D&D Next, 3d6?

[Grod_The_Giant]

Has anyone ever tried running 5e rolling 3d6 instead of 1d20? (With, presumably, appropriate adjustment for Advantage/Disadvantage) It seems to me that a bell curve would go well with bounded accuracy-- when the dice skew more strongly towards the average, small bonuses become more important. 3d6, then, would give more consistency in checks and emphasize the difference between specialists and nonspecialists a bit more strongly, without putting the rolls completely out of anyone's reach.

[MaxWilson]

Has anyone ever tried running 5e rolling 3d6 instead of 1d20? (With, presumably, appropriate adjustment for Advantage/Disadvantage) It seems to me that a bell curve would go well with bounded accuracy-- when the dice skew more strongly towards the average, small bonuses become more important. 3d6, then, would give more consistency in checks and emphasize the difference between specialists and nonspecialists a bit more strongly, without putting the rolls completely out of anyone's reach.

I experimented with it when 5E first came out. Decided that it felt too against-the-grain. Leave 3d6 to GURPS. To get the bell curve effect in 5E, just require multiple rolls. 5E already does this for combat, and I do for some skill checks too. E.g. "three consecutive DC 10 Arcane checks" is easy if you're trained, but hard if you're just defaulting off Int 13.

[Slipperychicken]

People tried it in 3.5, and some liked it. I think GURPS uses 3d6, and it works fine there.


I love the idea of a bell curve, but I'm not sure what we'd need to change to accomodate the removal of 1,2,19,and 20. Should criticals happen on 18? 16-18 so it's still a 5% chance? Should DC 20 become DC 18 instead, or DC 16? Armor class and tohit would gain huge importance too. Chance to roll a 15 or higher would drop from 30% to 10%, and getting the highest result would go from 5% to just under half a percent.

This might be a good mod if the details get ironed out, but I think it would look more like an overhaul. It could help deal with some issues that the d20 system encourages.

[pwykersotz]

I've done it, mainly because it was necessary. I have a player whose luck is abysmal. He regularly rolls single digits for an entire game. Every d20 roll. Two sessions ago I greenlit 3d6 for him, just because I was tired of poor luck failing him and causing him to have less fun since his character was less effective. He now rolls fairly average, just like expected, and he's having a lot more fun.

I tried it with another group a year or so ago, and I found that people just like using d20's more. Still, it's a variant that I think is perfectly fine to use.


I love the idea of a bell curve, but I'm not sure what we'd need to change to accomodate the removal of 1,2,19,and 20. Should criticals happen on 18? 16-18 so it's still a 5% chance? Should DC 20 become DC 18 instead, or DC 16? Armor class and tohit would gain huge importance too. Chance to roll a 15 or higher would drop from 30% to 10%, and getting the highest result would go from 5% to just under half a percent.

This might be a good mod if the details get ironed out, but I think it would look more like an overhaul. It could help deal with some issues that the d20 system encourages.

There's some guidance that the 3.5 SRD has. (http://www.d20srd.org/srd/variant/adventuring/bellCurveRolls.htm)

[JellyPooga]

It's an interesting idea, certainly. I've always liked GURPS for its relatively heavy bell-curve with 3d6, but I'm not sure 5ed is appropriately balanced for it.

DC's would have to be modified appropriately for most given tasks, in general, they would be slightly lower to compensate the comparative difficulty of rolling anything above 15. I'd suggest around a 2 point reduction, as a ball-park reduction (i.e. medium difficulty would be 13, hard 18 and so forth).

Now consider a 1st Level Rogue with +7 to two skills of his choice (proficiency +2, doubled for expertise, +3 ability score). +7 on a 3d6 bell-curve is a massive bonus, far beyond it's scope in a linear scale. Worth bearing in mind, I think, before rushing into it.

Perhaps a 2d10 curve would suit better; the impact of high modifiers is less, but not quite so drastic and you still get your curve.

[Grod_The_Giant]

I think you're right for hard checks, but not for medium or average. Check out what some probing with AnyDice gives us:




Schmuck +0
Trained +5
Trained and Expert +8
Master +10
Master and Expert +15


Average DC 8
84%
100%
--
--
--


Average DC 10
62%
98%
100%
--
--


Medium DC 13
26%
84%
98%
100%
--


Medium DC 15
9%
62%
90%
98%
--


Hard DC 18
0.5%
26%
62%
84%
100%


Hard DC 20
--
9%
37%
62%
98%


Very Hard DC 22
--
2%
16%
37%
91%


Very Hard DC 25
--
--
2%
9%
62%


Impossible DC 28
--
--
--
0.5%
26%


Impossible DC 30
--
--
--
--
9%



The question, I guess, is how hard a "hard" check should be. I think there's some virtue in keeping the DCs high-- a DC 20 check like, say, breaking a chain is unattainable (or nearly so) to a dabbler, barely possible for someone with training, a decent challenge for a master but still not much to a paragon with Expertise. Like I said, it emphasizes the importance of that numerical bonus in the way that the d20 doesn't.

