So, you want to build something for the new Player's Handbook? Honestly, I get it. The 2024 update to Dungeons & Dragons basically hit the "refresh" button on ten years of design. Some people are calling it 5.5e; others just say "the new core." Whatever you call it, the way we used to slap together a homebrew subclass has fundamentally changed. You can't just copy-paste a 2014 template and hope it doesn't break the game.
The math is tighter. The "swing" of certain levels is different. If you’re trying to create a 5e 2024 subclass homebrew guide for your table, you've gotta start by unlearning some old habits. Specifically, the "Level 1 dip" is dead. Gone. Buried. If you're building a Cleric domain or a Warlock patron and you put the big, flashy identity-defining feature at level 1, you’re already behind the curve.
The Level 3 Rule is Absolute
In the 2014 rules, Clerics and Sorcerers felt their subclass from the moment they stepped into the tavern. Now? Everyone waits until Level 3. This is the biggest hurdle for homebrewers who want to keep that old-school flavor.
If you're designing a "Path of the Neon Dragon" Barbarian or a "Circle of the Urban Jungle" Druid, your players aren't touching those mechanics for the first two levels. It feels weird, I know. But it’s there for a reason. It stops the "one-level Hexblade dip" that plagued every Charisma-based build for a decade.
When you're drafting your features, you have to pack a lot of "identity" into that Level 3 slot. It needs to be the "who am I?" moment for the character. For a Fighter, this is where you might give them a new way to use their Weapon Mastery. For a Rogue, maybe it’s a specific Cunning Strike option. Basically, if the Level 3 feature doesn't feel like a major pivot in how the class plays, the subclass will feel generic.
Mapping the Progression
Every class now follows a standardized skeleton. You can't just guess when the next feature comes. If you put a feature at level 5 because "it feels right," you’re going to mess up the power curve.
Class Feature Gaps
- Barbarian/Druid/Warlock/Wizard: 3, 6, 10, 14.
- Fighter: 3, 7, 10, 15, 18.
- Paladin: 3, 7, 15, 20.
- Ranger: 3, 7, 11, 15.
- Rogue: 3, 9, 13, 17.
- Cleric: 3, 6, 17. (Wait, level 8? No, that's Blessed Strikes now.)
See that Rogue jump? From 3 to 9. That is a massive gap. If you’re homebrewing a Rogue subclass, that Level 3 feature has to carry the player through the entire "Early Tier 2" play without them getting bored.
📖 Related: Why Candyland Couture Dress to Impress is the Hardest Theme to Get Right
The Death of "Magic Weapon" Features
Here is a mistake I see in about 80% of homebrew: giving a feature at 6th level that makes your attacks "count as magical for the purpose of overcoming resistance."
Stop. Don't do it.
The 2024 rules have moved toward giving players Force or Radiant damage instead. If you look at the new Monk (Warrior of the Elements) or the Druid (Circle of the Moon), they aren't dealing "magical bludgeoning" anymore. They’re just hitting you with pure energy. It’s cleaner. It bypasses the old debates about what "magical" actually means in the context of a non-spell. When you’re writing your 5e 2024 subclass homebrew guide notes, replace "counts as magical" with "deals Force damage." It’s punchier.
Synergy with New Core Mechanics
Your homebrew needs to talk to the new systems. If your Fighter subclass doesn't interact with Weapon Mastery, why does it exist? If your Sorcerer doesn't care about Innate Sorcery, it’s going to feel like a 2014 fossil.
Think about Cunning Strike for Rogues. A great homebrew Rogue subclass for 2024 shouldn't just add damage; it should add a unique choice to that list. Maybe your "Poisoner" Rogue can spend 2d6 of their Sneak Attack to force a Constitution save against the Blind condition. That feels like it belongs in the 2024 book.
The Problem with "Free Actions"
Don't give out free actions. Just... don't. The 2024 design team worked hard to clean up the action economy. Most things are now a Bonus Action or part of an Attack Action. If your homebrew says "as a free action, you can teleport," you’re creating a balance nightmare. Stick to the "once per turn" or "using a Bonus Action" limiters. It keeps the game moving and prevents one player from taking a 20-minute turn while the Wizard checks their phone.
Let’s Talk About Scaling
We used to scale things based on "Modifier times per Long Rest."
The 2024 PHB is obsessed with Proficiency Bonus or just flat numbers.
Look at the new Goliath or the Musician feat. They use Proficiency Bonus (PB) to determine uses. This is great for multiclassing, but be careful. If a feature is too strong, scaling it with PB might make it a "must-pick" for every other class.
If you want a feature to feel like it truly belongs to that class, scale it based on the class level.
"You can use this feature a number of times equal to half your [Class] level (rounded up)."
That way, a level 20 Fighter gets 10 uses, but a level 3 Fighter with a level 17 Wizard buddy only gets 2. It rewards staying in the class.
👉 See also: How to Actually Get the Power Cell Ancient Armory Armor in Horizon Zero Dawn
The "Feel" Test
A good homebrew subclass should solve a problem or fulfill a fantasy that the 48 official subclasses don't.
Don't just make "The Fire Wizard" but better. We have the Evoker.
Make "The Ink-Mage" who uses their spellbook as a physical weapon.
Make "The Oath of the Commoner" Paladin who smites with a pitchfork.
When you're finishing up your draft, ask yourself: "If I showed this to a player who only owns the 2024 PHB, would they be able to find where it fits?"
Practical Steps for Your Homebrew
- Identify the Core Loop: What does the base class do every turn? (e.g., Fighter attacks). Your subclass should enhance that loop, not ignore it.
- The Level 3 Anchor: Make it big. Make it loud. This is the character's new identity.
- Use the Glossary: The 2024 PHB has a massive glossary. Use terms like D20 Test, Heroic Inspiration, and Slowed. Don't invent your own terminology for things that already exist.
- Power Budgeting: If you give a strong defensive boost at level 6, the level 10 feature should probably be utility or social. Don't stack "Best in Slot" features back-to-back.
- Playtest at Level 3 and 10: These are the most common break-points. If it’s not fun at 3, no one will play it. If it’s broken at 10, the DM will ban it.
Creating content for this new era is honestly a blast. It’s like getting a new set of LEGOs that mostly fit with the old ones but have some cool new joints. Just remember: keep it simple, respect the level 3 start, and always, always check the math against a Battle Master or a Life Cleric. If yours is twice as good as the "gold standard," it's time to nerf it.
🔗 Read more: Papa Louie 2 When Burgers Attack: What Most People Get Wrong
Next Step: Take one of your old 2014 homebrew ideas and try to map it onto the new 3/6/10/14 progression. You’ll probably find that you have to combine two old features into one Level 3 "Super-Feature," and that’s exactly what the 2024 design is all about.