Monday, December 22, 2025

Archaeology: Dwarf Class for 5E

I had been playing RPGs for many years before I ever even considered getting into homebrew. Of course, I wasn't against homebrew. I routinely took homemade gameable content from my favorite RPG blogs and put them into the games I was playing. But I just never wanted to make it myself, y'know?

Boy how things change.

It was around 2016 that I started dabbling in game design, but even then it was all confined to 2014 D&D 5E stuff. That's when I came up with my first dungeoncrawling procedure, Advanced Darkness, some rules for 4E-style minions, and started working on my rules for mazes (all of which were essential houserules to bolt onto your 5E game if you ever wanted to run my megadungeon, I assure you).

Around 2019 I was getting into the really retro stuff, especially Greyhawk, and wanted to try making a "race-as-class" hack for 5E. The idea was that Dwarf, Elf, Halfling, and Gnome would all be made into a full class, levels 1-20. Any PC of a different class would therefore be a human (and should still get the human traits at character creation, obviously). I had lots of notes about how I was planning to create each one, but the only one I made serious progress on was the Dwarf.

You want to take a look together?

Overview

First, a link: this is unfinished, untested, and not very good. I don't stand by most of its design.

I made this using the Homebrewery, which is a really nifty tool for creating homebrew content in the style of 5E WotC products. I also referred to this guide on class design for 5E by Redditor u/JamesMusicus.

When I was a kid playing video games inspired by D&D, "race as class" was common. In Gauntlet you can play as a warrior, a wizard, a valkyrie, or an elf. In King of Dragons you can play as a fighter, a wizard, a cleric, an elf, or a dwarf. This made sense to me when I was young. 

When I was first getting into D&D and I learned that older editions had race-as-class, I assumed that it was like in those video games. When I got older and started looking into them, I was disappointed by what I found.

In my mind, D&D never actually had "true" race-as-class. From the beginning, you still chose a race and a class separately. It's just that your selection of classes was limited by your choice of race. If you play as a dwarf, then you're still just playing as a fighter, but with a couple extra dwarf-y traits to start with. And even that wasn't true anymore once they added thieves. In the video games I grew up with, dwarves and elves felt like their own unique classes, something totally different from human fighters and magic-users. I felt like this concept had still been unfulfilled by fantasy RPGs.

However, choosing D&D 5E as the basis for fulfilling this concept was, perhaps, unwise. In TSR D&D, the racial features were simple. "Dwarves get infravision, they resist magic, they can use this one really cool hammer that may or may not exist somewhere in your campaign, and they notice when architecture is fucked up." That's about it. And you also get all of your dwarf stuff up front, because it's pretty much just your inborn biological traits (or maybe cultural things from your upbringing).

But to make a 5E class, I have the burden of coming up with tons and tons of new features every level or two, all the way up to level 20. That's a tall order. I ended up making a list of every last dwarf trope I could find, pulling from it as needed just to come up with bullshit class features never before seen in any edition of D&D. I like leaning into weird biology whenever possible, to emphasize how not-human they are. But there's only so many new biological features you can accumulate over the course of your adventuring lifetime. You'll find that I very frequently just stole features wholesale from other existing classes.

And no, I did not account for multiclassing whatsoever. I've never been into it, and it feels especially weird in this context. So my plan was to just ignore it.


Deep Dive

Let's walk through the class and talk about each part, one at a time.

Level 1

First, I tried to write a 5E-style class intro description. I don't like it.

Next, note my first weird decision: I thought that the dwarf would be a good candidate for the fabled Constitution-based class, a popular subject of 5E homebrew at the time. Because no class in the game favors CON as their primary attribute, many had speculated on what a hypothetical CON-focused class would look like. I, like others before me, thought that they might drain their own HP to fuel abilities. This was my attempt at making that, also giving them a whopping d12 Hit Die (like the Barbarian).

For their 1st-level stuff, I just gave them all their normal racial traits. I worried that this wouldn't be enough, since human PCs would get both their human stuff and some class features at 1st level. But I threw in an extra ability for demolishing structures because I thought it would be cool. I'm sure that any amount of playtesting would have immediately revealed this power to be a total nightmare to DM for, as dwarf PCs just smash through a dungeon's walls like they're playing Minecraft.

Level 2

5E classes get a sub-class sometime within their first 3 levels, and for the dwarf I came up with "disciplines." I'll talk more about the these in the next section, since they were never finished. I do still feel like this is a good addition for a system using race-as-class, because it allows for some variety even within adventurers of the same species. The fighter-y dwarf should still feel meaningfully different from the cleric-y dwarf, y'know?

Level 3

Begrudged Enemy is actually stolen (I think verbatim?) from the first Unearthed Arcana attempt at revising the Ranger class. That's still my favorite version of the 5E Ranger, and I really like how it handled the Favored Enemy concept. Additionally, any class feature that contains another choice inside helps with providing more variety to members of that class, making our dwarves feel a little less monolithic.