[Slipperychicken]

I love that table for skills, but it also applies to AC. We'd probably want to re-examine the impact of AC on defensive CR. Also, PCs with good armor and shields will become extremely hard targets at low levels.

[JNAProductions]

2d10, methinks, is probably good.

[Talamare]

GURPS is a pretty disgusting system. It isn't hard to get like 15 in nearly everything making you good at everything
Once you reach like 15~16 in your Health stat, become this like never dying terminator. You will be at like -60 Hp and like, "Nah, yea, I'm at -60 but I'm fine"

The 3d6 curve is just too extreme
I suggest 2d10, the numbers are less extreme.
Not to mention a 7 dice set has 2d10 already

Crits happen on 19-20, 3%
Champion crit on 18-20, 6%

2d10 has nearly as much range as 1d20

while 3d6 has almost no range
If we use 3&4/17&18 for Crits, it only leaves us with 5~8/13~16 that aren't average rolls

1d20 / 9~13 = 25% chance
2d10 / 9~13 = 44% chance
3d6- / 9~12 = 48% chance

[T.G. Oskar]

IMO, if you want to see what D&D 5e would look on a bell curve, go check up the rules for Fantasy AGE. They're pretty much that, with a little extra tucked in.

The first setting that used the AGE rules, Dragon Age, uses classes and the skill system is pretty similar to that of 5e - in fact, it's a reason why it felt natural to play 5e. In the system, the range of difficulties go from 7 to 21, in 2-point increments; something "routine" would have a DC of 7, whereas something "nigh impossible" would have a DC of 21. Accounting for proficiency bonus, ability bonus and an average of 10 or 11 on the check, a 20th level character using proficiency can do something "nigh impossible" around 47.5% of the times (proficiency 6 + ability 5; average of 10.5), while someone just using its ability would make it around 1 out of 3 times (not exact, but a fair ballpark). Skills use a similar Ability (Skill) concept (for example, Charisma [Deception] in 5e would have its equivalent in AGE in Communication [Deception]), but progress differently; skills only grant a +2 bonus on the check, not a given "proficiency bonus", and can only barely increase to +3. Even talents are similar to 5e feats, and you could entirely replace superiority dice with stunts, where the Battlemaster learns more advanced stunts.

As it stands, it's doable, though you have to take the nuances of 5e into account. AGE is also mired in bounded accuracy, but its main defining factor is ability score progression, whereas in 5e it's a combination of factors. Magic items in AGE rarely, if ever, grant a bonus to attack and damage rolls, in comparison to 5e; on the other hand, magic items in AGE mostly grant "feat" advancements, which are often useful when you're attempting to dabble into another fighting style, make use of new talents, or just enhance what you already get. 5e, in theory, has bigger numbers for less effort, and because of that, you can see a reasonable gulf.

One final thing - AGE has no crits, and what a triple 6 does (and what a triple 1 does as well) is just giving you 6 or 1 stunt points (so rolling low isn't bad, but rolling high isn't exactly great other than essentially auto-succeeding)

So, in short - while it'd require an adjustment to fit the concept of 5e, a bell curve retooling can be done with minimal effort, so as long as you realize that you need to readjust the DC table to something more reasonable. Something around 5 (routine; someone rolling three 1s might not do it, but someone rolling a 4 with any sort of benefit would do it reliably) to somewhere around 25 to 31 for near-impossible (using 31 as the limit, someone with full proficiency bonus + ability bonus would never do it, and it'd require someone along the lines of a Bard/Rogue with Expertise on the skill or a 20th level Barbarian to actually have a chance to pull it off, and even then it has a massive chance to fail). Gauge the DC table well, and you might have an entertaining alternative.

[Foxhound438]

Has anyone ever tried running 5e rolling 3d6 instead of 1d20? (With, presumably, appropriate adjustment for Advantage/Disadvantage) It seems to me that a bell curve would go well with bounded accuracy-- when the dice skew more strongly towards the average, small bonuses become more important. 3d6, then, would give more consistency in checks and emphasize the difference between specialists and nonspecialists a bit more strongly, without putting the rolls completely out of anyone's reach.

nah, makes higher AC overpowered. all of a sudden everyone has a shield with defense style and nothing hits anything and combat becomes sluggish as hell. With an AC 18, something with only a +5 str will hardly ever hit you- and that goes both ways.

