Earlier this month, Darrington Press released the free playtest version of Daggerheart, their traditional fantasy RPG meant to go toe-to-toe with D&D. First with Pathfinder but now also with entries from MCDM and Kobold Press, we’re getting an awful lot of D&D-alikes, thanks to last year’s saga with the OGL. It’s now clear that a corporate game is a liability, so anyone making a livelihood in the gaming space is clearing out of the Halls of Hasbro. What makes Daggerheart, the entry from the Critical Role folks so special? I downloaded it for free, for one thing. In all seriousness Daggerheart is entering the public eye a little earlier than the MCDM RPG or Tales of the Valiant, both of which are currently fulfilling crowdfunding and doing any additional playtesting either contained to backers or within their own teams. The public playtest process is a great way to get a lot of feedback, and it’s worked well in the past; both 5e and the second edition of Pathfinder went through public playtesting.

It’s also caused some grief already. Darrington is somewhat in the crosshairs, between the moderate reception to their first game Candela Obscura and the relatively polarized fanbase that Critical Role has created by being the biggest voice in the room. Seems like a perfect time for someone like me to come in. I’m not the most impartial judge, given my growing disinterest in D&D or its cousins over the last five years, but I do understand what these games are trying to do. To that end, Daggerheart seems to have what it takes to grow a fanbase. It just needs to solve a few niggling issues with its own relationship to narrative mechanics first.

What’s in the playtest packet

The playtest materials that are being distributed for Daggerheart come with a sheaf of different documents, but there are three core elements here that are then subdivided or reproduced for ease of use. First off is the core manuscript. This is most representative of a traditional ‘sourcebook’ and is also the least polished in terms of layout; I think it’s pretty clear to readers that the manuscript does not and is not intended to look anything like the final product of the rulebook. I note this not as a criticism, but because it stands in stark contrast to the other two core elements. The character class packets are significantly closer to the look of a published product; the character sheet is laid out in a manner intending to emulate those from modern D&D, and the play aids and cards are all in what could be considered ‘final layout’. The final core element are the cards. Daggerheart clearly wants to provide cards as a way to provide flexibility to characters and cut down on copying abilities out of the book; I’m not necessarily against this so long as providing a printer-friendly or even text-only version of the cards continues to carry through into the final product. For now, GMs should know that reading the cards is necessary for getting their full text; the playtest manuscript does not currently reproduce all of them (which for playtest material makes sense). The cards cover character options like Ancestry and Community, as well as the Subclasses and Domains; I’ll discuss what these actually are when I discuss the breadth of character options. There is also a quickstart adventure and 5 pregens; given my interest I’m probably not going to read these but they are there.

It’s also worth noting the things that the current playtest material (this preview is built on the March 12th version) does not include. In part 5 is a list of empty chapter headings; planned sections of the text that are currently unpopulated and will either come later in the playtest or be held for the final product (more likely in my opinion). These headings are about homebrew, about writing custom versions of character options, equipment, and enemies. For this section, I’ll note three things. First, I’m very glad there are plans to put these items in the final product, as perceived hackability (perceived, not actual) is an important competitive element for D&D. Second, the execution of these sections will make or break them; good customization frameworks are harder to write than mechanics and usually done poorly, so there’s nothing to be gained by judging the mere intent to include them. Third, I completely understand why they aren’t included in a playtest.

Beyond what’s here and what’s explicitly outlined as coming later, we can only really guess at the differences between playtest material and final product; the class and heritage lists should probably be seen as final but that depends on playtest feedback as well. As I examine the core mechanics I’m going to do my best to acknowledge when something’s not final (like the layout), but for the most part I’m operating on the assumption that the broad strokes of the game are done. While there may be tweaks, it’s pretty safe to say that this game has six stats, race/class/subclass design, and uses 2d12 as the core dice mechanic.

