Deep Dive: Cowboy Bebop

I have a theory: there are two kinds of licensed games. Or, perhaps, licensed games fall on a continuum.

The first kind is where the rules provide as little friction as possible while you Do License Things. The game expects you to bring lots of knowledge of the license to bear, and the system does the bare minimum (usually some flavor of task resolution). They rely on the facilitator and players to know what works in the license. Doesn’t mean they’re all traddy by any means! On one end of the continuum you’ve got The Terminator RPG and a very straightforward skill plus die against target thing, and on the other you’ve got Tales of Xadia and an iteration of Cortex Prime. But the point is that in both cases, the game works best when the players know and love the material and come to celebrate the property.

The second kind is where the rules make an editorial statement about the material. The system models and directs the entire creative flow. You can learn something more about the topic than you might have going in. This is tricky! And it means you, the player, don’t have as much say if you came to it expecting to run a certain kind of session. You’re not celebrating, you’re modeling. Extrapolating. 

I’m not typically a big enough fan of a property to want much of the first kind (although Free League’s Alien: the Roleplaying Game is as close as I’ve ever gotten), but it is interesting when you turn yourself over to the second kind and let the game teach you things.

The first time I went into a game without already being a fan of the material was Kotodama’s English translation of Tenra Bansho Zero. True story: I know next to nothing about manga or anime, I kind of hate the frenetic yelling. I feel like a lot of the “big” shows sound and look like this, and have no idea what the appeal of the media might be. But TBZ isn’t just a “here are stats for transforming robots, magicians and ninjas” game. Instead, there’s an entire flow to the play and it’s all character-centered. You struggle with little character statements, turn them over quickly into a different kind of economy, take on new statements, advance your character and so on. This is not a deep dive on TBZ so I won’t go into further detail here, but I will say I came away after playing it with a much better understanding of why the shows feel so frenetic and the soap-operatic appeal. It taught me things.

This is what I was hoping for when I picked up Mana Project Studio’s Cowboy Bebop

Teach Me!

I suspect the corgi is everyone’s favorite.

What I know about Cowboy Bebop, the show, is that it involves bounty hunters and was both an anime and a Netflix live-action adaptation. That’s it. I’ve seen that there’s corgi involved (I do love corgis), a sexy lady, stylish clothes, and spaceships. I watched about half of one of the Netflix episodes and couldn’t get into it. But I adored Mana’s generic engine Not The End, which continues to be my favorite go-to at cons when I have some scenario I want to play out and don’t feel like building out a full Fate or Cortex hack. I trusted the publishers and hoped their game would teach me things.

Here’s what I learned after playing three bounty hunting missions with the IGRC home crew.

The tl;dr if you’re not into our deep-dive format: a tightly structured storygame with strong opinions about how to be a Cowboy Bebop writer.


Cowboy Bebop the Roleplaying Game is a co-creation of Don’t Panic Games (a board game design shop) and Fumble (Italian podcasters and designers of that Not The End game I mentioned) and feels like something a board game company and a high-abstraction RPG designer would create.

Pure style from top to bottom.

The characters are all futuristic bounty hunters in a retro-futuristic Solar System. The setting material is straight out of the anime, covering all the planets, some moons, asteroids, lots of fun options. To the designers’ credit, there is a lot of excellent setting material in the book, as well as essays about the broader themes: the nature of cyberpunk, 70s crime drama, spaghetti westerns, the style, the retro-futurism. At no point did I feel beholden to any particular canon, and that’s an excellent message to convey. Our game went darker and stranger than I thought it would based only on my half and episode on Netflix and flicking through the art. 

The theme of the game is that the bounty hunters are burdened by their history, and in the course of hunting bounties, we learn more about those burdens even while they accumulate new regrets and weights. Through a series of formalized bounty hunting missions, personal episodes, and seasonal finales, we find out whether the bounty hunters overcome their pasts or not. 

All this happens through a music-themed filter, where every action and every goal is framed as one of five musical genres: dance (taking forceful action), rock (withstanding opposition), blues (getting deep in your feels), jazz (figuring things out) and tango (interacting with others).

How It Works

Characters are made up of traits that are linked to one of those five approaches I mentioned. There is no trait list, it’s wide open, but it’s also pretty shallow: you mostly describe how things look “on screen” so to speak. So your mirrorshades might be a Dance trait, if that’s what you drop over your eyes as you go in to kick someone’s ass. Or they might be a Blues trait if that’s all we can see of your face as you think about your past. These traits provide little bits of description you’ll hit over and over with at least a patina of theme attached to them. 

The trait grid. Occasionally traits touching each other in the grid matters!

In play these traits work well. Our bodyguard mook was constantly invoking his cybernetic arms and metal mohawk, for example. Unfortunately it makes the scenes a bit one-note, since folks mostly want to roll the most traits they can so will constantly go back to the approach where they have the most traits. The only time the GM (the “Big Shot,” which is a thing from the show I guess) can mandate an approach is to finish ticking down a clock.

