D&D 5E Is expertise badly designed?

[Ristamar]

Regarding Reliable Talent, one interesting quirk in the rules is that it has absolutely no effect on passive scores.

If you want to really challenge a high level rogue with expertise, throw in an encounter that requires extended/repeated activities with with lots of opportunities to apply advantage or disadvantage on the opposing checks (roll for trap and hazard DC's instead of using static values). As the DM, roll in the open when possible as the rogue faces any obstacles. Avoid hard, binary fail states and be sure to include interesting choices and decision points in both success and failure so the player doesn't feel like the scene is on autopilot. And don't overuse this technique because players enjoy rolling dice.

 

[the Jester]

1: Bounded Accuracy isn't actually a thing in 5e. 5e uses 4e/2 math. Except when it doesn't and then things function really poorly.

No it doesn't.

3e assumes a base of +1 per level in things you're good at. 4e assumes +1 per 2 levels. Not to mention all the other boni you get, e.g. feats, magic weapons, stat bumps without a limit, etc. You can assume that the actual bonus a 3e or 4e character will have at any level above about 3rd will be significantly- perhaps even vastly- higher. The math is unbounded. Often, the difference between two pcs' bonuses for the same thing can be greater than the 20 pips on the die you are rolling. In other words, there are monsters that are a challenge to the fighter that other pcs simply can't touch.

In 5e, the math has real limitations imposed on it. And that doesn't make the game function poorly- it makes it shine. It enables mixed-level play. It maintains the worth of low level monsters even at high levels. It means that there is no case where the combat-dedicated rogue simply can't hit the bad guy that is a challenge for the fighter.

Now, that all may be not to your taste. That's fine. But that doesn't mean that it's functioning poorly- just that it's not to your taste.

 

[NotAYakk]

5e math for proficiency bonus is roughly 4e/2. 4e is Level/2, round down. 5e is 2+((Level-1)/4,) round down. The growth rate is two times higher, although the +2 means it's never actually twice as much within the level domain of 5e. (4e level 20 is at +10, 5e level 20 is at +6).

Likewise, stat bonuses in 4e tended to get in the +8-10 range, while 5e is limited to +5.

I think the assertion about "not using bounded accuracy" is that magic items are present in the game, but not accounted for in the challenge calculations of monsters. I think, anyway, MwaO would have to clarify.

4e was based off +1 per level after all modifiers.

+15/29 for half level, +4/29 for attribute, +6/29 for enhancement/item, +3ish for feat is +28 over 29 levels.

In 5e it is +4 prof/19 +2 stat/19 and optionally +3 enchant/19 for +9/19.

This isn't far off half 4e in one sense, but 1/3 4e in another (over domain of game).

ACs on monaters do not scale as fast by design. ACs tend to start off 'high', and don't grow as fast as ATK.

Skills in both are less controlled.

 

[tetrasodium]

Yes expertise is a mess & causes two problems as you advance. For example, an 8 int rogue or bard with arcana expertise will consistently put the dunce cap on a wizard with high int & mere proficiency in arcana. The second is that expertise on a skill that aligns with your prime stat amounts to always succeed against anything but plot armor An Arcane trickster with 20 dex & expertsie in stealth is going to wind up with+17 to the d20 & as soon as 9th level will have +13 making success all but certain sneaking through a gloomstalker's favored forest terrain while the gloomstalker's mere proficiency & 20 dex still has some chance of failing with her feeble +9 that will continue to fall behind

In my game I use proficiency dice(dmg263) & allow expertise advantage on the proficiency die

It has the effect at being more meaningful at lower levels (ie low chance of snake eyes) while the growing variance of the increasing proficiecy die shaves down some of the problematic parts of expertise that myelf & others noted already

 

[miggyG777]

In my game I use proficiency dice(dmg263) & allow expertise advantage on the proficiency die
View attachment 117482
It has the effect at being more meaningful at lower levels (ie low chance of snake eyes) while the growing variance of the increasing proficiecy die shaves down some of the problematic parts of expertise that myelf & others noted already

Very interesting approach thank you for sharing!

 

[Gradine]

Rogues being very good at skills is a problem... how?

Let players succeed. Your adventures should not hinge on the results of a single skill check (or even a handful) so I don't really understand what, if anything, this is supposed to be breaking in the first place.

 

[MwaO]

Wow. That's ... hmm.

Those are certainly words. Do you have a source for that? Either one?

Dungeons & Dragons Roleplaying Game Official Home Page - Article (Bounded Accuracy)

web.archive.org

"Since target numbers (DCs for checks, AC, and so on) and monster accuracy don't scale with level"

Page 274 details how monster AC and accuracy do scale with level. The DC guidance mentions about increasing DCs as PCs go up levels for bigger challenges. They don't mention the phrase 'bounded accuracy' anywhere in any of the core books. Nor do they mention how the game doesn't expect magic items to change the numbers and they're just something extra and unexpected and the game can handle it.

What they do say is that there's a typical campaign. And that typical campaign expects that PCs find magic items, almost eerily at a rate of 1 good number changing magic item per 4 levels. And that it is likely a 20th level weapon using PC probably finds roughly a +3 weapon. Which would mean they would adjust their to-hit by a total of +9 over 19 levels, exactly one half the likely +18 over 19 levels of 4e. And they would adjust their skills by +6, exactly one half the likely +12 over those same levels. And find a total of 5 magic items, exactly one half the 10 you'd expect to find in a 4e game using inherent bonuses.

D&D 5E - Magic Item Math of 5e

How many magic items should a typical 5e campaign have per PC? DMG, bottom of page 133: Over the course of a typical campaign, a party finds treasure hoards amounting to seven rolls on the Challenge 0-4 table, eighteen rolls on the Challenge 5-10 table, twelve rolls on the Challenge 11-16...

www.enworld.org

 

[lowkey13]

*Deleted by user*

 

[jayoungr]

Also, isn't Bounded Accuracy specifically about attack rolls and AC?

Expertise is about skill rolls.

 

[Tony Vargas]

I am a fairly new DM. I have heard a couple of people argue, that rogue and bard expertise is a bad design choice due to its impact on the modifiers in the bounded accuracy system. Is it too strong or just right? I would be interested in hearing your opinions.

It's not a bad design, it's just a place where 5e's goals of simplicity and BA are at slightly cross purposes.
Mainly, at high level, which doesn't generally seem to be a high priority in D&D game-design, anyway.

If you want to make Expertise more complicated to make it play nicer with BA, you could change it's progression, or have it replace proficiency with a different progression or cap total proficiency + stat modifier bonuses.

For instance, here Expertise replaces Proficiency:

Experience Points	Level	Proficiency Bonus	Expertise Bonus
0	1	+2	+5
300	2	+2	+5
900	3	+2	+5
2,700	4	+2	+5
6,500	5	+3	+6
14,000	6	+3	+6
23,000	7	+3	+6
34,000	8	+3	+6
48,000	9	+4	+6
64,000	10	+4	+6
85,000	11	+4	+6
100,000	12	+4	+6
120,000	13	+5	+7
140,000	14	+5	+7
165,000	15	+5	+7
195,000	16	+5	+7
225,000	17	+6	+8
265,000	18	+6	+8
305,000	19	+6	+8
355,000	20	+6	+8