D&D 5E Not liking Bounded Accuracy

[Ilbranteloth]

I like Bounded Accuracy, not so much on the PC side as on the world side. Having AC 15 means "my skin is as hard as rock, like a Gargoyle." Having AC 18 means "like a fighter in plate armor" and AC 20 is "like a fully armored fighter in plate mail with a shield."

AC 15 in 5E is basically like AC 5 in AD&D, AC 18 in 5E is akin to AC 1 or 2 in AD&D, and AC 20 is the new AC 0.

I skipped all the editions of (A)D&D that didn't use something like Bounded Accuracy (everything after 2nd and before 5th) and I wouldn't have started playing 5th if hadn't come back down to earth. I did play a couple of computer games from that era (Icewind Dale II is a 3rd edition game) but I can never figure out what AC 54 is supposed to mean or how it is different, in real terms, from AC 72 except that it is a known fact that most monsters in the game don't have a BAB over +52 so AC 72 is the target you need to hit for effective tanking. For a CRPG that kind of meaningless number-ism is tolerable but for an RPG it would be horrific.

That's exactly the problem I had. 3/3.5e was enough of an evolution that it made sense. But when playing a video game and I have a 163 AC I have no idea what it means. Same thing with the way 4e worked.

Really, I think the key for bounded accuracy is that the numbers reflect something in the world. A suit of armor = x AC. It completely separates the level of the characters and monsters from the world around them.

The abilities and skills of the characters and monsters then act upon those things. A higher level character naturally has a better chance of succeeding. The challenge in the design is working within a scale that makes sense.

In the past, other games used percentile dice to give more granularity. But the d20 is the die in D&D at this point. Even non-gamers know what a natural 20 is. So that limits the scale to 5% increments (which is not that bad actually). The bigger factor is that by using a d20, everything number has an equal chance of succeeding. A 3d6 approach with a bell curve would make critical successes and critical failures less common. But we have a 5% chance of either each time the die is rolled.

When you consider the numbers, a 20th level character with a 20 Ability Score and Expertise can have a +17 to their ability. With a 30 in the ability it could be as high as 22. And there are magical ways to increase that. That puts the ceiling well above the 30 that they've noted as Practically Impossible. So I would say that 30 and below are those things that a normal creature without magical assistance might be able to accomplish.

That +17 = 85% better chance than your average person without proficiency, and a 75% better chance than somebody with the most basic training. That's pretty significant.

Where I think it's potentially lacking is the difference between trained and untrained. The 10% difference at 1st level isn't horrible for some things. But there are a lot of things that it feels like an untrained character shouldn't be able to succeed at all, or it's nearly impossible. I've tried a number of possibilities to address this.

The easiest is to assign disadvantage when attempting an untrained skill. It's consistent, and there is no DM adjudication needed. You aren't trained? Roll with disadvantage.

But another option is to consider the DC. If the DC descriptions (easy, hard, nearly impossible) is describing the task in relation to a trained character (and it doesn't specify that it is), then the DC for an untrained character can be different.

For example, climbing a cliff that has an overhang that you must traverse, hanging from the bottom. For a skilled climber that might be considered hard. But for an untrained climber you could consider that nearly impossible. This solution seems like a good option, particularly with 5e, because the DM can determine under the specific circumstances whether the difference would be significant, and by how much.

 

[FormerlyHemlock]

Where I think it's potentially lacking is the difference between trained and untrained. The 10% difference at 1st level isn't horrible for some things. But there are a lot of things that it feels like an untrained character shouldn't be able to succeed at all, or it's nearly impossible. I've tried a number of possibilities to address this.

When I want a sharper bell curve, I add more dice. Either make it an opposed check, or a repeated check. If I think rock-climbing this here cliff should be non-hazardous for trained personnel with good equipment, but potentially deadly for novices who are just messing around, I'll make it a DC 7 Strength (Athletics) check per 10' of height you climb; failure means you make no progress this time; failure by 5 or more means you lose your grip and, if you're not secured to another climber, fall.

