores while also make the already good better and the already low worse - is your goal, then you have a good system in mind for it. Otherwise, I believe you should reconsider.

My dislike for the DC system is why I am looking for an alternate system. Not only do I get annoyed using it as the DM (for other reasons), I feel like the DC system is too "fair" and tries to give everyone a good chance at pulling off tests. It devalues the "awesomeness" of rolling an 18 for your strength and minimizes the "crapitude" of having 3 strength. What is the point of having scores from 3-18 when really all that matters is: +3, +2, +1, 0, -1, -2 ,-3? (I know they can break ties, for burning stats, and such but my general point stands). I want a score of 15 strength to mean that yes, you will be significantly better on strength tests than your 13 strength partner. In combat it may not mean anything (both are +1 bonus) but in pure raw stength tests, you will win most of the time.

When my players roll up an 18 intelligence wizard I want them to feel like they just won the lottery and will go to great lengths to keep that character alive. The whole "I don't have to outrun the bear, I just have to outrun you" joke comes to mind. I want them so crazed to save their character they won't think twice about tripping the party halfling and running out of the cave when they see a starving owlbear running towards them.

However I see your point about the 9-11 range. I have to admit that I don't think characters are all that different in this range. The problem is I want to keep it simple too...

[bholmes4]

Another idea I had was to apply the ability score modifier to the score itself when making a roll-under test.

Thus:

3(-3) = 0
4(-2) = 2
5(-2) = 3
6(-1) = 5
7(-1) = 6
8(-1) = 7
9(+0) = 9
10(+0) = 10
11(+0) = 11
12(+0) = 12
13(+1) = 14
14(+1) = 15
15(+1) = 16
16(+2) = 18
17(+2) = 19
18(+3) = 21

The only reason they are written with values below 1 or exceeding 20 is that the DM may apply situational modifiers based on the action attempted. Additionally when two characters are competing, they are used to determine "degree" of success based on how much they succeeded by. Otherwise rolling a 20 would be a fail and a 1 would be a success, regardless of actual score.

This system helps accentuate the differences between the extreme ends of the range but I am debating if it's worth the, albeit minor, extra handling time.

[beermotor]

I'm with you on wanting something better than the DC system. At root, all systems are "You have an X % chance of success, roll." Rolling a D20 yields results in 5% increments D100 in 1% increments. Stats are generated on a bell curve, not a straight roll where you have an equal chance of each individual result. So what about rolling stat checks with 3d6? Maybe you need a table to graph out the results, but I guess what you're looking for is deviation from the stat score... the further you get up or down, the more anomalous the result. I'm not sure how you'd model a high stat score, though, as for example an 18 strength would be badass but then you've got a pretty low chance of scoring very close to it. Also, if you had a really low strength, you'd always be above it, and potentially by a whole lot. So, this is probably more math that I am qualified to deal with, heh.

Whatever, I went to law school.

[sheriffharry]

Rolling 3D6 makes "average" character: most have 10 or 11 stats, exceptionally one gets a 17 or 18 or a 3 or 4.
So character are normal dude, with maybe one or two very "abnormal/special" stat. It's all good and well.

Now, you certainly could use 3D6 (or 3D8 or 4D10 or whatever combination) to make check/attack roll during the game.
But what will happen then is that most of the time you will get "average" results: usually about half the maximum roll.
The problem with that is that it would be TOO realistic! Just like in real life, every thing the character do would be done "average", no fantastic success, no disastrous fumble. End result: the game might be a bit boring that way.. (but realistic!)

Just my 2 cents.
(I got a degree in physics and stop working in that field 10 years ago. I discovered I hate mathematics.)

[beermotor]

That's a good point, something I lose sight of often.

[bholmes4]

I also want to keep this down to a single die roll. Most of my house rules are an attempt to make this game even simpler and play even faster so I have to keep that a priority.

