Solo play, the Czege Principle and oracles

I’ve been thinking a bit about solo rpgs lately, and wanted to hear your viewpoints on a couple of things.

I’m a firm believer in the Czege Principle: “When one person is the author of both the character’s adversity and its resolution, play isn’t fun.” (Link to Lumpleygames discussion). So, to make for fun single player games, you need to externalise either the adversity or the resolution, right?

The most obvious answer is to use oracles in some form: random tables, drawing cards etc. that provide prompts for the player to shape the adversity around. Then, the resolution can lie directly in the hands of the player without it feeling unfun or unsurprising. The most prominent example of this that I can think of is Ironsworn, which leans heavily on oracle tables.

From my admittedly very limited experience, the sweet spot for oracle prompts is to have enough meat on them to not feel too wishy-washy yet not be so fully fleshed out that they feel like scripted events. Of course, what feels like a good prompt is super contextual.

I was wondering if people on here have experiences with playing or writing solo rpgs, making oracles and coming up with good prompts. Or some completely different ways of handling this!


When some people play board games solo, they try to play both sides. This is very easy in turn-based games where all moves are public - you can pretend to be Person A, do the actions you think will help you succeed, then switch over to Person B, where you deal with Person A’s actions and make your moves, etc., etc. This really works if the gameplay is rather mechanical and crunchy…which DitV actually tends to be (fluff tends to come out as a way to justify the mechanical results).

So I’ve done this approach as part of a solo DitV campaign using the vanilla setting, and am currently running a quick solo DitV session set in Star Wars Legends, using this same approach. All I need to do is find an interesting Town someone else had already made, think of how my character would pass judgment, and then get myself in a Conflict to see if I can impose said judgment. Complications however can arise as a result of my intervention, which I then need to deal with - usually via another Conflict, and so on and so forth until my character dies, gives up, or declare the problem “solved”.

The reason it doesn’t violate “Czege Principle” is that I don’t author the character’s adversity - someone else came up with the Town, I’m just trying to pass judgment on it.

I do author the character’s resolution, sorta (though the dice tends to inform how I author said resolution…and naturally suggest a new complication I might have to deal with).

…but even that could be fixed if I can come up with a basic list of strategies that NPCs can use against player characters (and then randomly roll on the strategy table to see how the NPC would act in game). The idea of authoring standard tactics that an NPC can use in-game also has been used in ‘solo board games’.


On the last part of what @igorhorst says, yes, there’s a way to use pre-written algorithms (similar to IAs in videogames), specifically for agents, but not necessarily. Randomness or sufficient enough complexity (such as linguistic / emotional interferences) makes you don’t have control of resolution.
Also the “you” at prep time is not necessarily the “you” at play time.


I’d like to stress the last part of @DeReel’s post, especially as it connects to @igorhorst’s point about boardgame solo play.

From time to time, I play a lot of boardgames solo. Sometimes it’s just to learn the rules, sometimes it’s to enjoy a game I like a lot but don’t get to play often enough, and sometimes it’s because otherwise I wouldn’t get to play it at all. The latter is especially common with wargames, and solo play is a big part of the hex-and-counters wargaming hobby.

However, what is very clear to me is that solo boardgaming is much easier and much, much more satisfying when it’s easy to identify the sides in the game as their own individuals rather than as players external to the game. For example, it’s much easier to do this with a WWII war game where you may have the Axis and the Allies as the different sides compared to a card game like Dominion where there is no character or faction for the player to identify with in the game. (I believe this is what is called the “agent” in board game theory, but trying to google that veers off into other directions so I haven’t been able to verify that right now.)

So to relate this to roleplaying games, I believe solo play there may benefit from identifying clear agents the player can switch their point of view between. My experience here is much more limited, but I’ve had more success with solo play in games where there is a lot of character vs. character conflict than character vs. environment and I think the fact that the former better enables the perspective shift between different agents (i.e., the different player characters) is one factor in this.

Can this be applied in a way that makes more character vs. environment oriented games easier to play solo? Perhaps it helps if you imagine the GM as a separate character? This isn’t something I’ve tried myself, but I imagine it could be easier to split your perspective so that the imaginary GM character is the one who is the author of the character’s adversity, while you as the player of the character try to overcome that adversity.

For example, let’s say you’re playing a solo character in a space opera game. When you come to a situation where you need to make a GM type decision, I imagine it would help if you have a clear picture of your GM character as being Bob, who loves big space battles, chase scenes, and over-the-top meolodrama, or Frances, who’s really into politics, ancient precursors, and mystic visions, or Lisa, who’s always looking for a chance to bring up cool robots, intrigue, and psychic powers.

Anyway, just a thought.


Good post.

This reminds me of how authors talk about being surprised by what their characters end up doing in the novel they are writing. Yeah, I think it can work, but then you need to have a pretty clear picture in your head about who these agents are, just as you say. Then I would suppose you’re more able to disclaim decision making because you can go with what the characters and the conflict dictate would happen.


Two more thoughts, one that I forgot to put into the previous post and one that sidn’t occur to me until after.

The first is that this imaginary GM should have more of a personality and not just a set of content preferences. Like, maybe Bob is the kind of GM who will drop you into a heap of trouble without much thought of the consequences but be fairly generous when you throw together an implausible plan, while Frances gives you plenty of warning of what dangers lie ahead but is merciless when you make a mistake, and Lisa will go after the things your characters care about more than the characters themselves.