Underdark Culture is a great illustration of why designing mechanics for 5E is a pain in the ass. In the kinda of rules-lite OSR games I'm into nowadays, the way you'd design this sort of thing might look like:

If you can invoke some part of your background that the referee agrees would be relevant to the test, they may allow you to roll with advantage. 

+

If you can invoke some part of your background that the referee agrees would provide you expertise, they may reveal to you additional lore beyond the common knowledge available to other characters.

And that's it. "Hey, I'm a dwarf." "Oh, right, yeah, I think you'd be more familiar with this. Alright, have a bonus." Trying to codify that idea within the rigid mechanical framework of modern mainstream games is such a headache. The funny thing is, in a more rulings-over-rules playstyle, all the benefits of this class feature could plausibly be stuff that a chill DM simply assumes you're entitled to from day 1 of playing a dwarf, without being written down anywhere.

Level 5

Sigh. Never found a good name for this one.

5E classes get a major bump to their average damage output at levels 5, 11, and 17. For spellcasters, their attack cantrips scale upwards at these levels. For Rogues and Monks, that's when their Sneak Attack / Martial Arts die goes up. For other martial classes, they usually get Extra Attack at level 5, then something more unique at 11 and 17 (except for Fighters, who just keep getting more and more attacks).

I didn't want to settle for a boring default option, so my attempt at a more thematic feature for their attack bump is Taxing Strike. Spend Hit Dice to do more damage. You've got health to spare, and I know you're probably not going to take enough short rests to use them all, anyway.

Near the end of the doc, in my scrap notes, you'll find lots of debate about how I could/should revise this feature. But with the benefit of hindsight, I think it's silly that I tried to impose any kind of limit on how many times you can use Taxing Strike per day. After all, isn't your HD pool already its own limit? No need to overthink it.

Level 7

Elemental Resistance is very straightforward. I've always enjoyed tying dwarves with the elements, and anything that makes them even more tank-y gives you more room to burn those HD without worry. 

Stand Your Ground is inspired by a dwarf racial ability found in both 3.5E and 4E, but which didn't make the cut in 5E. Rereading it years later, I definitely prefer the version I was debating replacing it with, where you instead simply spend your reaction to automatically negate the shove. Aligns better with the general framework of 5E combat, I think.

Level 9

Stoneskin is absolutely buck wild lmao. By this point, I was already a fan of systems that model weapon and armor durability, but no such rules existed in 5E. Trying to houserule them in may be worth it, but I didn't want my dwarf class to require something like that. This was my attempt to simulate "weapon breakability" into 5E using the mechanics that already exist.

The shitty thing about this is that a dwarf PC probably won't get to benefit from using it against their adversaries very often. Sure, they might get a -1 or even a -2 applied to their foe's sword, softening the blow of one or two attacks. But that enemy is going to be defeated pretty quickly anyway. An ability like this is actually way shittier for a player to deal with when they face an opponent wielding this effect, since they're actually going to be maintaining their gear over many sessions of play. Most bad guys only exist for 1-2 rounds of combat, y'know?

Level 11

Greater Grudge is taken from the same UA Ranger. By my math, the improvement to Taxing Strike averages out to a damage spike comparable to an 11th-level Paladin burning a spell slot to smite.

Level 13

This is when we start getting features that came from a desperation to fill the quota. "What else are dwarves known for? Uhhh runes, I guess??"

Rune Artifice is adapted from a couple features taken from the Forge domain for the Cleric class, which was added in Xanathar's Guide to Everything, probably also influenced by the UA version of the Artificer class that existed at the time.

I have no sense for how balanced this is, but I think it's pretty boring.

Level 15

The improvement to Stoneskin is nuts. My rationale was that a feature this powerful wouldn't be justified earlier than 15th level, but by the time you've reached this point in a campaign, you might not still be fighting many mundane enemies anyway.

Level 17

Elemental Flurry is another result of grasping for ideas. Building on the elemental thing by adding an offensive component seemed simple enough. I have no idea if the Taxing Strike buff would be balanced at this level and I don't feel like crunching the numbers.

Level 18

Wizened is one of the dumbest ideas in this whole thing. Okay, so, the rationale was that I wanted to ground most of the class features in the biology of the dwarf. Things that a human character could never get. And one of the most prominent physical distinctions between elves / dwarves and humans is their longevity. Can we make a high-level class feature that reflects their old age, creating a mechanical benefit to represent the experience a centuries-old character brings to adventuring?

The result is me literally just stealing the Portent ability from the Diviner subclass of the Wizard. Fucking lame.

Level 20

I think that capstone features should be ridiculous. Too many of the 5E classes have a disappointing capstone. Almost nobody will ever use them anyway, so why not get a little crazy?

That said, the "ribbon feature" component of Legacy of the Ancients was a lot easier to come up with than the actual mechanical benefit. And for that, I really did just get lazy and say fuckit, more damage output, why not.