Core mechanics

While it would be cynical to assume anything because of it, I note off the bat that 2d12 is a dice combination that is not sold in traditional RPG dice sets, unlike 1d20, 3d6, and 2d10. I doubt the decision was made expressly to help sell dice, but it probably doesn’t hurt. The 2d12 roll has a player roll two distinct dice and add the results, looking to exceed a target number. The twist here is that the two dice have labels, Hope and Fear. If the Hope die is the larger die, the player is rolling ‘with Hope’, and if the Fear die is larger the player is rolling ‘with Fear’. If the dice are the same, the roll is a crit (and all crits are positive). The play guide lists the results of ‘success with Hope, ‘success with Fear’, ‘failure with Hope’, and ‘failure with Fear’ in a manner that looks somewhat similar to a PbtA-like success gradation. That said, this isn’t really anything like PbtA dice mechanics and I think the text of the play guide should be modified in that respect. Given that one die is Hope and one die is Fear, the dice coming up in the order they do is an independent event (that is to say the probabilities of rolling a given number ‘with Hope’ and ‘with Fear’ are equal), and this should not be treated as gradations of success given that there’s no dice manipulation in the mechanics. Instead, Hope and Fear should be treated like Advantage and Threat in Genesys; luck that occurs outside the core result, and for either good or ill. This isn’t a mechanical distinction; rolling ‘with Hope’ and ‘with Fear’ could do the exact same things, but how you write that will greatly change player engagement. “Sometimes luck won’t be on your side” is very different from “your successes will just be worse arbitrarily 50% of the time”.

The actual mechanical impact of these dice rolls is modeled through meta-currencies. Hope goes to players and is used to activate a number of abilities, while Fear goes to the GM and can be used to both activate adversary abilities and set circumstances against the characters. The closest mechanical comparison I have for Hope and Fear is Momentum and Threat from 2d20, and I don’t think it’s a bad starting point to use. Both GMs and players have a maximum amount of Fear and Hope they can have, which is a good way to ensure that the points get spent (earning one or the other every roll also makes them seem less scarce).

The roll modifier mechanics are not that far from D&D. Attributes are straight modifiers (i.e. 0, -1, +1 instead of 10, 8, 13) which are added to applicable rolls, and then by spending Hope a player can also add one of their Experiences (more on those below). Beyond that there’s advantage and disadvantage; instead of roll two keep one these add or subtract a 1d6 to the roll. Less elegant, but still designed to reduce exogenous modifier counting. Speaking of modifier counting, players are encouraged to grab tokens for their dice rolls to remind them to count their modifiers; an interesting idea that’ll probably help viewers of a game stream more than players at a table. Another interesting slant here is that while this game has parallel rolling for the GM like D&D, the GM rolls a d20 instead of 2d12. This both has the consequence of giving player characters a slight statistical boost as well as clearly broadcasting that the GM doesn’t generate Hope or Fear.

Like any good D&D competitor Daggerheart has most of its specialized rules around combat, and some of these are interesting. Daggerheart uses range bands instead of measured distances, a needed change for any system that purports to encourage theater of the mind play. The damage and armor system is interesting too. Daggerheart has moved away from an integrated ‘armor class’, a wargame shorthand that’s been typical in D&D forever. Instead, characters first have their Evasion, a flat target that enemies must roll over to hit them. Then, armor can reduce incoming damage by the armor’s rating…if the player chooses to spend one of their armor slots. Armor as resource management is only one of the big changes to combat math, though. Weapon damage ratings are similar to those in D&D, but damage acts completely differently. First, damage is partly decoupled from hit points. Sure, you may deal 2d8 with your sword, but how that affects a monster depends on a few things. Characters and monsters alike get three damage thresholds: minor, major, and severe. If the damage total exceeds the minor threshold, it deals one hit point of damage. If it exceeds major, two, and severe, three. Characters have six hit points to start, so these wound definitions make one-shotting a healthy PC literally impossible at any level (there is an optional fourth threshold at double severe, but that would still only deal four hit points of damage). It also makes starting characters significantly less, well, squishy, which is either a significant improvement or a disappointing flattening of the curve depending on your view of wizards having ¼ the starting HP of fighters. If an attack hits but doesn’t at least match the minor threshold it deals stress instead; stress is the classic mental track to parallel the hit point physical track and can be depleted by events other than combat as well as using abilities. The key thing with stress is that if you run out and must mark stress, you mark a hit point instead. This doesn’t exactly create a death spiral, per se, but it does mean that tired characters are at a much greater risk of getting hurt.

