It’s time for another installment of Ask The Angry GM. A column in which I, The Angry GM, answer questions I have been asked. Hence the frigging title.

By the way, I know this column has become less regular than my cheese-and-beer-powered Wisconsinite gut, but that’s changing in the next two months in a big, exciting, fun, awesome way. Stay tuned.

If you want a chance to get a question answered, and you don’t need an urgent reply, email ask.angry@angry.games. Don’t forget to tell me what to call you, and remember, it’s a question, not an essay.

A asks…

Assuming system is not an issue, what are some things to keep in mind when running games for one or two players?

I hate answering questions like this because I’m not actually the asshole that people think I am. I’m a totally different asshole. I hate telling people their idea is terrible and shouldn’t be done. See, I want everyone to live their best gaming lives and to have the most gaming fun they possibly can, and I’m not afraid to scream at and swear at and insult you to get you there.

I’m also a very optimistic asshole. I don’t use the word, impossible lightly. I’ve actually written many articles out of pure spite because someone was stupid enough to tell me something was impossible. Or tell someone else in my earshot that something was impossible.

So it pains me to say that running roleplaying games for one or two players alone is a terrible idea and that you shouldn’t do it. And it’s not just for all the normal reasons most non-sexy gaming non-geniuses will tell you.

But let me cover the obvious crap first.

Now, I don’t know what you mean, A, when you said, “Assuming system isn’t an issue,” and I want to be complete in my answer, so let’s start with why the system is an issue.

Every tabletop roleplaying game has a proper group size. The good games made by actual frigging professionals know they have a proper size, and they usually clue you in, even if they’re not ballsy enough to be firm about it. D&D (2024) says right on the first page of the Player’s Handbook that the proper party size has three to five players. Pathfinder 2E is less explicit, but does let Game Masters know it was basically built for around four players. So, you know, it expects three to five PCs in the party.

Other games, especially smaller games, also have proper party sizes. They might not say so, and the designers might not even know they do, but they do. It’s just up to you to discover the proper group size and where everything falls apart.

Now, the proper party size is partly a mechanical thing, but there’s more to it than just game balance. Of course, when you’re playing and running tactical roleplaying games with strong encounter-balancing systems, game balance is a huge concern, but people don’t appreciate the full scope of how and why that is, and they don’t realize it goes beyond combat.

Of course, in any action-adventure fantasy roleplaying game with combat as a pillar of play, which is basically the only kind of fantasy roleplaying game worth playing, combat balance is a huge issue. It’s also a lot more complicated than anyone realizes, and the extent of the issue is rarely covered in the game’s materials.

Encounter-balancing systems are approximations that allow adventure writers and Game Masters to predict the likely outcomes of combat encounters based on a very complex set of assumptions. That’s important because, for gameplay reasons, combats should make the players work just enough for their victories while feeling just threatened enough. That’s called a flow state. Combats in that magic zone feel exciting, engaging, and fun. Outside that zone lie the Jagged Cliffs of Frustration and the Sucking Mire of Boredom. Those aren’t places you want to explore. Both lead to disengagement, and one leads to asshole players keying, KILER DM on your car. By the way, I know that was you, Carl. You’re the only dumbass who would spell killer with only one l.

But these systems are just approximations. They can’t account for everything, especially in complex, open-ended systems like roleplaying games, and there’s a lot of emergent weirdness once you start to deviate from the conditions under which the systems were tested. Because, remember, these systems weren’t developed by game designers in a lab wearing white coats and putting centrifuges in their autoclaves. These systems were tested at game tables of the type the designers assumed matched the way most people would want to play and run their games.

Consider, for example, hit points. When it comes down to the balance math, you kind of have to abstract the party down to a single pool of hit points. That’s okay because you can trust players to play smart. Reasonable players let the tank take most of the hits and keep the squishies safe. So if you assume a party is going to lose, say, a third of its total hit points, that doesn’t mean that a six-person party is walking away with two corpses.

But when the group size gets odd, those abstractions just don’t apply. Larger groups, all things being equal, can distribute the damage they take among their members more readily than smaller groups. Meanwhile, when a smaller group loses a member, it loses a much bigger chunk of its output. A two-person party loses a lot when one character goes down; a six-person party not so much.