Oh yes, the game has clocks similar to what we’ve seen in many Powered by the Apocalypse and virtually all Forged in the Dark games so far. Having goals and threats with hit points is a good thing! I’ve written about clocks extensively. But Cowboy Bebop handles clocks a little differently.

A New Kind of Clock

Every session is an Episode, and there are multiple kinds of Episodes with slightly different rules. But in all cases an Episode consists of three acts, called tabs in the game because everything has a music theme attached to it. And to get through a tab, you have to complete an Objective clock before a Threat clock gets completed first. In both cases, the GM assigns an approach to every clock. In the case of an Objective clock, this approach – a genre of music – is the sort that a PC must use to complete it. In the case of a Threat clock, the approach gives the GM additional rules to apply pressure. 

If you’re not into clocks — that is, pacing tools external to the fiction — this is not the game for you. I’ve found a surprisingly large cohort that doesn’t love a clock!

Formally, any player can create a clock, and every clock will either be an Objective or Threat clock. Unfortunately, there’s no guidance at all as to when or how to create new clocks. The only requirement is that each tab must have one of each. So in our game, given the fact that every Episode really should be one session, we simply didn’t add more clocks to play. 

In a FitD game, adding a clock is a great tool in the GM’s toolbox to either give a “big” effort – something bigger than you feel like a single Action test should take – more time and scope, or in the case of a “bad” clock to bleed off complications while telegraphing a future major change in the fiction. In both cases, though, there’s strong interplay between the fiction and the clocks. I think if Cowboy Bebop gone with this application – the GM can require a clock for complex efforts (as written), the fiction as well as hits can tick a clock, and Episodes could be more expansive and run across multiple sessions – it might have felt more satisfying. But that would have meant the sessions felt less like watching a tight, snappy episode of Cowboy Bebop.

In Cowboy Bebop, though, the only thing that matters is running down the Objective clock. And the only way to do that is to make rolls. There is no dialogue between fictional positioning and the clock, so a character’s in-fiction efforts don’t ever tick any clock. The practical outcome of this is that entire transaction is a bit inverted from what you might be expecting: after rolling, the player might have one or two hits and they can use that to tick down a clock, and then the GM will ex post facto fill in what ticking that clock down looks like. 

There’s also the (very rare) occasion where the players don’t gain any hits. Whether the GM gets shocks or not (which we’ll talk about a little later) is immaterial to this! Unfortunately, there’s no real discussion of what failing a roll might look like. If you’re coming from games that have taught you to always make failure mean something, you’ll be good to go. But it’s a lacuna as written. Given that the fiction doesn’t ever touch the clocks, straight misses will not be reflected in the Threat clock.

Another interesting wrinkle that seems notional at best is that any player can tick any clock. Players can elect to tick Threat clocks and the GM can elect to tick an Objective clock. But that feels like a nod toward a theoretical freedom most players won’t find interesting. 

In any case, the time pressure of trying to get through an Episode in a single 3-4 hour session, combined with the disconnect between the fiction and the clocks, meant there was really only room for one of each kind of clock in each tab. We ran one session where I experimented with multiple competing Objectives, and that was interesting. But mostly everyone fell in line with one Objective after a roll or two and chased that.

Bounties Are An Excuse

Besides the iron grip the game has over the three act cadence of Episodes, the strongest editorial statement of Cowboy Bebop is that bounty hunting is just an excuse to get the characters to reflect on their feelings. This is the absolute highlight of the game for me. 

A true joy of self employment is getting to cry on the job any time you want.

Characters track two economies related to their histories and emotions. One is their Bullets, of which you start with six. You spend Bullets to decrease the GM’s effectiveness after a roll. When you mark your last Bullet, the mission ends in failure and the character writes down the name of the session as a “weight,” that is, a heavy memory. The next session will be a Personal Episode about that character’s weights, and will probably end in marking it. Marked weights play into the endgame of a Cowboy Bebop campaign: the more you have, the better your finale. 

Failing missions also leads to characters writing down weights. When everyone has a weight, that triggers a mid-season Episode with, again, special rules related to the three tabs.

On that note: the three-act structure definitely imposes a “writer’s workshop” vibe on the game, not only because of the formal structure but because of the guidance for running each tab (act). In Cowboy Bebop, the first tab (Get Everybody and Their Stuff Together) is about character interactions and, as a practical matter, finding the Bounty. The rules say otherwise but when you look at the breakdown of the entire season of Cowboy Bebop they include as an example, many of the Episodes feature that first-act investigation. The second tab (3, 2, 1) is focused on discovering the Bounty’s secret: every Bounty has one, and it’s tied into the riff (special ability) that changes the game a little. The final tab (Let’s Jam!) is for the confrontation, where the players decide what they’re going to do with the secret they’ve just uncovered.

This is perhaps where Cowboy Bebop most firmly asserts its opinion about the show. In a conventional RPG, you might first learn secrets to help find the bounty and then have a confrontation. That’s a logical causal progression. But the RPG cares more about the emotional progression, and that means uncovering the secret after finding the Bounty but before you can even attempt to capture them. Makes perfect narrative sense and fits the show (as I understand it), but it’s not at all a conventional RPG in that regard.

