2d10 for Skill Checks

[DEFCON 1]

I generally like this system, but there is a difference between a combat roll and a skill roll in general. A 20 is an automatic success / crit in combat but not in a general skill check. So, if you use 2d10 for combat, crits will occur much less often. Now in 3e this wasn't as much of an issue since the threat range could get as large as a 15 or better, depending on feats and weapons. But in 5e, using 2d10 it might be necessary to redefine a crit as hitting by more than x (probably 5), but could vary based upon house rules.

Heh, yeah... I had actually gone through the math when I was first considering whether or not I would use 2d10 for attacks as well. Had I done it, I would have had critical hits occurring for anyone who rolled an 18, 19 or 20 on the 2d10, as those three numbers produce a 6% chance whereas a nat 20 is a 5% chance. I would have been fine with that extra 1% chance over normal crit... but ended up not using 2d10 for attacks after all.

 

[UngeheuerLich]

Actually every approach is legit here.
I know a game that just uses 3d6. And 17 and 18 crit while 3 and 4 fumble.

I however tend to think that DnD stories play better with the randomness of a d20. It makes those moments where ysomething happens unexpected.
It also allows those moments where a different character suddenly gets the spotlight because he unexpectedly rolls a 20.
I also tend to believe that advantage helps mitigating bad rolls. When it is in place suddenly your bonus matters a lot more. It keeps disadvantage from making your rolls too low and it allows advantage to almost guarantee a good result. I think giving advantage a lot makes for a better game. I especially like the xanathar's rules that give advantage when tool and skill apply.


It keeps the

 

[Stalker0]

I don’t agree that all skill checks are binary, many of them are not.

When my investigation guy rolls a 30 to check a room (and he does...a lot), you expect to find a bit more than if you get a 20, and more than if you get a 10.

Knowledge checks are often the same way. So that’s why I think 2d10 works for skill checks in a way that isn’t needed for attacks.

 

[Yunru]

When my investigation guy rolls a 30 to check a room (and he does...a lot), you expect to find a bit more than if you get a 20, and more than if you get a 10.

You do?

 

[DND_Reborn]

I don’t agree that all skill checks are binary, many of them are not.

When my investigation guy rolls a 30 to check a room (and he does...a lot), you expect to find a bit more than if you get a 20, and more than if you get a 10.

Knowledge checks are often the same way. So that’s why I think 2d10 works for skill checks in a way that isn’t needed for attacks.

Actually, once the DC is set, it doesn't matter if you make it by hitting the DC or beating it by 20 points. However, people seem to think that a higher total is better and that "my 24 beats your 18" even though we both only had to make a DC 15 check.

One of the players in our group actually brought this up during our session on Sunday. He proposed something akin to the more you beat the DC or AC the better the result. The DM countered with "Sure, but the more you fail by the worse it will be for you as well. So, you roll a 1 on that poison save and you'll take maximum damage from the poison instead of average, but you roll a 20 and you won't take any. How's that sound?" It wasn't quite so appealing to him then...

That's why 5E stressed that a 20 won't always succeed on a save or skill check, even though many tables still play that way. Some things are simply beyond the characters and a lot of people don't like it so they play a 20 always makes the check or save, but of course then a 1 should always fail. 20 always hitting and 1 always missing is only true for combat, so that nothing is "unhittable" and no one "never misses".

And to DEFCON 1, for us we try to keep things lower key at lower levels. Typically your major "thing" will only be +2, maybe +3, so we do have a lot of +3's and +4's. Now that we are into Tier 2, we have a few +8's due to expertise, but otherwise +6 and +7 is more normal. We use a slightly altered ability score modifier system (9-12 = 0, 13-14 = +1, 15-16 = +2, etc.) so we'll have no +5's ever. Our DM tweaked the proficiency bonus a bit as well. You begin at +1 for first level, then +2 for 2-4, +3 for 5-7, and so on until +8 at 20th. We all agreed to these changes primarily because it seemed lopsided that a 1st-level character with a Str 20 (so +5) could have a +7 attack and damage, but a 20th-level character with Str 11 (+0) could only be +6 attack and no damage. With our changes, the maximum ability score modifier is +4 (at 19 and 20) and maximum prof bonus is +8 at 20th. The prof bonus is twice the best you could do with "raw" talent. It works for us and really doesn't affect the balance of the game any since your maximum adjustment is +12 instead of +11.