 

[Ilbranteloth]

When I want a sharper bell curve, I add more dice. Either make it an opposed check, or a repeated check. If I think rock-climbing this here cliff should be non-hazardous for trained personnel with good equipment, but potentially deadly for novices who are just messing around, I'll make it a DC 7 Strength (Athletics) check per 10' of height you climb; failure means you make no progress this time; failure by 5 or more means you lose your grip and, if you're not secured to another climber, fall.

I just prefer to avoid the multiple checks. Although I do occasionally use the 3 successes before 3 failures mechanic for skill checks. I suppose that would be another option for untrained individuals.

But I'm pretty happy with either the separate DC, or simply stating that an untrained individual can't do it. If you're trying to bluff your way through being your friend, a bard, and they request a song and hand you a lute, it really doesn't matter what you try. If you haven't learned how to play the lute, then it's either nearly impossible, or not possible at all.

I'm fine with going with it's not possible at all in my campaign, and that puts them in a difficult situation to attempt to continue to bluff their way through it...

"OK, a song then, any requests?"
Several songs are called out.
"Oh, yes I quite like that one"
I pretend I'm about to start the song, then stop -
"You know what, that reminds me of the time when my friends an I were in Waterdeep, and we..."

Not being able to perform the skill can be a very rewarding challenge to overcome. Just like when the rogue fails to pick the lock, so you must figure out another way to get in. It can be as simple as kicking in the door, or whatever other options they come up with.

I'm just not a fan of the idea that everybody can try every possible task, trained or otherwise.

 

[Tony Vargas]

Hi everyone.

One of my players is not liking bounded accuracy. He doesn't feel like his character is improving when his numbers only go up about once every other level.

Attack rolls go up slowly, but damage balloons - and you probably attack more often. It's not that you're not improving, just how you're improving is different... (In the combat pillar, anyway...)

His issue is that his skill bonus of +5 at 1st level, up to maybe +11 at 20th level, doesn't make much that he can do at 20th level that he couldn't do at 1st level. I sympathize with his position a bit. I can see the benefits of both a steady progression like in 3rd and 4th, and the benefits of the slower progression of 5th.

It does represent a different scope or sweep of advancement in the course of a long campaign. BA keeps the scope narrow, so you don't have to have wildly different adventures at high vs low level, and a campaign doesn't need to change tone and scope as you advance through Tiers.

It's like E6 that way, except your spells /do/ keep getting better. ;P

Is there anyone else who dislikes bounded accuracy? What do you not like about it, and why? Would you change it? How would you change it?

Mechanically it's OK, but it's not like it's hard to change. Just increase the proficiency progression: it's the same for everyone, so it's a very simple variant. Adjust monster stats and DCs to match.

If you want a more 3e experience, with classes, erm, out-classing eachother in terms of ranks and BAB, add more mechanics like Expertise. Let fighters have Expertise in their Style and/or weapon, so they can double and/or triple up on attack bonuses, and give Extra Attacks penalties to hit, so you /need/ those big bonuses to use it effectively, for instance.

 

[OB1]

I'm just not a fan of the idea that everybody can try every possible task, trained or otherwise.

I agree! And I think this is accounted for in the rules for the basic flow of the game.

DM Describes the environment
PC describes what action they would like to take
DM narrates the results, and may ask for a check if the outcome is uncertain

In my mind, what is uncertain, and how uncertain it is, may be different from PC to PC as what is easy for one character my be nearly impossible for another.

So I have no problem assigning DC 15 to a climb check up a steep cliff for a strong, trained PC, and simply telling the weak, untrained mage that they cannot climb it. If the strong, trained PC then wants to try and help the mage up the cliff, it might be very hard, requiring a DC 25 to succeed.

 

[GMMichael]

DM narrates the results, and may ask for a check if the outcome is uncertain

Thank you! It sure is fun to feel those multi-colored dice and watch them roll, but it's not even required by the game. The DM narrates; dice are only needed for uncertain outcomes.