Another key mechanic here is initiative, and in this case the PbtA mechanics are popping up again, but to better effect. While turn order is freeform and spotlight driven, whenever a player goes they place a token on the Action Tracker, which is provided as a card. The GM can then take tokens for enemy actions, balancing out PC and adversary actions. The GM can also spend Fear to add tokens, so there are chances to manipulate this a little bit. I personally think it’s a neat idea to add a little more structure to combat, and there’s even potential room to use the mechanism as a way to track player spotlight as well. That said it’s much less structured than the equivalent in D&D, and someone who hasn’t seen freeform spotlighting work in a PbtA or light OSR game will likely bristle until they try it.

I’m going to return to armor slots for a moment, because there’s an interesting shift in how the game treats armor and weapons. First, it’s notable that the hit point thresholds and evasion are not derived from stats, they’re based on character class. This means that there is no optimization in character creation that affects survivability. Second, armor slots and weapon proficiency are inherent to the character, not the item. The armor and weapon items determine the amount of damage an armor slot counters and base damage dice, respectively, but slots and proficiency are arguably more powerful than the items themselves. Armor slots indicate how many times you can use your armor before it can be repaired, so earning another armor slot (which you can do as a choice when levelling up) is significant. Proficiency is arguably even bigger; while a d8 is better than a d6 and a d8+2 is better than either, Proficiency modifies how many of those dice you roll. So sure, a level one character with a longsword will do 1d8 damage, but at level 5 with a Proficiency of 3, that same character is doing 3d8. That’s huge, and it makes martial characters feel more effective as they advance in a way 3/3.5 never really managed.

Discussing armor slots and proficiency starts to get into a big part of character options: Options for levelling up. The game provides a wealth of options for both character creation and advancement, but how they land and how effectively these options are all used varies from mechanic to mechanic.

Character options

There are a ton of character options in this game. Whether the breadth of options is effective or well-designed requires a bit more scrutiny, but there is definitely a lot of different possible character builds. Let’s start by looking at the direct D&D parallels: Class and Heritage. The classes are straight-up D&D classes, with some names changed to…maintain copyright? Confuse people? I’m being slightly unfair here; shifting ‘Fighter’ to ‘Warrior’ is completely whatever, but Guardian and Seraph represent somewhat different interpretations of the paladin and cleric, enough so that I’m willing to concede there isn’t a 1:1 mapping. That makes for nine classes compared to D&D’s 12; the monk, warlock, and barbarian are those without direct (or close enough) equivalents. When you get to Heritage, though, things go a bit fractal. D&D has nine baseline races (more in supplements) and a lot of backgrounds which are fairly minor as far as their impact on character builds are concerned. The Daggerheart playtest has a staggering 18 ancestries, which cover most D&D options (no half-elves or gnomes, full orcs instead of half-orcs) and then adds giants, goblins, cat people, frog people, mushroom people, turtle people, monkey people, constructs, fauns, faeries, and firbolgs. The other half of Heritage, Community, is less of a direct equivalent to Background and more of a counterpart to Ancestry that gives a similar mechanical detail based on the sort of cultural background you grew up in. There are nine of these, representing academic communities, affluent communities, criminal communities, literally underground communities, and others. This is neat, and frankly something D&D should have considered years ago. We’re not getting rid of the potentially loaded issue with ancestries (races in D&D, mind you), but at least there’s acknowledgment of cultural diversity here and some more interesting options to reflect where a character comes from. The actual parallel to D&D backgrounds come in the form of Experiences, which are freeform traits intended to substitute for a skill list, giving bonuses to tasks that align to the Experience.

