When Dungeon World was released in 2012, it slammed the door to Powered by the Apocalypse open so hard it broke the hinges. By taking the recipe crafted in the Baker House and mating it to the memetic power of Dungeons and Dragons, suddenly everyone could see what was so powerful about PbtA. Of course, Dungeon World was hardly a perfect recipe. Using the architecture of ‘moves’ established in PbtA but keeping both the stats and classes of D&D made for an incomplete match, and some of the mechanical choices made to get the two to pair up have received more significant criticism now that the design community has had a good decade and a half to really figure out what PbtA is. Still, the combination of a solid foundation and a lot of good ideas made Dungeon World into a rare specimen: The commercially successful fantasy heartbreaker.

Here in 2025 most people don’t talk much about Dungeon World anymore. Co-designer Adam Koebel made a right fool of himself on a livestream (it’s worse than that but I don’t feel like going into it), and PbtA has blossomed and fractured into a number of sub-designs. There are a couple of wrinkles to this story, though. For one, a second edition of Dungeon World is in development, spearheaded by none other than Burning Wheel’s Luke Crane. And for two, everyone and their mother still seems to want an indie alternative to D&D. You know, the same dungeon crawls and all the fun memes, just without anything designed by Wizards of the Coast. Into this environment comes a new game which started with a solid Kickstarter campaign and only grew from there. The designers weren’t conceited enough to claim to be the next Dungeon World but they definitely wanted to be, and with some solid post-campaign momentum (#1 on DriveThruRPG as of this writing), it’s not entirely out of reach. I am of course talking about Grimwild.

One thing that I think gives Grimwild a lot of potential (and from that a good amount of potential staying power) is that the designers have understood the assignment in a way I’m not entirely sure Sage and Adam did in 2012. Dungeon World was successful in part because the tropes and storylines of D&D are in demand regardless of the underlying mechanics. When it comes to engaging with mechanics, though, you’re serving a fairly demanding group of players: You both need to provide the experience that they’re envisioning (heavily influenced by D&D), while also not providing any undue friction towards getting to that point. And, just to make it that much more complicated, you’re also going to be constantly bombarded with the question of why your game works better than D&D, and when the question asker is a diehard 5e player, that’s a tough one to answer. 

Game mechanics

Grimwild’s core mechanics are derived from Blades in the Dark. The player rolls a pool of d6s, looking for the highest result in the pool. A 6 in the pool means the result is ‘perfect’, a 4 or 5 means the result is ‘messy’, and a 1, 2, or 3 means the result is ‘grim’. These three results map to the PbtA/FitD results, moving from a clean success to a partial success or success with complications and then to a ‘miss’. For advantageous conditions a player can add dice to their pool; most commonly this happens because of an assist from another character or a player spending ‘spark’ (a meta-currency described as ‘pure protagonist energy’ in the text). For disadvantageous conditions, the mechanics are a bit different. Depending on how difficult the roll is, the GM can add ‘thorns’ to the roll. A thorn is an eight-sided die, and if the die rolls a 7 or an 8 then the result is stepped down or ‘cut’ (from ‘perfect’ to ‘messy’ or ‘messy’ to ‘grim’).The thorns also interact with two other roll results, ‘crits’ and ‘disasters’. In the event that a thorn makes you step down a ‘grim’ result, you get a ‘disaster’, which is essentially a fumble. The roll goes from a miss and a complication to the worst possible outcome occurring. On the other hand, if you roll two or more 6s in your pool, it’s a ‘crit’ and, in addition to a couple specified mechanical effects, you’re also immune from having your result get ‘cut’. Having five distinct die results makes for a fairly complex die roll, though it’s both nowhere near as particular as something like Cortex Prime and, due to the FitD resolution, requires no math.

The core mechanics on the GM side of the table are clearly structured like those in PbtA, albeit with significantly more mechanization. While each PbtA game varies in particulars, the two broad constructs around which most GM-side mechanics are built are GM Moves and clocks. GM Moves specify what the GM can do and when, creating a clear exchange of narrative control. Clocks are used to both organize the threats and other narrative aspects of the setting while providing information on the timing of certain events coming to pass. In Grimwild, both of these are firmly mechanized, giving clearer guidance on how to use each element. GM Moves in Grimwild are broken out into Story Moves, Suspense Moves, and Impact Moves. A PbtA veteran would immediately note these three categories essentially move from ‘soft’ to ‘hard’, but what makes this delineation important is that specific rules call for specific types of Moves in response. Story Moves also have complementary Story Rolls, which allow a GM to leave certain circumstances up to the dice instead of having to choose everything behind the scenes.