 

[Saeviomagy]

I feel like the problem with 5e's skill system is that the difference between (ignoring class abilities for the moment) the best of the best and the worst of the worst is the difference between a -1 modifier and a +11.

If you pick a hard DC, you can't tell the difference between the two on 40% of rolls.

That's why the d20 feels too swingy.

Compare with 3e where the difference would have been between a -1 modifier and a +28.

 

[DND_Reborn]

I understand what you mean. With Bounded Accuracy they inadvertently created another issue for many players. That said, more than enough are happy with the d20 as is. Pick your poison I suppose.

 

[Laurefindel]

To add to the "swingyness" of 5e skill checks, we, as players and often as DMs too, like to roll the dice. Rolling dice is fun; there is a certain thrill to know whether they will come up in your favour (heck, there are many casino games based solely on that principle).

But technically - and 5e insists more on than point than any previous editions of D&D - a skill check should only be initiated when a) the outcome is uncertain and b) failure implies a significant consequence. Otherwise, the characters are assumed to be successful in their endeavours. But many DMs are quick to call for a roll when the outcome should have been pretty much set, and players are only happy to comply. Because let's face it, rolling is fun.

This of course does not create the perceived issue here, but I think it exacerbates it by creating situations where the able fails more often than it should.

But back on the subject, I like the 2d10 for skills. The DCs are mostly unaffected, and advantage/disadvantage plays out well. Since it's not an attack roll, critical range is irrelevant, and while it is not a true bell curve, it distributes the odds nicely toward 10s and 11s while retaining a certain swing. So while I like the "one d20 to rule them all" design behind D&D, 2d10 ability checks is a nice, un-intrusive fix to the lack of reliability based on training for skill checks.

 

[DEFCON 1]

I don’t agree that all skill checks are binary, many of them are not.

When my investigation guy rolls a 30 to check a room (and he does...a lot), you expect to find a bit more than if you get a 20, and more than if you get a 10.

Knowledge checks are often the same way. So that’s why I think 2d10 works for skill checks in a way that isn’t needed for attacks.

I am also a DM that gives out more information for better success on information checks. I've never really bothered with the whole trope of "only roll checks if there is a consequence for failure" thing that many of the DMs here swear by. After all, there's most likely never a consequence for not knowing some piece of esoterica... but at the same time I'm not going just give out all the information about something just because I don't want to have a skill check that doesn't have a consequence. To me that seems unnecessary.

Especially if a PC has proficiency in some subject (like Arcana, Nature or whatever)... the height of their check just tells me how much info about the subject the PC knows off the top of their head. And if there's more info to be gleaned and they feel like they need it... then they can go to a library and research. But I have no desire to just narrate several paragraphs worth of trivia about the subject because the player asked and I didn't want to ask for a check that had no "consequence for failure".

Others do so and that's fine... I just choose not do so.

 

[5ekyu]

I am also a DM that gives out more information for better success on information checks. I've never really bothered with the whole trope of "only roll checks if there is a consequence for failure" thing that many of the DMs here swear by. After all, there's most likely never a consequence for not knowing some piece of esoterica... but at the same time I'm not going just give out all the information about something just because I don't want to have a skill check that doesn't have a consequence. To me that seems unnecessary.

Especially if a PC has proficiency in some subject (like Arcana, Nature or whatever)... the height of their check just tells me how much info about the subject the PC knows off the top of their head. And if there's more info to be gleaned and they feel like they need it... then they can go to a library and research. But I have no desire to just narrate several paragraphs worth of trivia about the subject because the player asked and I didn't want to ask for a check that had no "consequence for failure".

Others do so and that's fine... I just choose not do so.

To me, this is where the 5e rule for failure - "some progress with setback" is goldmine.

Almost every failed knowledge check in my game has gotten some good info and some bad info that were either indistinguishable or maybe even the bad info seemed more plausible or more useful. but since the character also gets some narrative hooks for "confidence" (was it common knowledge at the academy, proven and documented multiple times over, did it only appear in that one wizard's toimes that kept his records mostly focused on the mating proclivities of young bugbears or did you overhear a drunken dwarf in a bar trying to impress some barmaid?") they can act on the information or not as they see fit.