Getting into Domains and subclasses, the design intent starts to become confused. The very first criticism I heard about Daggerheart was a complaint that the classes only had two subclasses each. Considering that D&D classes had zero subclasses for the majority of the game’s history, that seems to be the RPG equivalent of complaining that an Olive Garden doesn’t have as long a menu as a Cheesecake Factory. I also have an issue with the subclasses, it’s not based on how many there are but rather how they’re executed. Class abilities are broken down by two different subdivisions: Subclass and Domain. My main question is why. The Domains provide the bulk of abilities to each class; each Domain has 21 ability cards broken out over 10 levels, giving each class 42 abilities to play with as they progress. Why aren’t the Subclasses at least aligned with the Domains? For the Sorcerer, why isn’t the Primal Sorcerer ‘Midnight’ and the Elemental Sorcerer ‘Arcana’? The Subclasses seem like a somewhat underwritten way to shoehorn in several ‘big concepts’ for each class, like making sure that Ranger could mean both tracker and animal companion dude, and a Bard can go hard on either music or lore. And as I’m reading these, those are the strongest pairs, really. If the whole Subclass to-do was folded into the Domain rules, if those big class-specific starting abilities were still there but without an additional label, it would have been more cohesive and given an opportunity to make multi-classing even more interesting. As far as the Domains, I like those a bit more. Each class is a mix of two Domains, which means that classes have overlapping lists of abilities. That makes a lot of sense to me. Additionally, multiclassing by adding one of the new class’s Domains onto your two starting ones is a really cool way to both handle the inherent balance issues of multiclassing as well as have it create optionality rather than make the classes blur together as everyone dips between multiple.

Game mastering

Daggerheart has more complete GM guidance than D&D but that’s not saying much. Advancement, for instance, is completely milestone-based, which is a weird pick for a game that also provides granular level-by-level progression choices broken out into three tiers. The game does seem to have a much better grasp on how math should or should not affect combat, and doing away with any notion of encounter-based progression is undoubtedly a good thing. On the other hand, characters have background questions, connections to other party members, and Experiences that represent swathes of relevant skill. There’s plenty of room to add things like session questions, drives, something that makes a player think about what their character is doing. Almost anything is better than “I don’t know, whenever you feel like it”.

On the opposite side of the coin is what I call “Apocalypse World Ctrl-C”. I don’t intend this to mean that Daggerheart plagiarized Apocalypse World in any way. What I do mean is that a large swathe of the GMing section was lifted from a PbtA context and dropped into this hyper-trad heroic fantasy monster slaying context and there was some translation that never happened. First, the good. The adaptation of Agendas and Principles, the distillation of the job that the GM is trying to do, works very well and honestly I’d argue it would work very well in every single RPG out there. When you take GM Moves, though, and put them in a game without attendant PC Moves, you’re screwing up. First, MC/GM Moves in Apocalypse World aren’t perfect, and many contemporary PbtA games have gone for much more specific outlines of what Moves to use and when (Fantasy World is a solid example of this). That said, in the context of Apocalypse World, where the GM never rolls dice and can do literally anything at any time, Moves are a godsend of both a pacing mechanic and plain old headcheck. In a D&D-like game, it is at best an overly prescriptive and at worst an actively confusing way to explain how the GM should be administering events in the game world.

Now, with that said, I understand the temptation. In a trad fantasy game you still have the narrative license to make a whole lot of things happen out of thin air, and it’s still useful for a GM to have some sort of headcheck and backstop for understanding when and how it’s narratively appropriate to throw something into the mix. I wouldn’t even say that the specific Moves in Daggerheart don’t work for this context, per se. The problem comes about when you’re mixing a mechanical lever to challenge players, Fear, and a mechanical resource, Stress, into a non-mechanical kind of ‘best judgment’ zone. And this kind of goes into the same issue I had with structuring the dice rolls like they’re PbtA rolls (besides the fact that Hope/Fear being an independent variable means that statistically they’re nothing like PbtA dice rolls): There is a strong desire for the designer to use these mechanics, but they haven’t really adapted them for the kind of game this is. Like the dice, I think this ultimately comes down to writing, which is why I called this all ‘Ctrl-C’ in the first place. The desire to use these ideas, to use GM Moves and teach reactive GMing, is a good one. That said, it needs to be actually written for Daggerheart, not just borrowed from an indie game.