The Writer’s Workshop

In actual play, there’s a productive tension between pursuing your Objective clock and getting deep into your character’s feels. This is mostly due to how the traits work, and how they keep everyone focused on a particular vibe. 

If you don’t choose “headphones” as a trait you’re doing it wrong.

Every Episode even comes with its own trait, Fate style, that anyone can tag if they can narrate it in. In one Episode our trait was obligations, and that was very effective at making folks think about their obligations and seek out opportunities to feel and act obligated. In another, the trait was guns guns guns. I’m sure you can imagine the difference there.

By constantly iterating and reiterating the music thematic elements of play – the clock approaches, the traits attached to approaches, the riffs that let you criss-cross these things – the game does a great job of acting like a writing coach. Find symbolism and hit it in different ways. Constrain your style to the type of scene you find yourself in. Put everyone on the same stylistic page. 

Occasionally character play felt a little one-note. Our retired cop would regularly fall back to his two Blues traits: his personal code of honor and a tattoo he had received from his (now dead) beat partner. So when it was the cop’s turn, it was usually expected he’d find a way to make those fit. If you’re even mildly skilled, you can always ensure you’re going to roll as many dice as possible. Same with our mad-genius decanted clone, our angry cyborg bodyguard, and our pretty-but-uptight young woman who inherited her captaincy. 

Math, Smoke, and Mirrors

The central rule is that players (and only players) roll a pool of dice against a target number. That number, and the starting pool, is connected to the current tab (act). In the first tab, you start with 1D6 and are trying to roll a 5 or better. Second tab is 2D6 and a 10. Third is 3D6 and a 15. Easy enough.

Mo dice, mo problems!

There are lots of ways to add dice to your pool: invoking a trait, matching the Episode’s approach, spending a Bullet, and so on. So you roll all your dice and are looking for, at most, two successes: one by meeting or beating the target number, and one more if you roll a pair or more of 6es. Meanwhile, the GM is adding up shocks, one per 1 rolled with a minimum of one shock every roll. Once the player has their hits and the GM has their shocks, the player can cancel the GM’s shocks by either wounding one of the traits they just rolled (which you can’t roll again until you fix it) or by spending a Bullet. But you can only cancel 2 shocks at most.

Players can apply their hits to clock ticks or to reducing the target number of the tab. The GM applies shocks in the opposite way, ticking Threat clocks or increasing the tab’s target number. A couple bad rolls, or some specific combinations of effects from Threat clocks or the Bounties themselves (they also get riffs!), and things can go badly for the bounty hunters very quickly. 

I mostly like the math behind the game. I like that outcomes get swingier the bigger the pools get, for example. At one point folks were rolling 6 or more dice and getting the occasion 4 or 5 snake eyes. Combined with effects that produce two shocks per snake eye, suddenly I’d find myself swimming in bad news and having to think long and hard about the actual pacing of the game.

Which brings me to my conclusion about the system. Much like FitD games, the GM’s available tools are mostly a lot of smoke and mirrors. Like after the catastrophic roll I just described, I could have simply ended our Episode an hour in and there’s nothing the players could do about it. But of course we’re there to play, so I looked at my options and decided on things like creating another Threat clock just so I’d have more places to put lots of shocks, and cranked the heck out of the difficulties. Which just made it even worse later for them. Fun?

At the end of the session I’m describing, I trivially engineered both the Objective and primary Threat clocks to pop at the same time. Honestly? That feels like an awful lot of control on the GM’s part. Since the players can see me applying my saved shocks (called Risk) in such a precise manner, a lot of the magic of the game I think disappeared. They knew I had my thumb on the scale the whole time. Deciding not to make them lose too quickly makes the victories feel unearned, I think.


Cowboy Bebop absolutely delivered on what I wanted it to: a distinctive kind of play with a lot to teach me about its subject matter. I might actually go back and watch the show at this point. I went in not knowing a thing, and my hardcore show fans both told me yes, absolutely, you’re nailing the vibe. That’s some damned good design that can do that. 

I love the way the setting materials are presented, a mix of canon and fan service and broader essays about the thematic and aesthetic through-lines at work. This is a great way to handle play material. Modiphius’ Dune does something similar, lots of high-level essays rather than a big setting dump, and it really works for me. Genuinely hope that as more IP-driven games appear, publishers will consider this approach. It takes a bit more work and maybe doesn’t directly tickle the ultrafans’ fancy, but it’s just so much more effective than a list of locations and characters and weapons.

The play required to make the game work – invoking traits that align with one of five musically-themed approaches – I feel is author stance all the way down. It was hard for the players to scratch their actor-stance itch because nobody is actually motivated to do much of anything, and every game is heavily structured around bounty hunting. The players are motivated to pursue their play aesthetically more than the characters are motivated to act on their needs. Like the show, as I understand it, things are meant to look fantastic and flashy and fashionable.

In the end, you will get a very Cowboy Bebop experience out of this game. If you’re already a fan you may learn a lot about what makes the show tick.

2 thoughts on “Deep Dive: Cowboy Bebop”

Leave a Reply