Clocks are replaced by Diminishing Pools, a mechanic I rather like. Each pool is a certain number of d6s, and whenever a Move is triggered that affects the status of the ‘clock’, the GM rolls the dice in the pool. All dice which come up 1, 2, or 3 are removed from the pool, and once the pool is empty then something is triggered, either in the fiction or mechanically. Diminishing pools look more like clocks in some cases (a 6d pool indicating when guards arrive on scene) and more like usage dice in others (the Fighter’s ‘bulwark’ ability, which allows you to ignore damage so long as there are still dice in the pool), but I think it works very well as a unified mechanic that’s easy to understand and adapt for different circumstances.

The game’s mechanics expand into what you’d expect for a dungeon adventure game: combat and damage rules, an exploration system, and of course character creation including a full (and very familiar) set of class options. When considering how well these work, though, it’s best to look at them in context. Grimwild isn’t just trying to be a fantasy game, it’s trying to fit the mold of D&D, just like Dungeon World did before it.

Whither Dungeon World?

The Grimwild designers are calling their core rules the Moxie Ruleset and, as seen above, it offers some improvements over PbtA and FitD for more delineated, action-driven gameplay. While I’d argue this ruleset is more of its own thing than, say, Wildsea, it’s still very much in the family and still following in the ‘D&D but PbtA’ mold that was cast by Dungeon World and then followed by others like Root and Fantasy World. I personally think Grimwild is the best executed of this type that I’ve seen so far, though the reasons for this aren’t strictly mechanical.

The reason that Dungeon World succeeded like it did, really, boils down to one thing: Dungeon World cast D&D character classes as PbtA playbooks. It was by far the most prominent area of mechanical alignment between D&D and Apocalypse World, and also was the smartest way to enable the same sorts of character decisions that most people think of when they think D&D (cementing your character through race and class, choosing new and interesting character abilities when you ‘level up’). The choice to use the same six core stats helped from a marketing perspective, but when it came time to play the game I think the issues with execution hampered much of the potential upside.

Grimwild makes its throughline back to D&D classes as clear as absolutely possible. All of the classes have the exact same name save the Barbarian (here a ‘Berserker’), and all of the classes have a central, defining class feature that makes them unique. It’s also worth noting that the magic classes here are artfully done; the sorcerer, wizard, and warlock gain access to their magic abilities in completely different ways and give different hooks into the world (from eccentric wizard colleagues to wild sorcery to your warlock pact). It is fairly important that there is a big, distinct ‘thing’ for each class, like a lot of PbtA games there are many opportunities to “take a move from another playbook” so there needs to be a balance between choices and character differentiation.

Many of the other key mechanics lean into a PbtA trope, that of player-driven setting building. While backgrounds are mostly flavor they give Wises, which like the identically-named Burning Wheel mechanic allow players to declare key setting details and give access to relevant information (and, in some cases, permission to make a roll). Bigger than this, though, is exploration. The exploration mechanic is a significant optimization of the map-making rules from Dungeon World but also gives some of the map-making power to the players, a rule that may not be to taste for every table. Whenever a map needs to be expanded, the GM makes the appropriate move (called Expand the Map) and gives everyone (including themselves) 3 tokens, which are spent on adding pointcrawl-style nodes and paths to the map. Players can also earn these exploration tokens in other parts of play, but when Expand the Map is used, all tokens must be spent. What’s a little wild to me is that these rules apply to the region map, to settlement maps…and also to dungeon maps. Yes, the players are partly responsible for mapping out the dungeon, which is an aggressive choice even within the realm of player-facing mechanics.