My newest idea is to do a roll-under system but grant a -2 bonus to ability test rolls for your prime ability. This means warriors are more likely to win strength contests, thieves to win agility tests and so on. For some reason I can easily accept a 17 strength wizard being shown up by the 9 strength halfling when trying to bust down a door but I have a harder time when the 17 strength warrior is shown up. Don't get me wrong, I want the odd gonzo situation to happen from time to time, but right now it happens too often under the normal roll-under rules. With this rule addition though you can have effective test scores between 3-20, providing an even larger range between the scores and reducing the likelihood of "fluke" results.

The way I justify these bonuses it is that by the nature of our class, you are constantly using and training your prime ability. So while yes, as a 16 intelligence wizard may have the same IQ as the 16 intelligence warrior, because the wizard is constantly reading, expanding his knowledge, and so on, his brain is just a bit quicker, a bit more finely tuned. If it came down to an intellectual contest the wizard would likely win (thus the -2 modifier to his roll).

[beermotor]

Actually, why not just skip rolls altogether? Door is heavy and takes 14+ strength to bash down. The brazier is large and awkward so you have to be taller than 5' and 15+ strength to pick it up a few inches and move it. The crowd is drunk and easily amused, and will be wildly entertained by personality 8+. If your stat is equal to the requirement, you succeed, but that's it. Much greater than, and you succeed in fantastic fashion. Slightly lower, you fail, but not horribly... very much lower and you fail spectacularly (pull a muscle, break your lockpick, the town rises up in arms against you for insulting their heritage, etc).

This actually comports with what someone said on another thread about not making people roll to search, but instead ask them "what are you doing specifically?" You could apply that here as well. So, maybe the crowd will be easily amused by some jokes and juggling and personality 8+, but if you're trying to teach them about particle physics, it doesn't matter if you have an 18, or if you roll a 1.

So you get realistic results and you get them fast (no rolls required).

[Vanguard]

Those kinds of requirements are even worse than the DC system, honestly. From a player standpoint, the most important rolls you make are during character creation, and you have no control over them. The only way this works is if you allow point buy stats, which is very contrary to the spirit of DCC. Why not play 4E or Pathfinder or any other version of D&D where this is the norm?

At a base level, there are two kinds of mechanics: linear systems and dice pool systems.

Linear systems balance the randomness (dice) versus raw skill (modifiers). There is not really any room for degrees of success in these kinds of systems because DCs = meet/exceed or fail. The issue I have is that the DC system is so swingy, particularly at low levels.

Dice pool systems are ideal for degrees of success. They balance the randomness (success on a X+) versus skill (the number of dice rolled). Essentially, your intended action has a total number of success to go off as planned. Fewer successes means a partial accomplishment, whereas no success equals unequivocal failure.

Honestly, I think you're better off leaving DCC RPG as is. Hacking it to function like a dice pool system is going to be as much work as designing a whole new system. Besides, there are already plenty of great games that have this kind of work for you.

[bholmes4]

I assume you are replying to Beermotor there Vanguard so I won't say much on the bulk of your post, other than I agree with everything but the statement that DCC is better off being left as is.

I have no desire to add a dice pool system and will stick with a single die roll system (likely a roll-under on d20). The problem is I simply can't stand the DC system and I have to find a way to house rule it out. I feel like one of the reasons D&D continues to click with me is that with all the disjointed and jumbled mechanics, it always felt like the "machinery" was kept a bit of a mystery. With the DC system (and unified mechanics of the d20 system) I feel like you are exposing the innards of the game, reducing it to down to dry and static numbers.

It's all subtle tricks on the mind but for whatever reason, roll-under disjointed systems work for me while DC systems can ruin a game. I have my theories why, I even started to write them up but that discussion has been done to death on message boards. It's a matter of opinion and either you like DC systems or you hate them. Just like ascending AC or descending AC...

[caveman]

I've been messing about with tests a bit in recent games. I too am frustrated by the DC system mostly because I find that as a DM I don't often have a clear DC in my head before I ask for the roll.
So I thought about trying to ask for d16s for simple rolls, d20s for average, and d24s for hard rolls, with a move up or down the dice chain depending on if the test is appropriate to the character...
But I've found I mostly have folks roll a d20... maybe step it down to d14/d16/d20?