As soon as you’re off the map in terms of what the designers assumed, it’s impossible to predict the outcomes of combat, which is bad even if it never leads to a party wipe. You still run the risk of falling off the Cliffs of Frustration or watching your horse drown in the Swamps of Boredom.

Note, too, that balance isn’t just about combat. In a well-made roleplaying game that’s based on challenging gameplay and isn’t just a storytelling performance engine for fanfic romantasy bullshit with a thin veneer of dice-rolling slathered on top, everything’s been balanced to some degree or another. Anything that costs resources, provides resources, or offers rewards has been tweaked to provide a proper group with the right amount of stuff. Anything from trap damage to treasure distribution to the number of downtime hours you get to go to shopping might be part of the balance.

But let’s move beyond mechanical balance. There are other issues here that affect the gameplay that aren’t just about numbers, math, and damage.

Consider, for example, how many roles, skills, and resources a party has. A four-person party might have access to between sixteen and twenty skills to help them solve the problems they encounter on the road of adventure. The same party might have four different strategic roles filled in combat and might be able to operate effectively at three different broad ranges. They might have a certain number of social skill options. Maybe they have a good balance of resources that refresh on three different schedules. You never think about this shit, but good designers do. At least, you hope they did in the game you’re running.

The thing is, this shit matters when it comes to open-ended play and creative problem-solving. The more access the players have to tools and resources, the more solutions they can find for a given problem. The more available solutions that exist, the more likely it is that the players will be able to come up with a solution when they need one.

See, very small groups are often resource-starved in a way few people notice. They’re starved for available options, and that can mean they’re also starved for kinds of options. They might have only two operating ranges instead of three or four, they might not be able to field enough strategic roles to manage a combat, or they might not have enough social approaches to handle a broad range of interactions.

There are ways to cover all this crap, all the balance issues, and the lack of operational diversity and available solutions. You can use hirelings, NPC allies, and pets. You can let players play multiple characters. You can give player-characters twice as many skills. These are solvable. Technically. But the one thing you can’t give the party is a third voice.

We can fight for days about what makes a roleplaying game what it is. I know. My supporter-only Discord community loves to argue about crap like that. I don’t want to have that fight here, though, so I’m just going to assert a fact. I can do that because I don’t ever say wrong things.

The interactions between players are a vital, defining, core part of the roleplaying gaming experience. You can’t do roleplaying without characters, right? But fictional characters are defined foremost by their interactions. The players do most of the interacting between themselves because they’re the ones who are always on camera together and because the game relies on their ability to unify around a common goal or purpose despite having different goals, desires, personalities, and beliefs. That’s all gold when it comes to character play.

Now, this is going to piss some people off, and I don’t really care: a duet game — a game run by a Game Master for a single player — is so fundamentally lacking in proper interplayer interaction that it isn’t really a roleplaying game experience anymore. No matter how much back-and-forth, in-character crap the Game Master and the one player get up to, it’s fundamentally different in kind from interplayer interaction that it’s no longer the same kind of gameplay experience.

I don’t care how long you’ve been doing duet games and how much fun you’ve had and how you’ve never heard complaints from your one player. You’re not doing what roleplaying Game Masters and groups of players do anymore. You’re doing something different. You’re allowed to do it; you’re not bad for liking it, but you ain’t doing the same hobby I’m giving advice about to someone who asked for it.

But that’s one player. What about two? And what about that third voice thing I mentioned?

This is all down to social dynamics. It’s down to how groups of people interact and make decisions, which are the two things players do together all the time in roleplaying games. They interact with each other, and they make group decisions.

When it comes to two-person groups making decisions, agreement is often trivial, and disagreement is crippling. If one person proposes an idea that the other person mostly agrees with, there’s usually no further discussion. There’s no tension. There’s no exploring of the idea. It’s often just, “Yeah, sure, that sounds good.” And even when there is some tension and some partial disagreement, it either defuses quickly or else it accelerates into full-on disagreement.