Unfinished Subclasses

Sentinel

This one needs to serve as our basic "dwarf fighter" option. As you can see, that mostly just means stealing the basic mechanic from the Fighter class: the maneuvers from the Battlemaster subclass. This is almost entirely unchanged from its source, except that I was planning on swapping out a handful of the maneuvers for some dwarfier variants. For example, instead of "Evasive Footwork," dwarves would use "Bull Rush." Instead of "Disarming Attack," dwarves would use "Sundering Attack." Instead of "Parry," dwarves would use "Shield Splinter."

In addition, Sentinels get the rest of the proficiencies needed for a proper front-line tank build, plus yet another "I will fuck up my opponent's gear" feature that I thought would be exciting and fun (but would probably just be a pain for the DM to keep track of).

I also gave all three disciplines something they could craft, which is probably not very useful in most 5E campaigns but which feels appropriate for a class that's just about being a dwarf.

Idolater

This is my alternative to being a "dwarf cleric," but I really didn't want to make some kind of shitty 1/3 spellcaster like the Arcane Trickster or the Eldritch Knight. Instead, I opted for something wildly ambitious and misguided for a homebrewer of my meager experience level: designing an entire alternative magic system for use in just this one subclass alone. What could go wrong?

The basis of this was the Cleric and Paladin's "Channel Divinity" ability. What if you had a whole subclass based around fleshing out this one existing class feature? As I discovered, trying to write a magical class feature for a modern, crunchy game like 5E is exhausting. Trying to comprehensively answer every question that might come up, consider every limit necessary to keep it from breaking the game, phrase everything consistently with how other 5E mechanics are phrased, etc. And having to make a judgment call on all those various issues feels arbitrary and swingy and ugh.

As you can see, some of the miracles I had in mind were things like "turn undead," a healing aura, and light aura, a fear aura, a shockwave effect, etc. Offensive miracles often work best when you toss your idol into a group of enemies like a grenade. Support auras often work best when you hold onto them yourself, or hand them off to an ally to carry around.

In hindsight, having even just 3 of these active on a battlefield simultaneously would probably be a pain.

Prospector

Lastly, I wanted a subclass that 1) fit in all the other dwarfisms I could, and 2) didn't just feel like a miniature version of an existing class. But of course, the blank canvas is a tyrant.

On one my main inspirations was this Miner class for B/X made by Emmy Allen. I probably should have stolen the name, too.

After that, it falls apart. Their primary feature was going to be almost exactly the same as the Dungeon Delver feat (i.e. advantage to detect secret doors, advantage on saves to avoid or resist traps, resistance to trap damage) which is another lazy bit of design on my part.

As you can see, I had some thoughts around giving them a pretty simple bomb-based attack, with different types they could learn to craft (sort of an Arcane Archer-y idea), but I mostly wanted the Prospector to focus on utility. Detecting the presence of treasure, appraising its value, having a much higher carrying capacity for equipment, etc. These all make sense thematically, but they're so OSR, so irrelevant to 5E adventures, y'know?


Overall Thoughts

This was a positive learning experience for me, but it's also probably for the best that I never finished this little project. There's a reason you're only supposed to make subclasses for 5E, not full classes. I learned a lot about 5E's design that you don't pick up on just by playing the game every week for years. There's important elements of the writing you only notice by trying to replicate it very authentically.

It's easy to criticize bad design, but it's so, so, so difficult to produce good design. Every feature I added, I had to consider:
  • Does this fit into the greater mechanical framework of D&D 5E? Not just ability checks, skills, and advantage/disadvantage, mind you. I mean, does it fit into the bigger picture of the adventuring day and the resource cycle? Does it concern itself with the factors relevant to typical gameplay experiences players will actually play out in this system? Does it adjudicate through codified rules the parts of the game that this system normally handles via codified rules, rather than the parts that are left freeform?
  • Is this phrased in a way that's consistent with similar rules throughout the rest of the game? Does it coin a new keyword when appropriate?
  • Do the numbers fit into the right balance of numbers relevant to this? Is the damage output right? The number of uses per day? Are there similar abilities elsewhere I can compare this to?
  • Will it actually be usable during play? Will it be a pain in the ass to adjudicate when it comes up? Does it force people to track information they otherwise wouldn't even think about?
  • Does it reinforce the themes of the class? Does it lend itself to a playstyle that feels appropriate to the fantasy it describes?
  • Does it have a clear role to play in an adventuring party? Is the class versatile enough to accommodate a variety of distinct builds, yet focused enough to have clear mechanical identity?
  • Does it make any goddamn sense? Can you understand how to use it when you read the instructions? Are there any ambiguities or game-breaking exploits?
Almost every feature I came up with fails on several or all of these considerations. The ones that don't were literally just copy/pasted from existing material in 5E anyway. And yet at the same time, one of the most important lessons I learned is how to recognize which of those considerations were worth keeping and which were holding me back.

Keep making sloppy art, friends. DIY art is a beautiful thing. And check out the beta of Rivers & Lakes, our wuxia tactical combat game.


-Dwiz

No comments:

Post a Comment