To be clear, these exploration mechanics are all optional; it would be entirely possible to port in an existing hexcrawl or other map and use it to play Grimwild. But when taking them as part of the entire experience, it’s clearer that Grimwild is positioning itself to be a sandbox game, to give opportunities to have a more PbtA-like game than D&D would give them. Dungeon World was designed in a similar way, minus the player control; you started writing a Dungeon World map with the location of the encounters of session one (usually a dungeon), and then one nearby Steading where characters could rest and resupply. Everything else was drawn out organically as the game continued. It’s worth noting that Grimwild doesn’t really bother with the sorts of rules that Dungeon World had for steading tags and steading-steading interactions; it replaces this set of emergent mechanics with handing the pen over to the players. In a way, this is more aligned with the PbtA ethos than what came before; it’s not the only place where Grimwild both improves upon Dungeon World and also goes just a bit more ‘indie’.

---

This discussion was more about Grimwild than Dungeon World, but through contrast it helped me understand where Dungeon World really sits among PbtA games. Going back and reading Dungeon World probably around a decade after I had last played it really did impress upon me that it’s an odd duck. There are moves and playbooks, albeit saddled with the D&D stats. The use of other PbtA rules in places they weren’t really meant for led to some incredible (and potentially unintended) crunch; having every monster get a threat instinct was not really a translation of Apocalypse World, nor was taking the settlement detail reserved for the hardholder and applying it to every village, town and city. Although critics talk a lot more about the interfaces, like using moves in combat or writing classes as playbooks, upon rereading Dungeon World it was clearer to me that the appropriation of Apocalypse World mechanics therein was done with a very, very different intent than the design of Apocalypse World.

Note that I didn’t say that mechanics appropriation was done poorly. One thing that strikes me about Dungeon World is that it reads, density-wise, kind of like Torchbearer? Not a one-to-one alignment of course, but enough of an alignment that the current fate of Dungeon World is very, very interesting. Dungeon World’s second edition is being designed by Luke Crane, and enough time has passed that there’s a lot of newer PbtA design work and theory that could apply to the project. At the same time, I’d think that Dungeon World as a continuing product may benefit from some consideration of deliberately trying to be a very dense PbtA game. As much as the rules probably could have been written better I don’t see the density as a problem…if done well with the core issues resolved, the second edition of Dungeon World could be a fantastic vehicle for emergent sandbox gaming, the exact sort of thing that Worlds Without Number wants to be but can’t so long as it’s still effectively D&D. As far as Luke Crane designing within the PbtA sandbox…we shall see. I have a copy of Miseries and Misfortunes, Luke’s D&D-based game, sitting on my desk; I’m reserving any judgment until I read it more closely but it’s at least an indicator in the broadly right direction.

Whenever the second edition of Dungeon World comes out, I don’t necessarily think it’s going to compete with Grimwild. Grimwild has internalized years of design theory around PbtA, including Blades in the Dark and its fork of the system. It’s also internalized years of D&D and fantasy gaming trends, and as a result it seems uniquely well-suited to delivering on the moment-centric, character-centric play which dominates RPG discourse (due both to gaming and Actual Play). Ultimately, I think that’s why Grimwild is going to succeed; it’s not the next Dungeon World. It’s not an indie version of D&D, it’s a version of D&D that’s delivering on the playstyle and the tropes of D&D in a distilled and easy to grasp package. The really indie stuff in Grimwild, like player-facing dungeon design, is probably going to turn some people off. That said, it’s easier to exclude or houserule than virtually any of D&D’s worst sins, and certain groups will end up discovering they like it. Grimwild hits differently than Dungeon World did in 2012, in part because the environment has changed. Daggerheart is a mainline D&D competitor, and although Grimwild is significantly closer to a PbtA game than Daggerheart, the jump is still an incremental one. So no, Grimwild isn’t the next Dungeon World. Grimwild is going toe to toe with all of the other dungeon games, and no one’s going to send it to the kid’s table because it’s “indie”.

Like what Cannibal Halfling Gaming is doing and want to help us bring games and gamers together? First, you can follow me @LevelOneWonk@dice.camp for RPG commentary, relevant retweets, and maybe some rambling. You can also find our Discord channel and drop in to chat with our authors and get every new post as it comes out. You can travel to DriveThruRPG through one of our fine and elegantly-crafted links, which generates credit that lets us get more games to work with (which is eactly what we did here)! Finally, you can support us directly on Patreon, which lets us cover costs, pay our contributors, and save up for projects. Thanks for reading!