[JoeJ]

If rolling an 18 on 3d6 equals a critical, there's a 1:216 chance of that happening, as opposed to 1:20 using a d20. That will tend to make combat slower and less exciting.

Also 3d6 makes the -5/+10 feats all but worthless. (Some people might consider that a good thing, though.)

[JellyPooga]

GURPS is a pretty disgusting system. It isn't hard to get like 15 in nearly everything making you good at everything
Once you reach like 15~16 in your Health stat, become this like never dying terminator. You will be at like -60 Hp and like, "Nah, yea, I'm at -60 but I'm fine"

Um, what? Unless you've invested heavily in Advantages that make it hard to die or your HT is much higher than 15, if you're at -60hp in GURPS, you're almost certainly dead.

Also, whilst I agree that GURPS can be a "pretty disgusting system" at times, it's only considered so because of the extent to which it can be abused if given free reign with it. 15 in nearly every stat is very achievable with enough points, sure, but if you've only got 150pts to play with and have a cap of 30pts of Disadvantages, it's not even a possibility (DX and IQ 15 is 200pts for a start).

On topic; yeah, 3d6 makes those -5/+10 Feats look pretty horrible!

As for combat bogging down, I'm not so sure. Hitting AC 18 according to the table Grod posted, with +5 to hit will happen about 25% of the time. +5 is a very common "to hit" modifier at level 1 and hitting 1 time in 4 sounds fairly reasonable. Sure, compared to the (approximately) 1:3 you get from 1d20 it's lower, but not drastically so.

[Cybren]

GURPS is a pretty disgusting system. It isn't hard to get like 15 in nearly everything making you good at everything
Once you reach like 15~16 in your Health stat, become this like never dying terminator. You will be at like -60 Hp and like, "Nah, yea, I'm at -60 but I'm fine

You do realize that you have to roll to stay concious every turn you're at 0 or less hp and take any actions or make a defense roll, and roll at every - multiple of your HP to not die. Your roll to stay concious is at -1 per multiple of HP below zero you're at, too, and you make it every second. Did I mention when went to 1/3rd your hp your move and dodge were halved so you probably aren't getting away.

Your issue is basically that people that spent points to be good at surviving are good at surviving

[Theodoxus]

I've done it, mainly because it was necessary. I have a player whose luck is abysmal. He regularly rolls single digits for an entire game. Every d20 roll. Two sessions ago I greenlit 3d6 for him, just because I was tired of poor luck failing him and causing him to have less fun since his character was less effective. He now rolls fairly average, just like expected, and he's having a lot more fun.

One option for your unlucky player, might be to use the d10/d6 combo. It perfectly simulates a d20, the d10 provides the base number, and the d6 is the control die. 1-3, you add nothing, 4-6, you add 10. So, if you rolled an 8 and a 3, that's the same as an 8 on the d20. If you rolled a 3 on the d10, and 5 on the d6, that's a 13.

I have a friend who uses this method all the time and we've never had an issue.

[Talamare]

Um, what? Unless you've invested heavily in Advantages that make it hard to die or your HT is much higher than 15, if you're at -60hp in GURPS, you're almost certainly dead.

Also, whilst I agree that GURPS can be a "pretty disgusting system" at times, it's only considered so because of the extent to which it can be abused if given free reign with it. 15 in nearly every stat is very achievable with enough points, sure, but if you've only got 150pts to play with and have a cap of 30pts of Disadvantages, it's not even a possibility (DX and IQ 15 is 200pts for a start).


15 costed 60pts

You do realize that you have to roll to stay concious every turn you're at 0 or less hp and take any actions or make a defense roll, and roll at every - multiple of your HP to not die. Your roll to stay concious is at -1 per multiple of HP below zero you're at, too, and you make it every second. Did I mention when went to 1/3rd your hp your move and dodge were halved so you probably aren't getting away.

Your issue is basically that people that spent points to be good at surviving are good at surviving

That was MY character, and we checked the rules and we were playing it correctly

[Cybren]

15 costed 60pts


That was MY character, and we checked the rules and we were playing it correctly

In order to avoid continuing to derail this thread i won't post after this but it seems you were playing 3rd (or 2nd etc) edition, which I actually don't have any experience in. In GURPS 4E stats have a linear cost, 15 HT would cost 50pts, and in 4E HP is tied to ST and not HT, so you don't get to double dip; A ST 10 HT 15 character in 4E GURPS is pretty survivable; they are really good at their rolls to stay conscious/alive but would still be automatically dead at -50HP. You'd get more survive-ability from ST 12 [20] HT 12[20] Combat Reflexes [15] Fit [5] than just HT 16 [60], and even your HT 16 abomination isn't that much protected against crippling injuries, damage modifiers, and major wounds. Get hit in the skull for 5 or more points of injury, for example, and you automatically get to make a roll for knockdown at HT-10. Enjoy your success on a 6 or less, because if you fail you drop whatever you're holding and fall prone, and are stunned until you recover (which you should do quickly, since you have a high HT, but good luck not being already riddled with bullets by the time you stand up).

