Citizen Sleeper 2 Wants You To Fail Your Way To A Better Story

Ahead of the release of Citizen Sleeper 2: Starward Vector, I spoke to developer Gareth Damian Martin about what it’s been like to come back to the world of the dice-driven RPG in a post-Baldur’s Gate 3 world and how this sequel ensures the action remains interesting even during failure. Citizen Sleeper 2 uses a very aggressive auto-save feature, preventing players from recovering from their mistakes or circumnavigating difficult rolls through save-scumming. As Damian Martin puts it, they’re “trying to be a game master for a million people,” and that means honoring player’s choices and incentivizing them to keep going even when it might feel like the deck is stacked against them.

In Citizen Sleeper 2, you once again play as a sleeper, sentient android that’s an emulation of a human mind that’s then housed in an artificial body. The story begins with your body on the fritz and a bounty placed on your head, and no immediate means to repair yourself or escape the asteroid belt-located settlement you find yourself on. You pick a class to curate your ideal playstyle and then set out, using dice rolls to determine the outcome of the choices you make and shape the relationships with the crew you begin to recruit. Each class now also has its own unique push ability, allowing you to potentially reroll the dice in certain circumstances to transform a failure into success.

Citizen Sleeper 2: Starward Vector launches on Xbox Series X|S, Game Pass, PS5, Switch, and PC on January 31.

GameSpot: To create the most narratively satisfying consequence, when is the best moment for a dice roll to happen? How did you determine what actions in Citizen Sleeper 2 should be a dice roll and which actions should just occur?

Gareth Damian Martin: From the very beginning, it’s always been about giving a constant feeling of choice and expression for the player in all of their individual small interactions rather than it all being loaded onto dialogue.

I try to think about the potential choices as much as possible, even if they’re only small choices within the dice system, because I think that’s the thing that I’m really excited to do. Because one of the reasons that I wanted to make a dice game was to allow myself to tell stories [and] I don’t have to make a combat system, I don’t have to make a work system, and I don’t have to make any system. I have one system, and that system could tell any story that I can imagine and that I can think of. So I spent a lot of my time trying to think about what’s an interesting story.

With Citizen Sleeper, the fun for me is that I get to sit down and I go, “Okay, what are the cool things that happen in episodes of ship and crew stories?” …Because I know that my dice system will make that mechanically interesting, especially in Citizen Sleeper 2. And so there are small, expressive choices like, “Am I the kind of person who hacks this door? Am I the kind of person who’s saying, ‘Chop down this door?'”

So I really wanted to highlight all of those little choices and how the contracts are a real opportunity to then make them culminate as part of a story. While in Citizen Sleeper 1, you had this longer, more languorous pacing the longer it goes on, in Citizen Sleeper 2, the contracts give this opportunity to actually have a series of dice decisions evolve and mutate and go wrong or go well and create all these little dramas. That’s what I think about when I’m thinking about dice decisions.

Why a six-sided die system? Is that just your favorite? Did testing reveal unfavorable results with a four-sided or eight-sided die?

I wanted to use the dice that was the most immediately understandable, and [for] most people, [that] is a six-sided dice–it has the most associations with chance and luck, [so] when you show someone a [d6] dice, they know that luck is going to be involved. They know that there’s a danger of things going bad. There’s a danger. It’s like when you see two 1s, I don’t know about you, but in any game I play, when I see two 1s, I have to say snake eyes. It’s just a thing that’s in me. And I think that’s the universal quality of dice, that they can be very clear mechanically, especially six-sided dice.

I sometimes forget that for the grand majority of the world, the six-sided d6 is all there is. There is no other die.

I’m sure [if] those people [saw] a d4, they’d lose their minds. “What part of the triangle am I looking at? There’s so many numbers on this right now, what’s happening?” And d20s are associated with D&D and they’ve become the brand image of D&D. And I think Baldur’s Gate 3 really [solidified] d20s’ [association with D&D] into people’s minds.

It’s so interesting to have [released] Citizen Sleeper 1, [then] Baldur’s Gate 3 [comes out, and now] there’s a Sleeper 2, and to notice the difference in the way you can talk about dice [with people who play video games]. One thing we started doing quite late in the [marketing] campaign [is] calling Citizen a dice-driven RPG. That’s now our line for describing the game. And I wouldn’t have done that if it wasn’t for Baldur’s Gate 3. And I felt like Baldur’s Gate 3 opened up this space for me to be able to say that because I would be worried that people wouldn’t [get it]. When Citizen Sleeper 1 came out, people would not understand what that meant.

What are some of the challenges in designing a single-player game so that it can still incentivize a player to continue through failure? There aren’t additional players or a Game Master to lessen the sting of a bad roll.

[From a developer’s perspective,] you wouldn’t believe how much harder it is to do something when you don’t have manual saves. It’s not just something that’s like, “Oh, I forgot to add manual saves,” or it’s a quality-of-life feature or something. It’s a genuine massive design factor [to] design around it. I don’t know about you, but there’s plenty of times I’ve played Baldur’s Gate 3–and this will be the last time I’ll talk about Baldur’s Gate 3; I’ll talk about my game–but where I just went into a fight and I was like, “Oh, I’m going to die.” And the only [auto] save I have is at the start of the fight.

