Backing Agon on Kickstarter was a no-brainer. The genre -- ancient Greek heroes -- is appealing to me and the group I usually play with. But also, it’s a John Harper game. Harper has a long pedigree before Blades in the Dark, but it was that game that really defined his place in the independent RPG scene. He’s a triple threat -- game design, layout, and illustrations -- and the initial materials for Agon showed the same level of quality and thought.
The final book did not disappoint. My first impression of Agon, besides the overall visual aesthetic, was that it was vastly different from Blades in the Dark. Where Blades has a lot of mechanical complexity, Agon appears streamlined. Where Blades has a robust fantasy setting, Agon relies on existing mythology. Blades’ familiar task resolution mechanic is nowhere to be found in Agon, which instead employs a unique vignette-style approach. At the same time, there are some elements in common between the games. Both use a dice pool and both incorporate a repeatable structure to the game play.
Like many games that introduce a new system, Agon is the foundation for a whole collection of games. Dubbed “the Paragon system,” the publisher opened the door for designing new “playsets” -- that is, games built using the same system. In any other setting you might call these skins. The underlying mechanics of a Paragon playset stay the same. Each playset documents its genre-specific equivalents to the elements of the Agon system. With few exceptions, the playsets -- in particular those released by Harper and his co-author Sean Nittner -- offer very light documentation, relying heavily on the Agon book.
For my two-month run on the Gauntlet, I ran a different playset every two weeks, scheduling four different flavors of Paragon over May and June. It was perhaps a little ambitious, but I was really interested in experiencing the flexibility of the system.
If you’re thinking about running a Paragon game, this article offers a few words of warning. It is, however, largely a celebration of Paragon. I had tremendous fun running this game in all its guises. One of its most distinctive and enjoyable features is the use of vignettes to structure the game. While vignette-style play may not be new or unique to Paragon, it is the system where I learned it first and fell in love with it. So, that is where we will start.
Every scene in a Paragon game is a contest. String five or so contests together and you have an “island” -- a single, self-contained story arc. In the Agon setting, your group of Greek mythic heroes embarks on an odyssey and find themselves dealing with challenges on a series of islands. Each island is more or less an episode (or string of episodes). Other playsets make use of this structure, too. Chamber (the X-Files-flavored Paragon) and Storm Furies (space battles) both use missions as the episodic unit of measure.
Agon’s book provides a set of 12 different islands, each episode summarized in three pages. The descriptive elements are short, and while there’s an outline of the key scenes and there’s some text offered for flavor, the onus is on the GM to improvise much of the setting. The key scenes include an arrival, a series of challenges, and then a final battle.
Each scene is resolved with a single die roll. That is, after the GM determines the difficulty level, the players each roll a pool of dice to see if they’re successful or not. The results drive how the scene resolves. And then we move on to the next scene.
Compared to most other tabletop roleplaying games, this scene-by-scene approach can feel a little weird. At the same time, this structure aligns with stories we see in media: tightly crafted scenes that come together into an episode. The connective tissue between scenes is hidden away, largely irrelevant or uninteresting to the viewers. Even though it’s unusual for a roleplaying game, it’s not entirely unfamiliar.
Because this structure is well understood, it isn’t difficult to make the transition. Because my style of facilitating games involves lots of improvisation, I love working with the players at the table to establish a scene and place their characters in it. We freeze time momentarily to determine the outcomes of the dice rolls, but then resume our narration of the scene. Depending on the number of players and how much you elaborate, an entire contest from beginning to end is about 20-30 minutes of game play. Because the table isn’t resolving every impactful micro-action via roll, the GM can maintain control of the game’s velocity.
In Surge Protectors, one of the playsets created by Nittner and Harper, the players play giant transforming robots in the vein of the 1980s cartoon. The genre is perfect for this structure of play, where the robots simply go from one fight to the next. At the same time, this playset suggests that each episode should focus on taking down one of the major “bad guy” robots, who are determined to wreak havoc on Earth. In Paragon, adversaries both large and small are defined by sets of dice.
Adversaries in the Paragon system -- any character or environmental challenge that the characters might engage in a contest -- are represented by two or more dice. To determine the difficult level of a contest, the GM rolls these dice, taking the highest value and adding the strife level. Strife level is 4, 5, or 6 -- an essentially arbitrary number defining how bad things are.
Agon suggests that the dice assigned to an adversary reflect different aspects of it. The king’s royal guard captain, for example, may have trained (d10) and loyal (d8). On the surface, this gives the GM a nice way of representing the scale of different threats. On the other hand, there isn’t any guidance for GMs on how to assign dice to these descriptors.
