Back in 2020 I reviewed the newest edition of John Harper and Sean Nittner’s Agon. Agon is a fascinating game, taking the characters on an Odyssey-like journey of myth through a number of islands. Like Greek myth, though, the game has a strict structure and, barring a small chance of premature retirement, usually ends in the same way. It’s great for generating stories, but not what I’m typically looking for.

Deathmatch Island is based on Agon’s mechanics, but casts the strict structure differently. The structure of each island is because the characters are contestants in a game show, a twisted game show where physical challenges and loot boxes give way to a literal battle to the death. Survivors make their way from one island to another until they reach the end game with Production, the shadowy administrators of the whole thing, shaping the game based on how many social media followers each contestant gets. The last surviving contestant may win a big prize…or wake up on yet another island with a job offer they never could have imagined.

What makes Deathmatch Island is not the translation of Agon into Squid Game, it’s the ever present ability and temptation to break the rules and try to find out what’s actually going on. There are restricted areas, secret maps, and forbidden gear all just waiting for characters to find. That all said…only one character makes it out alive, and they’re barred from competing again (…or are they?). Finding the full truth in a campaign of Deathmatch Island requires new characters, going to the same islands, to dig a bit deeper. It’s a nontraditional campaign structure, but one that holds a lot of promise if the players are up for it.

Mechanics

Deathmatch Island is built on the mechanics of Agon, and other than renaming everything it follows them pretty closely (there’s even a guide in the book directly comparing the two). There is a mechanical tweak regarding the Fatigue and Injuries mechanics; compared to the equivalent in Agon, characters earn more Fatigue (than they would Pathos in Agon) and have less to go around. This thematically makes sense; we’re stepping down from mythic heroes to schmucks in a game show. And in a game show to the death, the stakes don’t really hit right unless there’s a good chance your character’s going to die.

Overall, though, Deathmatch Island borrows Agon’s structure to a tee. The characters arrive on an island, undergo a number of trials, and then have a final battle. Instead of following a mythic archetype, though, you’re literally on a game show. All of the superstructure from Agon is transformed into an actual superstructure within the game world, helped by both Agon’s relative simplicity and some bulking up using ideas from other RPGs. The translation is direct, though. You still have a name die, but instead of it being your reputation amongst the Gods it’s your follower count on social media, influencing how Production shapes the show and your time on it. Instead of your Epithet, you have your Occupation from back home. Instead of Divine Favor, there’s extra gear that Production may hide away in different places on the island.

Let’s talk about these islands for a bit. The Trials portion of an island in Agon was left vague; the GM had the freedom to determine what conflict would lead up to the final battle and how to write that out into 2-4 different trials. Deathmatch Island takes a much more grounded approach that is easier to wrap your head around. Each island is designed as a point-crawl, and each node has one contest. The PCs are still aiming for 2-4 contests, but the intent here is for pacing. Once the GM thinks enough time has passed, the klaxon on the island sounds and it’s time for the Battle Royale, where the PCs have to best the other teams of contestants. The Battle Royale is divided into three: in Agon, you had the Clash, the Threat, and the Finale, while in Deathmatch Island you have the Scout, the Scramble, and the Battle Royale. The idea is ultimately similar: jockey for advantage, determine the stakes, and then resolve once and for all.

Contests, like in Agon, are relatively straightforward. The GM (also called the Production Player) will announce the contest and what Capability it tests, and then roll a dice pool for the opposition. The single highest die in this pool adds to the Danger Level (which is determined by either the island the PCs are on or a specific part of the island in certain cases) and the final sum is the difficulty level the PCs must beat. The PCs assemble their dice pool and take the two highest dice, plus the highest of any of their Acquisition dice (for extra gear they’re using). The primary differentiator for PC dice pools are going to be Occupation and Capability. Each player rolls a random Occupation in character creation, and that Occupation can provide a die for contests should it be relevant. The Occupations are also tied to a favored Capability; one of the five Capabilities that they’re better at (meaning they get a d8 for that Capability instead of a d6). The Capabilities are the five overarching skills of the game: Social Game (relatively self-explanatory), Snake Mode (deceit, intrigue, trickery), Challenge Beast (athleticism and physical ability), Deathmatch (live combat, coolness under fire), and [Redacted] (anything having to do with getting into restricted areas and behind the scenes). The PCs also get their name die, which is going to increase in size based on how many Followers they have; each contest gives characters some number of followers, but winning the contest obviously gets you more. Finally there is an advantage die; as the name would suggest this die is intended to model all the favorable or unfavorable environmental dynamics which give one side or the other an edge; if the advantage helps the PCs they get the die, while if it helps the opposition then they get the die. The dice pool for the opposition is a bit simpler; in addition to the advantage die the opposition will roll any dice that are defined by Traits that the primary opposition character has; those traits are the main mechanical elements of all NPC descriptors in the game. If the PCs win a contest, they win all the awards defined at that node in the island, which are typically different types of gear and in some cases special advancement.

If the PCs get through their contests and survive the Battle Royale, they’re off to the next island. In between islands they can heal some fatigue and do trust-building exercises (to earn Trust with the other PCs). There are also ‘jury votes’ where PCs award (very Survivor-like) superlatives to each other, and even theorycrafting, where the PCs are encouraged to hypothesize what’s actually going on behind the scenes at Deathmatch Island.

Meta Enough?

