D&D 5E Going for 3d6 instead of d20

[DemonSlayer]

Welcome to GURPS. That system uses 3d6, except lower is better.

I really like this concept, making the average happen more often. The swingy d20 is really prevalent in 5e.

 

[Bawylie]

I just use a magic 8 ball to determine if something works or not.


-Brad

 

[DEFCON 1]

I've also been thinking about the 2d10 method for my next game, mainly because I like the idea of the static bonus for ability + proficiency having more meaning and impact. I was expecting to probably go with 18-20 for a critical hit (that being a 6% chance to roll vs the normal 20 which is 5%.) I hadn't thought about Ad/Disad, but highest or lowest 2 of 3d10 probably will work fine for me.

The one thing I like theoretically about 2d10 is the idea of triggering/creating new Aspects for characters when they roll Double 0-- like if they are proficient with a skill and rolls 00 on the check (a 1 in 100 chance)... then whatever that check was about might be a hyper-focus or knowledge situation that could lend itself to their characterization. So for instance a wizard was rolling an INT (Arcana) check for info about some magical sigil they discovered and rolled 00... and we'd narrate it that they know the sigil is from a certain cult and that the character knows almost anything about this cult because they used to be a member, or their mentor was a member at one point, or the cult did something to them in the past and they learned everything they could about it for the future, or whatever we could come up with. Then in future rolls, that aspect could be used to gain inspiration for hard checks or automatic knowledge on easier ones.

Good thread! I'm looking forward to hearing other's experiences with the 2d10 concept.

 

[MNblockhead]

Gary went to the d20 system (originally drawing from a bag of numbered poker chips, and starting with a WWII tank combat war game) to get away from the "goofy bell curve."

Far be it from me to tell you how to play your game, but I wouldn't play D&D without the d20. Partly this is because the game's mechanics have been designed and play tested using the d20 and I don't want to do the math to rebalance everything. It would make more sense to just play another system, maybe one with a d6 dice-pool mechanic.

But it is mostly because the d20 is iconic. Nolstalgia plays a big role of course, but there is a very positive aesthetic to playing with all the platonic solids in the game. There is a kind of symbolism in the collection of D&D polyhedral dice that helps create the ambiance of the game. There is a pleasant tactile aspect to selecting a different die for a different kind of role. The funky dice are part of the "magic" of the game. I've pretty much done away with graph paper at the table. I use digital tools more and more to run aspects of the game and can live without a pen and paper character sheets, but we roll physical dice at my table, especially the beautiful icosahedra.

 

[Blue]

The problem is the bell curve's low granularity in the most common area will engender the opposite of what I want.

With bounded accuracy, we're supposedly looking at a "regular character" having somewhere near a 50/50 chance to succeed at what they are doing. Being a little bit better (say a +1 weapon) gives you a small increase - 5%.

Now let's look at the same thing with a 3d6 system. Rolling an 11 or higher is 50% both with d20 and 3d6, but if you add a +1, it's a 55% chance with d20 but a 62.5% chance with 3d6. The change for common rolls was 12.5%, two and half times it was for a d20. A -1 does the same thing, dumping the chance to 37.5%.

This loss of granularity makes every math advantage extremely important. Any inequality in character builds or optimization will lead to a wide discrepancy in what characters can do. You lose out on the character who only has a 16 ability score when others have 18s because it's crippling. Everyone will always be going for the races that best match their class. Trying a skill that isn't aligned to the ability scores you are increasing is so much more likely to fail that it's likely not worth taking, and the flip side is that expertise is virtually guaranteed because a +3 is godlike.

It's really the exact opposite of what I want, where players can make the characters they want and not have to sweat if they are a point or two behind or ahead of other characters.

 

[OB1]

Great timing for this thread, I was just thinking of trying something similar this morning!

I think I might test out 2d10 for skill checks and contests only as a way to emphasize ability and training for skills.

In contests, this would make it much more likely that the person with the higher modifier will win the contest.

In checks, the standard 10, 15 and 20 DCs now differentiate between an unskilled average person attempting something tricky, to something only someone with a lot of natural ability and training will regularly be able to accomplish.

I don't use crits (fail or hit) with skill checks, so the non existant 1s and very rare 20s won't hurt me there.

 

[Yunru]

The problem is the bell curve's low granularity in the most common area will engender the opposite of what I want.

With bounded accuracy, we're supposedly looking at a "regular character" having somewhere near a 50/50 chance to succeed at what they are doing. Being a little bit better (say a +1 weapon) gives you a small increase - 5%.

Now let's look at the same thing with a 3d6 system. Rolling an 11 or higher is 50% both with d20 and 3d6, but if you add a +1, it's a 55% chance with d20 but a 62.5% chance with 3d6. The change for common rolls was 12.5%, two and half times it was for a d20. A -1 does the same thing, dumping the chance to 37.5%.

This loss of granularity makes every math advantage extremely important. Any inequality in character builds or optimization will lead to a wide discrepancy in what characters can do. You lose out on the character who only has a 16 ability score when others have 18s because it's crippling. Everyone will always be going for the races that best match their class. Trying a skill that isn't aligned to the ability scores you are increasing is so much more likely to fail that it's likely not worth taking, and the flip side is that expertise is virtually guaranteed because a +3 is godlike.

It's really the exact opposite of what I want, where players can make the characters they want and not have to sweat if they are a point or two behind or ahead of other characters.

Why not use a d10?
There's still no bell curve, but static bonuses (such as proficiency) are much more impactful, as they represent a larger slice of the pie.

 

[der_kluge]

Great timing for this thread, I was just thinking of trying something similar this morning!

I think I might test out 2d10 for skill checks and contests only as a way to emphasize ability and training for skills.

Agreed. The first thing I thought of when reading the OP was "I think 2d10 would be easier to convert". So yea, 2d10. Still keeps everything 20-based, so conversions would be easier, and gives you some bell curve.

 

[Horwath]

Why not use a d10?
There's still no bell curve, but static bonuses (such as proficiency) are much more impactful, as they represent a larger slice of the pie.

That is also an idea.

I remember that we played in 3E/3.5E with d12 for initiative and opposed checks, so skill gets more vote than luck.

With only 1d10 all DC's must be lowered by 5, but as max roll would be only 4,5 pts higher than average roll static bonuses would be stronger.

crits would be 10%(20%/30% for the champion) of the time, but they are fun. Maybe put crits at static max normal damage, so we don't get massive damage too often.

 

[thexar]

We basically tried the 4d6-4 method (we simply counted 6 as zero). This was before 5th, so advantage wasn't part of the experiment. Also, bonuses greater than +2 were reduced to additional d6's. It started off as fun, but anything mildly difficult on a d20 was nearly impossible.

If it came up again, I'd try 3d8, counting 8 as zero (0-21). Or try GURPS, after the new fantasy ships.