The Agon book offers numerous examples, but the assignments are at the discretion of the GM. Ultimately, the dice produce a result somewhere in the 9 to 15 range, which makes me wonder why the GM doesn’t just pick one of these values. Perhaps there’s something exciting about rolling for difficulty. Some mechanics allow the GM to add another die to their pool, representing some disadvantage faced by the characters, or some ill-tempered god.
In the play set that I designed -- Rising Tide, about ecoactivists seeking climate justice in a post-collapse world -- players have the option of letting the GM add a die to their pool. This die represents the character’s flaw or complication. In exchange, the character gets some bonus experience points. In play this worked really well, because it allowed players to activate their characters flaws in the fiction.
Earlier I mentioned that the Agon core book describes each island in 3 pages. The island and contest descriptions are evocative, but sparse, and depend on the table elaborating the setting and environment, the adversaries and NPCs, and the actions of the characters. The playsets for Paragon -- Chamber, Storm Furies, and Surge Protectors specifically -- offer even less guidance. Ultimately, the Paragon system tests the ability of the GM to draw fiction out of the players around the table.
Using just the text on the pages would make for short sessions and make the game seem very mechanical. Paragon depends a lot on improvisation and collaborative storytelling, perhaps even more than a run-of-the-mill Powered by the Apocalypse game. Even though PbtA games don’t have much in the way of setting or adversary descriptions, the moves provide leverage for improvisation. In PbtA games, however, the dice rolls resolve the narrative at the action level. In Paragon, the dice rolls resolve the action at the scene level.
Scene Resolution
The limited detail in the island descriptions and on the character sheets can make it difficult to get the improvisation engine going. Unfortunately the dice rolls don’t offer much help. Instead of rolls providing useful levers to provoke improvisation, the dice rolls simply indicate whether a character beats a difficulty score or not. I described the results of the dice pool as success vs. not success, but this isn’t quite right.
The Agon book describes rolling under the difficulty number as “suffering.” I’ve interpreted this to mean that the character “fails forward” or “succeeds with a cost” to use the parlance of collaborative storytelling games. Some contests have mechanical consequences, forcing characters to mark off Pathos (stress) or lose some other currency. The rules are not clear on how much discretion the GM has in doling out these penalties. Personally, I avoid them as the players are afforded minimal opportunities to affect their dice pool as it is. Taking these away from them limits what little control they do have.
Building the dice pool is really the main mechanical thing players need to do. The characters get certain dice based on three key elements of the character sheet. In Agon, these are the character’s Name, their Epithet, and the Domain. If the Epithet isn’t relevant to the character’s action, they do not include that die. The GM determines the Domain. With the first set of dice out of the way, the player has some choices to make, largely whether they are spending currency (in Agon these are divine favors or bonds with other characters) or marking tracks (usually Pathos) to add more dice.
As many times as I’ve walked players through this process, it always strikes me as a little disjointed. Even players who have been through the process several times struggle to remember it. It’s quirky. Some currencies you spend and some you mark off. One currency adds d4s and one adds other dice. The lack of parallel internal structures means determining the contents of the pool isn’t straightforward.
And yet despite all the machinations in the procedure, the player’s dice pool usually ends up with the same 4-6 dice. Early in the campaign these are mostly d6s with a d8 or two and a d4. Because of this, it can sometimes feel like the players’ choices -- really the only ones they can make mechanically -- don’t have a lot of bearing on the outcome.
The result of the d4 gets added to the total of the highest two dice. That d4 can make all the difference. Since you’re trying to beat a score of around 10-12 , it’s difficult to get to that on two dice, even if one of them is a d8. That d4 pushes you over the edge. In Agon it represents divine favor, and it fits nicely. That last little bit of success you owe to some higher power (or luck, if you prefer).
One aspect new players note right away is that there is a competitive element. In resolving a contest, the player who roles highest is considered the ultimate winner. They literally get all the Glory -- the points distributed at the end of contests. If the roll isn’t the highest but still exceeds the difficulty, they get half the Glory. Even if the character suffers, they get one Glory. There’s still some reward for emerging from the contest, even if you didn’t win.
Glory is one of two experience tracks. As characters complete contests, they earn Glory. As characters suffer Pathos, they also earn Fate. As they earn Glory, the characters start their dice pool with better dice. As they earn Fate, they grow ever closer to “retirement.” I like that these two concepts are disconnected from each other, but neither is particularly relevant for short games. Players need to earn 80 Glory before they can increase their name die to a d8. Since they only accumulate Glory at about 10-15 points per contest, it takes at least two islands before getting there.