Theorycrafting is a great jumping-off point for talking about where Deathmatch Island comes alive. Now, this isn’t exactly intended to be like Paranoia, and in fact, things like theorycrafting place the intended dynamic squarely at the feet of the players from the beginning. During character creation each character is given an Initial Motivation, and there’s a 1-in-12 chance that one character starts the game with the Initial Motivation to sabotage the game and bring an end to it. The superstructure of Deathmatch Island isn’t necessarily meant to be gated behind spoiler tags, though the actual truth of what’s going on will vary from game to game and is mostly left in the hands of the GM.

Mostly. Theorycrafting is essentially a feedback loop where the players indicate what kind of conspiracy they think is happening (and hopefully what kind they’re interested in engaging in), and in turn giving the GM more ammo to drop hints and and nudge the nature of reality towards what the players are hypothesizing. It’s ‘Don’t Give the GM Ideas’ as a formalized mechanic. I think it’s neat, and I especially like that the GM retains some control over the game’s ‘truth’. Everyone knows what kind of game Deathmatch Island is, but Theorycrafting helps the players and the GM customize the exact nature of their dystopia.

I keep on going back and forth on the [Redacted] mechanics. From a pure mechanics perspective it’s very smart; engaging with what’s going on behind-the-scenes is different from playing the game as intended, and spending your time on nodes that give [Redacted] awards is a tough call, because those awards are harder to use in the standard contests (which may still very well kill you). On the other hand, the trad gamer perspective immediately bristles at it…wouldn’t Snake Mode and Deathmatch be equally relevant when trying to break the game? It’s a distinction that requires acceptance as opposed to more justification.

There’s also the pacing of the game. Deathmatch Island is intended (at first) to have the same length as a shorter arc of Agon, going through three islands and having somewhere between six and twenty contests (the contest numbers go up if your team gets access to a vehicle) and three Battles Royale, including the End Game. In the End Game, if things go according to how Production intends, there is one survivor and everyone else is dead. Of course, you don’t necessarily get through a Season (the arc of sessions that make up the three islands) and find yourself where Production intends. PCs are likely to get maps and intel from [Redacted] contests, and those may significantly change how each island is approached. When it comes time for the End Game, each player is handed a Prisoner’s Dilemma: Each player may choose to Play to Win, or Break the Game. If everyone chooses Break the Game, then the entire End Game goes very differently, and the PCs are working together to bring the game down. If even one person chooses Play to Win, though, the End Game continues as normal and everyone who chose Play to Win gets a massive advantage die. For multiple reasons, the odds of the team coming together to Break the Game in Season One are very low.

And this is where I’m less certain about the pacing. Each subsequent Season is played with completely new characters (with potential exceptions) who get all of the secret maps and unused [Redacted] gear from the previous Season. This means that even by Season 2 the PCs may approach the game completely differently. However, the game is supposed to take place on the exact same islands as before. There are enough nodes that you can run each island two or three times without overlap, so this is a place where taking some time to run the game at full length would help me understand if the repetition works. If I’m understanding the game correctly you would end up running two to three normal Seasons, and the contestants would align towards breaking the game faster each time. Depending on how those Seasons played out you would then engage the ‘New Game+’ mechanics. In ‘New Game+’ the characters are already aware of what’s going on, and are playing the game with the intent of breaking it and destroying Deathmatch Island from the beginning. I really like this arc, and I want the continuing wave of dawning comprehension to drive my players. I’m also aware that playing in the same areas with brand new characters each time is not the campaign structure most gamers are used to. To say the least, Deathmatch Island will require buy-in from your players, and not just in the realm of safety tools.

---

Deathmatch Island took me a few reads to really internalize the intent of the game. While the Agon rules are ported with minor modification the overarching game structure is completely different; while Agon is the story of heroes growing in skill and stature Deathmatch Island involves killing and retiring PCs at a steady clip, short-circuiting the normal advancement cycle as we understand it. Instead, the PCs keep their forbidden knowledge between ‘Seasons’, and the information about Production and what the game actually is gets closer and closer.

The starting islands and casts are very well done, and provide solid fodder for a campaign of Deathmatch Island. I expect that more islands and other pre-written material are likely in the works (there is already a Backerkit stretch goal with nine more islands); I also assume that an enterprising GM could write their own, though there isn’t much guidance towards that activity in the book. Overall though I’m not too concerned about this; running the game and understanding how the nodes play should provide enough of an instinct to write more material should that be your interest.

From reading Deathmatch Island the game has shot up my ‘to-run’ list, but one of the biggest reasons is so I can definitively find out if the game works as intended. At first I was thinking this may be a semi-adversarial game like Paranoia but it really isn’t; naming the GM the ‘Production Player’ is mildly disingenuous as the GM is still guiding the players towards their goals in a fairly impartial way. To be clear this is a much easier and safer way to structure the game, but I was kind of looking forward to things getting weirder than they did. Even without the high weirdness of adversarial asymmetric rules, Deathmatch Island still provides something very different by highly deemphasizing character advancement, relying on carryover information both in and out of character, and basically playing like a legacy version of an RPG in a lot of key ways. Like Agon, Deathmatch Island plays very differently than most RPGs and carries that difference very well. Unlike Agon, though, Deathmatch Island has teeth. Within a palette of limited mechanics, Deathmatch Island gives players life-or-death decisions, some of which they’re not even going to be aware of, and those decisions often have weight well after death and into the subsequent character’s lifetime. After reading both, Deathmatch Island has piqued my interest more than Agon. Both games are narrow; you roughly know how the arc of the game is going to go. When it comes to a dystopian game show to the death, though, I already know it’s more about the journey than the destination.

Deathmatch Island is coming soon; it’s available for pre-order from Evil Hat Productions. You can also check out a free preview at DriveThruRPG.

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!