See, the ideal starting point for a group decision is partial agreement. That leads to friction, discussion, expansion, exploration, and debate. But two people rarely hit that stride because the social energy is only flowing along one path. Everything is just back and forth between the two parties. It’s a linear social dynamic.

A triangular social dynamic, one between three parties, changes the entire process. Obviously, the first big change is that you’re adding another perspective when you add a third voice, but you’re also increasing the odds of a sweet, sweet partial disagreement that invites expansion and exploration and creates social friction. But a third voice also creates a tie-breaker voice. That prevents disagreement from stalling the process. A third voice means that, in every exchange and every firm disagreement, there’s a spectator. There’s someone relegated to listening to both sides and who is available to interrupt the exchange.

There is also a diffusion-of-responsibility effect that kicks in when you add a third voice, and that can grease the social wheels in very important ways. You see, in a two-person exchange, everything is personal. One person owns the idea, and one person owns the agreement or disagreement. When you start considering personality types, especially those with regard to agreeableness and assertiveness, that personalization can cause a lot of problems. For example, less assertive people are less inclined to disagree, either out of politeness or because they don’t want to be the one stopping the game by saying, “No.” On the flip side, because every idea is owned by one person, anything that goes right or wrong becomes that person’s responsibility. If my idea screwed us, and you disagreed but relented, we might have an “I told you so” moment if you’re assertive, or we might have a silent, “I knew this would happen, and he never listens to me” moment if you’re not assertive.

When a third voice joins the group, every decision magically becomes a consensus. It ceases to be one person owning an idea that another person can veto if they’re willing to halt progress. Two-person decisions are authored while three-person decisions emerge.

But there’s more to social dynamics than just decision-making processes. For example, in two-person groups, whenever there’s any interplayer interaction at all, both players have to be on. No player ever gets to be in that partial state of semi-attentive rest while two other players burn through their social energy. There’s a subtle social fatigue that happens here. Especially because, for weird social reasons, the Game Master doesn’t always quite count as a social party.

A bigger issue, though, is the lack of a social sidechannel.

When you’ve got two players and a Game Master, any time there’s an exchange between just two parties, the uninvolved party has no one to react with, to engage in side chat with, or to crack an occasional joke with. Now, I know it’ll sound weird as hell for me to say a lack of a social sidechannel is a bad thing, but that’s because we only really notice it when it gets out of hand. The social sidechannel can get distracting, but when it’s moderated, it animates and elivens the social experience, it helps burn off tension, and it increases engagement. When two players are watching an exchange between the Game Master and a third player and making occasional comments, they are watching the exchange. If you have someone to react with, you’re more likely to pay attention. If you don’t, you start reading your character sheet or leafing through your rulebook or you pick up your phone and I have to smack it out of your hand and then I’m back at the auto body shop getting more scratches buffed out of the Angrymobile.

A third player adds a fourth social party to the table, and four people is the minimum you need for a social sidechannel to exist.

Honestly, it’s the social sidechannel that really makes the two-player problem intractable. In theory, you can overcome the whole linear social dynamic and two-person group-decision-making thing simply by making yourself a character and being both a player and a Game Master. In fact, if you do get stuck running a game for two, I recommend you do exactly that. I would never, ever run a game for two players without acting as a part-time player and a third member of the party. And, yes, I have done it. I’m not speaking out of my ass here. Of course, you’ve got to compartmentalize really well, and you’ve got to show some restraint, but you absolutely should make yourself a third-voice by joining the party as a player and pulling double-duty.

Even if you do that, though, understand that decision-making will end up feeling stilted and awkward, even if only slightly. It’ll be lacking something. Also, be alert for interplayer conflict due to personalization that either stalls the game or creates silent resentments. And finally, recognize that your game is going to feel a little bit socially flat. It’ll be a little quiet. Muted. Maybe even a little social fatiguing. So keep the sessions on the shorter side.

I really can’t in good faith recommend running games for just two people, but if you must, you can do it, and you can even have a pretty good time. That said…

Absolutely never under any circumstances run a two-person game for a romantic couple. Just don’t. Never.