[Part 1] So, earlier this week an user made a thread about how they didn't like Bounded Accuracy. Most of the thread was a clusterfuck of namecalling, but *some* actual discussion did occur a bit throughout, and after bump cap was hit.
, if you want to read the clusterfuck.
The complaints about BA were twofold: >i.Success Rates overall too low for Proficiency until lategame (Unless you add Expertise) >ii. Success Rates too high for a group of dabblers/amateurs.
People in the previous thread argued quite a bit about >i. Some people thought roughly DC10 = 15% Failure for an average Level 5 PC in a thing theyre trained in was fine. Others thought that by level 5 you should be able to succeed at tougher tasks. People also argued about the published DCs in modules, some of which were pure bullshit.
[Part 2] Anyways, I ran some numbers based on what I read in the thread to see what they were talking about (will post as pictures in the next couple posts). I ran the numbers on >ii, as that was the part that seemed most interesting.
The best example given was when a trained PC attempts a DC 15 task and fails, a party member without skill will almost certainly succeed, carrying the group, but making him look like an idiot.
So I thought about how to examine this, and the way I see it, there are basically 3 types of checks: >A. Only one attempt - either you can't try again, or it gets harder with each failure, or failure results in serious consequences. This one isn't "giving us trouble". >B. Everyone has to pass - Whether its jumping across a pit, or outrunning a bear, or the party is separated and being interrogated, or you're trying to all stealth into a castle - if one person fails, the whole group suffers consequences, and the whole group has to try independently. This one seems like it would "give us different trouble", based on the math I ran. Not what I'm going to talk about first though. >C. If one person passes, the group is good. This is generally your knowledge checks. With some DMs it might also be social checks and investigation checks (With others the social and investigation checks might be type A) This is the one where the trained guy looks like a chump next to his allies. This is the one that was called out as not fun in that thread.
Jack Rogers
[Part 3] >C. So I ran the numbers (Anydice, GSheets) for DCs 5-24, for some plausible parties. For all DCs between DC5 and DC17 (except DC 10 and DC 11), the weaker (at skill bonuses) party rolling is like having a +11, and the stronger is like having a +12.
The AutoSuccess rule was posted (which is apparently a kludgey approximation of a "Autosucceed on
Robert Harris
[Part 4] It's a tricky situation.
I don't have the answer to make trained characters not feel like chumps in C. The frequency this would happen at level 17 isn't so high, especially if the autosuccess rules in the DMG are used (then you only need to worry about it happening at DC 16-24) but before then, it's rampant.
It also emphasizes two things: >I. Unless it's for DC18+ checks, there's no reason to take Knowledge skills. Investigation and Social Skills are the same, unless there are escalating penalties for each failure. >II. You should never use "the whole group needs to individually succeed" checks, it's basically a guaranteed failure.
Here's what I've got for a conversation starter/suggestions to consider: >a. Knowledge checks are trained only? >b. In C type scenarios do not allow all party members to roll. Allow only a single roll for the whole party - maybe grant advantage if the party has someone else with at least a +3? >c. in B type scenarios, again, single roll, and give disadvantage if a party member has a penalty?
Easy answer is to just double or triple or otherwise more heavily weigh proficiency bonuses for skills a character is trained in.
Ryan Wood
You'd need your level 1 proficient characters to net a +11 for it to balance out with type C checks.
And that doesn't address the problem of failures for type B checks.
It also means either really high endgame numbers, or it means giving the everyone the endgame numbers at level 1 and dropping the per-level proficiency scaling. Either option has cascading side effects to address.
Because of those things, I'm not convinced this is a problem that's best addressed by increasing the size of proficiency.
Anthony Robinson
But I mean, if you wanted to both set up your chargen so you start with a 20 in a stat of your choice, and then say Proficiency = +6 full stop, I suppose that would do it.
It also gets rid of the "gradually increased specialization" thing that D&D normally has.
Still doesn't address the Type B issue though. Group stealth checks are still a shit.
Brayden Lopez
Stealth checks aside, Type B situations are generally handled best by having the first character who succeeds do something to help the others succeed.
>Jump across chasm >Can now catch other players who are attempting to jump or secure a rope to something on your side >Check gets easier for everyone
Caleb Green
Sure, that's better than trying to have everyone jump. But it's not always an option. That staircase scene in LOTR comes to mind.
Josiah Richardson
So, I asked this question last thread, but everybody was too busy namecalling to answer:
If you dropped the number scaling in your primary stat and proficiency, and started both of them at the max value, what would be the consequences?
>Youd need to raise AC values. >The players would have much less need for ASIs, and would take more feats. >The gap between good and bad saves would always be at endgame levels, increasing the need to target the right saves of an enemy. >Trained Characters will be much better at skillchecks and you can have tougher challenges. >No number scaling: Characters stay in the same groove the whole game, easier to balance around. >No number scaling: Characters may get bored at the lack of number advancement, if they're big into number growth.