Earning Fate happens a lot faster, and designated spaces on the Fate track trigger an advance. Since most one-shots won’t get players past one advance, I’ve started suggesting that characters take an advance at the start of the game and, if necessary, between missions.
The player who rolls best gets some other advantages in different kinds of contests. During the series of rolls constituting the final battle, the player who rolls best can set the stakes for the adversary or earn an advantage die for a later roll. The other main benefit in regular contests is that the player who rolls highest narrates last. The rationale for this should be clear: If we’re watching a show about Greek heroes, we save the most dramatic action for last.
This competitive aspect is discouraging, especially if your players are used to intensely collaborative fiction. In fact, I didn’t even include this part in the game I ran with my home group, knowing full well that the uneven distribution of Glory would make people dislike the game. On the other hand, I used it when running games on the Gauntlet and it was virtually a non-issue. I even ran Surge Protectors (the Transformers playset) with my kids’ group and they barely noticed. I mean, when you get to play giant transforming robots, everybody wins.
The ranked outcomes of the dice rolls gives the narration a structure, which is a welcome change from other roleplaying game systems. Consider most tables, where the GM is responsible for ensuring everyone has equal time in the spotlight. For me as a GM this is one of the most stressful parts of the job: I want everyone to have the same screen time. In the Paragon system, we are compelled to go around the table to describe the outcome of the scene. We even know what order to go in.
Characters Helping Characters
All this talk of competition might make you wonder whether characters should help each other. It’s a good time to point out that the characters aren’t really competing with each other: their goals are driven by the fiction and they generally have the same goals. (Some of my games involved some interesting tension between the characters, but they were largely supporting each other.) I suppose in the fiction the “winner” of the roll is the one whose actions are most notable, who people will be talking about the next day.
Because of this ambiguity, I’ve heard that some players aren’t sure whether to help each other or not. That is, the competition of the mechanics worms its way into the fiction. This makes sense: why bother helping someone if all your reward is minimal? The Paragon system has a mechanism for characters to help each other. A player can opt out of the contest and instead give their Domain die to another character in exchange for a point of Glory and a Bond. I dunno. I may be missing something but this doesn’t seem like a fair trade. Perhaps this strategy makes sense if you think you cannot beat the difficulty number with the resources you have available.
On the other hand, having such a mechanic in vignette-style play doesn’t make sense to me. Everyone is playing a role in the scene. A character’s role might be supportive, but they are still participating in the contest in their own way. Some of the most creative storytelling my groups have done involve the characters coming at the problem from different angles and relying on each other in different ways. In short, after trying the support mechanic twice I decided to instead encourage players to find creative ways to participate in the contest. Hanging back? Creating a distraction? Opening an opportunity for another? That’s still participating in the contest. Let’s roll to see how you do.
And this is the main impression I want to leave with you. Despite the quirky design choices, Paragon draws the table in. Everyone has a stake in every scene. Everyone has a role to play in dealing with the contest. From the first contest on a new island or new mission, or the building climax of the final battle, everyone has a chance to elaborate and embellish the narrative. Those mechanical quirks are some of the same things that make this game engaging and enjoyable: they draw the entire table in. More than anything else, this is my goal as a GM, to ensure that everyone is fully present even when they’re not speaking.
Some tips to help you GM Paragon, with references to the Agon rule book:
- Embrace the vignette- or scene-driven approach. Focus on completing about two or three scenes per three-hour session.
- And then embrace the episodic nature of the game. Give yourself ample time to do the three-contest final battle.
- Indulge in session zero discussion. Ask about character backgrounds. Delve into their bonds. Imagine their base of operations.
- Take your time setting up the scene (page 68). Ask players to contribute details.
- Don’t forget about Advantage dice (page 57) -- this is a mechanical way for you to reward contest success or just great roleplaying.
- Be patient in helping players assemble their dice pools (page 29). Take it step-by-step and try to do it the same way every time: first the basic dice (name, epithet, domain) and then the currencies to add dice.
- Let the final battle draw out (page 32). Use each of the three contests to build to a climax, starting with how the characters gain an advantage.
- Switch up the contest types (page 25), so that players aren’t relying on the same domain again and again.
- Consider dropping the competitive aspect, such that every player beating the difficulty earns the full Glory and every player missing gets half. It’s better to play than to turn people off because of one rule.
Dan is a regular player and GM on the Gauntlet Gaming Community. By day he runs a boutique UX design firm and by all his other free time, an amateur game designer. Follow him on Twitter @brownorama. Follow his game design work at cosmicbeagle.itch.io.
Thanks to Alun for the nudge to write this.
Thanks to Matthew, Darin, and Joe for slogging through two months of Paragon with me.