I must create a system, or be enslaved by another man's. I will not reason and compare: my business is to create.

- William Blake

Thursday, January 12, 2017

Grokking 5e D&D (I) - Skills, and how they (don't?) work

The Fifth Edition of D&D took me a while to fully understand. One reason is that, contrary to what it may seem, 5e is in some ways different from all other editions (although probably not as much as 4e, for example), even though it keeps enough classic features to allow most fans to recognize in 5e a (somewhat) familiar version the game they always loved.

Another reason is that 5e is a bit opaque about some of its design. For example, it is clear that skills are, by default, more important than tools, and some tools (thieves' tools) more important than others. The designers realize this, and create backgrounds (and other rules) accordingly, but at the same time create a feat that lets you choose tools OR skills, as if they had similar importance, which can be confusing to newbies.

I like it more now than when I first started playing (j fact, it is probably my favorite game at this moment), so I though I would share some little insights about 5e's philosophy. They might be obvious to some (specially if you're a long time 5e fan), but they may also vbe helpful if you're trying to understand the system a bit better.

The first things that comes to mind is how skills work. Or don't work, depending on what you're looking for.

The thing is, you cannot rely exclusively on skills do do things in 5e - including the things skills are supposed to do.

The proficiency bonus is small, from +2 to +6. I do like the idea of smaller numbers in general, but this means your proficiency will only make the difference between success and failure 30% of the rolls, that's on level 17. For most of your characters' career, proficiency will only be useful about 20% of the time.

This causes other unexpected effects. For example, clerics are not that good at religion, since it is Intelligence-based. Nature is also based on Intelligence, making wizards better at it than rangers and druids. Wizards do not have the option of being great at arcana; the Arcane Trickster is probably better (if that is what the player wants), since he gets Expertise, even if the Wizard casts more powerful spells.

Fortunately, the designers know this - and they have taken several steps to "fix" it, by using different features other than skills.

A cleric can be good at religion - if he takes the right archetype. A Champion fighter can a "remarkable athlete", not because of his athletic skills or the (quite unremarkable) feature of the same name he gets on level 7, but because he gets more points to put in his abilities as he levels up. Even a wizard is decent in arcana, not only because of the skill, but because he will often have better Intelligence than other classes (even if they have the skill). Druids and rangers can be good at dealing with nature - by using the appropriate spells and features.

The rogue, being the best at skills, gets not one, but two fixes: expertise and "reliable talent" - because skills, by themselves, are unreliable.

Copyright: Wizards of the Coast
Of course, a Fighter gets better at fighting (which is NOT a skill or ability check, but worth mentioning here), not (unlike earlier editions) because he has a better chance to hit with every roll, but because he gets more attacks (as seen in some of the earlier editions), better criticals and other features that have nothing to do with attack bonuses (I do acknowledge this option has some significant upsides, such as allowing the level 20 wizard to have a good chance of landing a blow if he attacks a powerful foe with his staff, even if the damage is minimal; on the other hand it leads to hit point inflation instead of skill inflation, which deserves a different post).

As you can see, if you look at the big picture, the game works as intended. But, in order to keep the "bounded accuracy" idea, it has sacrificed some degree of simplicity.

If the designers had chosen to make proficiency equal to half-level (or half-level +2, of course) for some "expert" skills, or just let advantage from multiple sources stack, or just allowed more possibilities of expertise, you could do away with multiple class features at once - making the game a lot simpler.

On the other hand, this would make classes too similar to each other for some players. Unfortunately for me (since I like simpler systems), many people seem to prefer having different ways of relying on skills instead of having more significant skills (i.e., greater bonuses) to begin with.

Like many others, this is an aspect of 5e that works as intended, although it might look a bit strange at first sight. Even if you disagree with some of the designer's choices - and I often do - the results make plenty of sense.


  1. I'm definitely a fan of the small proficiency bonus - heck, if I'd designed the game, I might've gone with a +1 through +5 range instead!
    There is some wonkiness in which skills benefit from which stats, though, as your examples illustrate. (Personally, I think it has a lot to do with Wisdom's poorly-defined nature - like, what the jank is Wisdom, anyway? Somehow clerics and druids are supposed to be "wise" but not necessarily intelligent? *shrugs*)

    1. +1 to +5 would elegantly mirror ability scores. It might work, I think, if expertise were more common.

  2. The d20 inflates the effects of modifiers such, that if one were to map it onto a d% scale, it would become obvious that most modifiers are meaningless, and that the variable effect of Advantage/Disadvantage is only approaching the necessary weight on tasks. For systems like BRP (Chaosium and other d100 systems), where skill percentages can exceed 100 (permitting multiple attacks and so forth), this disparity between the truncated d20 scale grows even larger.

    I think that a lot of what you have been attempting to fix in 5e can only really be meaningfully addressed on a d% scale. ...and a system which has a significant number of skills, and where combat is skill-based, and where skill can lead to expertise and special techniques, etc.

    1. I agree, this works very well for BRP/Runequest etc.
      I think things can be fixed in 5e, you'd just have to use 3d6 or give everyone expertise...
      About combat, 5e tried to fix this with feats. They didn't succeed IMO - but I'm giving it a try!