valis
Explorer
I still need to clean up the formatting, but this should be a more useable adaptation of Courtney Campbell's social action system to 5e. It could probably still be streamlined further. I tried to maintain as much of the flavor as possible while minimizing the tracking of pluses and minuses to player rolls.
https://www.icloud.com/numbers/AwBU..._u9BVMOQrYVuFq#On_the_Non-Player_Character_5e
First of all, thanks for the wonderful comments. I worked a long time on this. It's based off the old Arabian Nights game, and is designed to preserve agency of the player characters and make social interactions less arbitrary.
It was designed originally for older styles of games (Basic/Expert) using the bell curve reaction table, which means their are some wonky interactions when converted to a straight 1d20 system. I've noted the same issues many people in the thread have noted.
However, having a long running 5th edition game, I do have some words of advice. The range of bonuses to stats is not +2 to +11 (Proficiency + Stat modifier). It's +2 to +17. A first level thief with expertise and a 20 charisma has a minimum value of 10 on a Charisma (Persuasion) check. They might not normally do that, but if they know you're handling these interactions mechanically, they would be more encouraged to.
What's more likely is that your bard will have a Charisma of 20 at the same time they are picking up Expertise, meaning that an mean roll of theirs will be 21, (+6 proficiency, plus 5 from stat), not considering their ability to roll at advantage or handing out inspiration dice to other people making ability checks.
What I'm saying is that difficult checks in the original system, at 11, (or 12 to get a friendly reaction) had a ~3-~8% chance of succeeding, pushed up to ~25% for characters with a very high Charisma (16+). 16 on Charisma on 3d6 down the line is very unlikely. A DC of 25 for a mid level party has a realistic possibility of having a 40%+ chance of success with very little work. Letting the Warlock roll at advantage and being granted a bardic inspiration die by the bard isn't that unlikely. Neither is a bard with expertise in persuasion when you're using this system (and they know how powerful it is).
Now, I encourage everyone to use it and change it for their own games. But know that it is trivially easy for your players to have only a slight chance of failure at middle or higher levels at tasks that should be challenging or nearly impossible. You may find if you use it as an objective resolution system that the lower difficulties cause players to become incredibly persuasive speakers, causing guards to leave their posts and flock to their cause with little difficulty. That's not a bad thing, but it is something you should be aware of.
Personally I'm a fan of objective difficulties, not level based ones (like in Pathfinder). A DC 20 cliff to climb is DC 20 for first and 20th level characters, rather than having it be a DC 15 cliff for low level characters and a DC 25 cliff for high level characters.
This also discounts the possibility of magic items or other effects that can influence the roll (which you can bet the players will leverage). It's important to keep in mind what bounded accuracy means, and how expertises and skill rolls can push the range of values. My 11th level group can routinely get +15 or more to persuasion skill checks. I asked for an investigation roll and the response was "32" today. The player isn't spec'd for investigation.
For reference:
+2 to +6: Proficiency
+4 to +12: Expertise
-2 to +5: Stat
~+1 to +4: Advantage
+1 to +6/8/10: Bardic Inspiration
Additional bonuses and effects due to magic items.