The second is that you could use this to make your own personal GM oracle. I suppose this may be easier with a PbtA style game with clear GM moves, but on the other hand that principle is easily portable (even if figuring out the specific GM moves for a non-PbtA game make require some thought). So continuing the example above, Bob is always looking for an excuse to make a dramatic soft move, Frances likes to announce future badness and will use harsh hard moves to follow up on that.

Structuring this, you could make a set of random tables or just a list of priorities for this imaginary GM, perhaps as simple as putting the various GM moves in a particular order.


Ryuutama has that sort of GM character built into it.


Interesting! From what I could gather it’s not exactly the same, since the Ryuujin is an in-game character, but it’s in the same direction.

Curiously, I don’t think the Mythic GM emulator follows this. It is possibly the most successful modern single player rpg. What makes it work is than you play almost entirely as the protagonist…and it still provides surprises. Basically, as a player you ask the Oracle a yes/no question, then decide the odds on a d100 table.

The trick is the yes or no question can be something like, did my nemesis send his followers after me? Or, is there a hatch I can climb in to escape this hallway ? If a success./ fail resolution is needed you do the same thing (or use your fave system).

I think it works because as the player you can still be surprised AND when you are not, you get something you were expecting/wanting.


I confess, I have never touched the Mythic Emulator because honestly, it doesn’t sound like fun, but just from reading your description, how can it surprise you? You’re asking yes/no questions, based on things you are “looking for” or “expecting to happen”. So either you’re not surprised because you created the thing, or you’re not surprised because the situation hasn’t changed?

With Mythic GME, a lot of the fun comes from the unexpected Extreme Yesses and Extreme Nos, which you then need to dig further into when that flies against what seems obvious. Suddenly you have to come up with the non-obvious and ask about that. (The interrupted and altered scenes are similar in this regard.)

I mean, was it coincidence or was it enemy action?:smile:

It tends to get even further afield when the first question asked, about the most obvious cause for something happening/not happening ( phrasing matters), gets nixed.

This then tends to build and spiral outward over time.

MGME also tends to work because humans are pattern recognition monsters ( even when the pattern isn’t true, something that may not be much of a problem in this context), and will begin to identify and solidify patterns that make sense in context from these semi-binary results.


This is very true! So, I technically speaking most of the fun comes from the game system becoming more intrusively pushing the player to come up with something unlikely after a random number of likely responses are confirmed, or after an extreme Yes or extreme No is confirmed.

Oracles from Mythic GME variations can also help with this. So, while the Czege Principle can apply to Mythic GME, the main fun is pushing your imagination when the expected result is not confirmed as is.

I haven’t designed games specifically for solo play, but I have designed games that also can be played alone. If you want a traditional roleplaying game experience, then you’re thinking is on the right track (check out Pandemic), but if you want something else – like engagement in fiction – then you don’t need an adversary. What you need is a structure of play that shines a light on how you can give your ideas a twist.

The Murder of Mr. Crow is basically using a structure of how to create a murder mystery. You can solo play that game and come up with a playable murder mystery; I just turned that structure into a game, where the participant(s) come up with suspects and clues and prompts questions in a specific way to detail what really happened (improvised).

Imagine is take a routine from Impro, that I today use to describe an environment as a game master, to describe a situation. The structure of the game is taken from kishotenketsu, that takes what previously have been established and then gives it a twist. This is game that you need to play, not read, to be able to grok.


That’s a good point, that it can be about engagement to the fiction. And thanks for reminding me about The Murder of Mr. Crow. It’s relevant to something I’m working on. Another game that comes to mind is Gentleman Bandit by Allison Arth.

I still think you usually need some kind of prompts to nudge the player to interact with and engage with the fiction in a certain way, I guess, but I agree you don’t need to provide the answers necessarily. In fact, if you have the prompts be interesting and provocative enough, it might be better to not provide canned answers.

I’ll check out your other links later tonight, for sure. Tack!


Hi! I’ve designed quite a few solo RPGs myself and to be honest, I’m not sure if the Czege Principle applies to any of them! :smiley:

But this might be a case of different design goals at play. For me, the purpose of a solo RPG is less about having an enjoyable evening and more about a kind of internal alchemy, where the self comes into contact with the not-self in the form of the text and emerges different from the experience. In that regard, my goal as a designer is to create unusual circumstances which set the player off balance and then I ask them interesting questions and have them cogitate on the answers as the game progresses.

Now some of my solo games are a bit more mechanically intense, but in my mind, those mechanics serve the purpose not of driving the narrative but of driving the approach to the inner self. Mechanics create comfortable boundaries which can then be worked past through deeper, inner interrogation.

Not sure if that makes any sense, mind you, but I would argue that the world of solo RPGs is quite a wide one! Where Is Your Anger in particular has struck a chord with me lately with how it helps to create a greater understanding of the place and impact of righteous anger.


That’s a really good way of putting it, I feel. Or, as you say, there are lots of different ways to do solo RPGs and not having traditional adversity/conflict resolution be a thing is extremely valid.

I just finished God Damn Them All, which is a journaling game, that is to say it’s mostly a writing exercise where the game prompts and steers the player to write stuff they might otherwise not. I enjoyed playing with the ideas of how you incentivize certain actions when the player has almost total narrative control.


I’ve been (very slowly) prepping for some solo play with Legacy and I think I’ll try out having a “GM player” to refer to if I need it.


Ooh, that sounds very interesting for sure! I will definitely check it out!

1 Like