If that was my game, that would be a game-breaking bug. That auto save would be the only thing you had. So I have to design around every possibility of failure to keep the player playing. That’s always been my focus, but that’s how I work as a [Game Master] as well. It is like I’m always trying to keep the players at the exact level of tension and freedom and constantly trying to adapt around that. And so the thing that I found really effective for achieving this, especially in Citizen Sleeper 2 where the game has multiple difficulty levels and it dares to be harder than Citizen Sleeper 1–which makes it even harder to have an auto save-only game–is that I found that pushing your luck as a concept is incredibly effective for letting a player balance themselves within a spectrum of failure and success.

[In Citizen Sleeper 2,] any job can fail and succeed on so many different terms. For example, you might succeed at a job but you might use five supplies on the job. Now, supplies cost 15 [each]. That’s a shitload of money that you’ve just thrown away. Could you have done that contract faster [and saved money]? Maybe. You might succeed at the contract, but there’s always, like in most contracts, an additional resource option. There might be, for example, in a falling-apart derelict, you can gather more scrap. But if you do gather more scrap, that derelict also becomes more unstable. So you cannot [gather in] that bit [anymore]. But also interestingly, you could ignore the main objective and just try to grind as much scrap out of that derelict before it collapses, then go back to base and try and sell that. And you might end up better off than if you did the job [safely or more slowly].

And then also because the stress system and the push system are tied together, when a player has a nice clear empty stress bar, they’ll be like, “Yeah, I’ll push this cycle. Why not?” But that push starts them on a spiral. The moment you get one stress, then you get more stress. And so then you start to get sucked into that and then you’re like, “Oh, I won’t push because I’m being cautious,” or you’d be like, “I will push because I don’t care. I’m going to risk it anyway.” And so you set your own difficulty level as you go as well. I always like offering opportunities for the player to get themselves in trouble. And I think that’s a big part of the design of Citizen Sleeper. And so a lot of the focus when I was working on it was thinking about what are all of these different ways that a player can succeed and fail? And for the game to keep playing, how can the game keep playing in almost every circumstance?

Now, this time we do have the dangerous difficulty where there is a permadeath. So that’s the only version of the game where the player can’t keep playing. And that is an honor mode-style game. And I would never recommend that for first-time players unless they really love that stuff. But that was me saying, “Hey, yeah, okay, I’m going to offer this opportunity to play this game on the highest intensity where you are really having to test yourself,” because I think mechanically it stands up to it. I don’t think Citizen Sleeper 1 would’ve benefited from a permadeath mode. I don’t think it would’ve done anything apart from end the game early and it wouldn’t have led to an interesting story. But I think in Citizen Sleeper 2, because there’s so much going on, so much more going on mechanically, and you have active upgrades rather than passive upgrades this time.

I think engaging with some more traditional RPG elements such as jobs [for] your party of people has allowed me to up the mechanical side of the game, but in a way that a game of Blades in the Dark feels, not like a way the game of D&D feels. And I think that means that the game has kept its really unique DNA, because there aren’t really any other Blades in the Dark-inspired video games.

No, I’ve looked.

But there are shitloads of Dungeons & Dragons-inspired video games. We drown in them. It’s an infinite number of games with health points and magic points and all the other stuff we know. Part of doing a sequel was just having the opportunity to actually pull in more mechanical stuff from tabletop games as well whilst still keeping that DNA.

How does Citizen Sleeper 2 react to the player’s luck or decision-making, if at all? Will the game offer a helping hand if a player happens to get extremely unlucky or misplay a ton of dice while acclimating to the game’s mechanics, for example?

When I’m making a dice game, luck is going to be a big factor and dice are going to have to be honest because I think that’s how I would be in a tabletop game as well. I think when you start fudging dice rolls, I think you’ve got to start asking yourself, “Why are you rolling dice at all?” like when you’re running a tabletop game.

[The game] will never fudge the rolls for people, but what I will do is create a world where failure is going to be interesting. It’s going to lead to interesting things in the story. And that’s something I tried to do early on in the game and communicate really clearly. [After] one of the first big contracts in the game, if you succeed, a big choice comes up about what you then do. And that choice can lead you to then choose to fail the contract on purpose. And if you fail the contract, the contract can go another way. So even when I’m offering the player success, I want them to think about what success means in the terms of this bit of the story.

Because at the end of the day, I don’t know about you, but the most interesting things that happen at the table tend to happen through a combination of successes and failures.

If it’s all successes, it gets boring. And if you stack it all in your favor, it loses life. It doesn’t feel weighty or meaningful. So for me, I’m always playing that game. And yeah, it’s very scary to make a game like that, but there are lots of design systems that I try to use to support the player in that. I wish every player who plays Citizen Sleeper 2 gets at least one moment where they go for it or they use a push in that key moment and they were one roll away from success and they did it. I want that to happen at least once for every player. So that’s what I’m focused on. Imbalancing, it’s not about making fair or unfair or getting people out of holes or putting them in holes. I just want to have those moments where it’s all resting on the juju, the prayers that you are making to the dice gods. I want that to come through.

This interview was edited for both brevity and readability.