Dungeons & Dragons Playtests Four New Mystic-Themed Subclasses

Dungeons & Dragons has dropped their first Unearthed Arcana playtest of 2026, with four brand-new subclasses being tested. Today, Wizards of the Coast posted a Mystic Subclasses Unearthed Arcana playtest to D&D Beyond, featuring four magic-themed subclasses. The new subclasses include the Warrior of the Mystic Arts Monk subclass, the Oath of the Spellguard Paladin subclass, the Magic Stealer Rogue subclass and the Vestige Patron Warlock subclass.

The Warrior of the Mystic Arts is a spellcasting subclass that grants Monks the ability to cast Sorcerer spells up to 4th level spells. The Oath of the Spellguard is designed with protecting magic-casters in mind, while the Magic Stealer Rogue targets spellcasting and can empower their Sneak Attacks with magic stolen from nearby spellcasters. The Vestige Patron Warlock forms a bond with a dying god, with the god taking on a vestige form as a companion. The Vestige companion grows in power with the spellcaster. Notably, the Vestige Patron draws inspiration from the Binder from past editions of D&D.

There's no indication when or what this new Unearthed Arcana could be related to. There are several Unearthed Arcanas not currently attached to an announced D&D product, although two almost are certainly tied to a Dark Sun sourcebook.

You can check out the subclasses here. Feedback opens for the playtest on January 22nd.

 

[Kurotowa]

It was leaked, then quickly removed.

Which is why you never trust leaks. Very often they're full of inaccurate or outdated information. I've seen this with video games, plans change all the time. They run out of time, they run out of money, the lead dev gets a hot new idea six months before launch and re-writes half the story, someone key suddenly leaves the project and the rest of the team has to work around that absence.

This is why when WotC is announcing books for the year, all they do is give a title and general subject. We don't get hard details on what's in it until a month before release when the book's already at the printer.

 

[Parmandur]

That's not exactly true. The Ravnica book sold in line with the other setting books (not as well as Eberron). The others sold from slightly worse (Theros) to very much worse (Stryxhaven).

Per Nate Stewart in early 2019, Racnica was at that time the fastest selling book in the 45 year history of D&D. You are, mp doubt, referring to the Books an data. The Bookscan data that does not include data from game stores where Magic fans hang out and buy Magic products...

It was leaked, then quickly removed.

The only information I have ever found was for the Lorwyn card set boxes that were delayed until Spring. There is no indication that the 32 page document for Lorwyn was cut down from something larger...though again, it would fit well into a larger print product that covered a number of Magic Settings in a broader framework.

 

[Paul Farquhar]

plans change all the time.

Indeed, which is why I cited this as an example of how plans changed. We can only guess as to why of course.

 

[Paul Farquhar]

The Bookscan data that does not include data from game stores where Magic fans hang out and buy Magic products...

And no doubt those, relatively small number of stores, had much higher sales than those overall.

 

[Parmandur]

And no doubt those, relatively small number of stores, had much higher sales than those overall.

The Wizards Play Network has over 6,000 stores. Former WotC employees have gone on record that the Books an numbers are woefully incomplete and don't give an entirely accurate picture, and a product aimed at Enfranchosed Magic players that sells in Magic card selling stores that cater to that audience would seem a likely place for the data to be skewed, particularly when the Brand Manager at the time says it was the fastest selling book D&D had in the first 45 years of the game.

 

[Paul Farquhar]

The Wizards Play Network has over 6,000 stores.

Which is a tiny fraction of online sales.

 

[Parmandur]

Which is a tiny fraction of online sales.

Not necessarily of the Bookscan sales, which have been established as being a fraction of what D&D is selling. Per the numbers that Teos Abadia ran, the Booksscn numbers for D&D books seem to be conservatively like a fifth of the total sales. And that is small enough for the sampling to be off, considering that people buying from Barnes & Noble don't necessarily have the same interests as the other 80% of customers. It is self-selected.

Heck, at this point, over 60% of D&D sales are apparently direct sales from WotC with no stores in between.