[JellyPooga]

15 costed 60pts

I'll quote you again:

[quote]. It isn't hard to get like 15 in nearly everything making you good at everything[/qupte

A 15 in HT costs 50pts, not 60, for starts (as referenced by Cybren). Assuming you're using 4ed, of course.

Second, if you're going to be "good at everything" with 15 in all your base stats; which is what I took your meaning to be; 15 in each of ST, DX, IQ and HT will cost you 300pts. That's without Skills and without Advantages. An average "adventuring" character, regardless of genre or setting, is around the 150-200pt mark, including Disadvantages. Your 300pt character who's "good at everything" is far and above what might be considered "normal".

Third; 15 HP means you can technically survive up to -150. At that point, you've suffered complete bodily destruction; you're ash, you've been gibbed into a red and bloody mist, you're less than a skeleton...there is nothing left to revive. However, as soon as you hit -1, you're testing every second to see if you suddenly and inexplicably pass into the beyond. Well, maybe it isn't that inexplicable; it could be the bullet hole in your chest. :smallamused:

[AstralFire]

IMO, if you're going to use 3d6 with 5E, random criticals as a whole should be dropped -- unless you're fully automating your rolls with a tool like Roll20 it's very annoying to get a 5% critical chance properly working on 3d6 in a way that doesn't feel kludgy -- and replaced with some sort of hero/action/fate/force point system to allow you to break through the monotony. But really, I don't recommend it -- the stability is better done with advantage/disadvantage and careful setting of DCs unless you're willing to do some major system work.

[Talamare]

I'll quote you again:

[quote]. It isn't hard to get like 15 in nearly everything making you good at everything[/qupte

A 15 in HT costs 50pts, not 60, for starts (as referenced by Cybren). Assuming you're using 4ed, of course.
As Cyb wisely assumed, it was 3rd.
ST was a useless, IQ determined 95% of what you did, unless you're playing in a stone age tech. You should just use a bow/gun, which got bonuses for having high IQ. Then you just needed to spend maybe like 1~2xp on the bow/guns you're actually using to have like 90% accuracy

Now then, to attempt to pull this thread back to where it belongs.

Someone mentioned a fairly comprehensive AGE system that uses 3d6 and has been extensively balanced for using 3d6, so instead of playing DnD. Just play AGE, the core experience is the same. You get to RP fantasy heroes with friends.

Otherwise, just use 2d10. It will give you a softer bell curve towards average results. It will give you clean percentages for crits. It will allow you to use similar expected DCs that a d20 would. Finally, everyone has 2d10.

[Grod_The_Giant]

Rolling two 6's has a probability of... I think about 7%, making it a good crit stand-in if that's your concern. You could do the same with snake eyes for critical fumbles.

You can get some stability with rolling multiple d20s, but setting DCs? D20 is swingy because of the large dice size. When you add in the fact that bonuses are only between 25% and 50%?

[AstralFire]

Rolling two 6's has a probability of... I think about 7%, making it a good crit stand-in if that's your concern. You could do the same with snake eyes for critical fumbles.

You can get some stability with rolling multiple d20s, but setting DCs? D20 is swingy because of the large dice size. When you add in the fact that bonuses are only between 25% and 50%?

Yes, setting DCs. With 3E, the assumption at a lot of tables was that a task was relatively easy if you could do it while taking 10, and I feel this was reflected in a lot of the DCs. The problem was that if you wanted to do most of those things in combat, a sure-fire success out of combat became a 50% chance of failure in.

With 5E and no "taking 10" (although a few skills have 'passive' modes similar to taking 10) I simply set the DC based on how often I think someone with a +3 bonus should be failing it. So if I think they'd only fail 10% of the time, instead of setting something at a DC 10 and rolling 3d6, I'd set it at DC 5 and roll a 1d20. In 3E, you could make a strong argument for 3d6 in some regards because of uncapped modifiers going crazy and skills that have effects based on flat, linear degrees of success -- but in 5E you can adhoc everything, and I freely do so to smooth out results when I think it's appropriate.