---

The fact that most of my concerns about coherence are in the GMing section encapsulates my concerns with the game. This is another fantasy game, and I think any reader who goes in expecting something other than revisions to the D&D framework doesn’t understand the hobby very well. That said, what’s going to make this pop off, what’s going to make it a viable competitor to D&D, comes down to two things. The first is the ability to run this game and have the things you want to run feel good. That sits partly in pre-written content, yes, but it also comes from a game where the GMing guidelines make sense and the rules act the way you expect them to act when you change something. The second is the ability for any table to take stuff they see out in the wild, be that from Critical Role, another streamer, or hell, just on Reddit, and put it into their game. All those rules, the custom content rules, aren’t written yet, and while that makes sense for a playtest, it ends up being a fairly significant gap in assessing how this game will go over.

From the player side, I think this has a lot of the right ingredients. The character options are balanced away from core viability attributes like hit points and evasion, so there’s much more breadth to make the character you want without worrying about handicapping them in play. There are a ton of tangible character options in the form of the classes, ancestries, and communities, and it hasn’t escaped my notice that the anthro options are in there to broaden the appeal compared to D&D’s more hidebound worldbuilding. Even something like having players roll 2d12 but giving the GM a d20 is a recognition of how these games play at the table and that swingy can be fun but a bell curve is less frustrating.

In a way, all of my observations about the game so far, good and bad, build up to the game’s construction in response to Dungeons and Dragons. A lot of the good mechanical choices here are made because they address something D&D does poorly. A lot of the questionable mechanical choices are in areas where D&D is either profoundly bad or just absent, and the lack of stylistic direction ends up being a major problem.

I think the Daggerheart team has put some good bones together in ways that players will tangibly notice: The character options, the dice, the meta-currencies all work, and yes, I know a lot of them work because I’ve seen them before in other games. That’s not a sin, and trad game design at the systemic level isn’t so complex that the buffet approach won’t cut it. Besides, it’s a playtest. To the degree that any one part doesn’t work it’ll be in the execution, not the idea, and that can get changed.

I just hope that the design team revisits those PbtA-like concepts, both in the dice mechanics and the GMing mechanics. Not enough development was done in either case; the GM Moves kind of hang loosely next to the more developed mechanics, and the dice mechanics will, from writing as opposed to statistics, evoke severe frustration every time a player rolls a 22 and it ends up being With Fear anyway. The main thing the design team needs to do is stop thinking about if a mechanic has worked well before and start thinking about how it would work in a game that’s supposed to feel like D&D.

Because this game is supposed to feel like D&D; that’s basically what ‘high fantasy’ actually means in an RPG context anyway. The good news is that I think Daggerheart could really take off. There are some very smart choices made in this system. There are also some very short-sighted choices made, but (not at all) coincidentally they’re around game elements that D&D sucks at; at least the bar is very low. My personal hope, though, is that Daggerheart, success or not, does more than step over the low bar. The narrative rules brought into this game were brought in with intent but not redesigned or executed with that same level of intent. That is the one thing Daggerheart needs to truly stand on its own and compete. I only hope that the design team is willing to take my challenge to them on, to do more in their playtest revisions than nerf domain abilities, adjust numbers, and fix typos, and instead get this game to the level it needs to be to stand toe to toe with monopolists giants.

Playtest materials are available from the Daggerheart website and DriveThruRPG. Header art by Jessica Nguyen.

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! Finally, you can support us directly on Patreon, which lets us cover costs, pay our contributors, and save up for projects. Thanks for reading!