Running an RPG Solo as a Player


This aims to be a document with the purpose of providing some guidance on several topics with regards to how to run an RPG, Solo and GM-less. The scope of this document is focused on the player-driven approach.

The information provided is collected through own experience, discussions in social media platforms and inspiration provided by existing authored works.

Some text may be copy-pasted from own previous articles and provided as-is or edited as required.

I intend to have this document, live, and make changes as required. Excuse the editing. I think the content is more important at this time.

Hope you find it helpful.


So, what is Solo Roleplay Gaming?

I’ll start by answering the second part first. What is Roleplaying? There are so many people who know nowadays, so the relevant introductory text in RPG books is oftentimes omitted. So roleplaying means to undertake a fictional role. Usually it means in a fictional world as well, but not always. Let’s put the third aspect now, gaming. Since this is a game, roleplaying is covered by certain rules that need to be followed, and there is a line between fiction and reality, which is blurred through immersion, but still remains there. Traditionally, roleplaying games, have Players who play one character each (PC) and Game Masters, also known as Narrators or Storytellers who are tasked with delivering the experience and having fun in the process. The GM narrates everything but the PC intentions and actions, describes the world, builds up the story, runs the NPCs and also has referee duties. Some RPGs may split some of the GM duties across the players or even be completely GM-less. Now, let’s kick in the first aspect, let’s make it Solo. This means it’s definitely GM-less. The (one) player is tasked with everything. Running the entire game on their own with the objective of having fun in the process.

If I could categorize the different solo styles that appeal to this community, I’d put them in the spectrum of two axes. Journaling vs Traditional and Player vs Story Driven.

Journaling style: when the game mechanics are used as an inspirational prompt for narrative. The player has the narrative freedom to write pretty much anything they want, so long as it fits the concept of the prompt. The interpretation of this concept can be very fluid as well. At the end of the spectrum it treads very closely towards ‘Writing with dice’.

Traditional style: Playing an RPG of choice with the use of a set of tools commonly known as a solo engine. The solo engine can be as light as a single oracle, or it could be coupled with random tables, generators and whatever else needed to provide prompts. The RPG system could be united with the engine (solo rpg system) or could be a standalone regular TTRPG. At the end of the spectrum it treads very closely towards being ‘a solitaire board game’.

Player driven style: When the game revolves around the player’s actions. What are their objectives? What do they do to accomplish them? Similar to how a first person video game would be run. What do I see? What do I hear? What does my opponent do when I confront them? At the end of the spectrum it leans towards minimal metagaming. The player’s knowledge and the protagonist’s knowledge should be as close as possible.

Story driven style: The game is set up in scenes and threads and the game revolves around building a story using the protagonist. There is a lot of metagaming knowledge involved which is used to create an interesting story arch. At the end of the spectrum, it leans towards world building and emulating the characters instead of the world. Asking the Oracle what does the protagonist do when faced with the odds designed by the player.It’s not black and white, and many solo styles fall somewhere in between.

I haven’t touched all the styles, but I consider that my favorite styles are at the end of Player Driven, Traditional style.

Player Driven

In Player Driven style, there is no need to try to generate an adventure. No random events to get everything going. No seeds from an online generator. No cards with descriptors to point somewhere.
No, instead, they avoid completely the GM side and are focused on the Player side. The player has an objective, and actively tries to make it happen. There it is.

In my failed solo attempts, I tried so many times to view everything from a GM perspective and used a multitude of tools to get creative juices flowing, and it was okay until that point, but when I tried to have my player act through it, it just wouldn’t work. I hit mental blocks. It’s as if my Player was waiting for the GM to drive the action.

Almost all of the social TTRPG GM material is focused on how to make adventures and worlds and how to have the players run through them. Some GMs railroad the players into the story, others don’t need to as the players get the necessary signs and play through the story. In the solo community, traditionally we use those same materials to generate solo adventures, and here’s the catch. When you’re the same person running those sides, you end up testing how will your protagonist react to what you throw at them as a GM. It’s like puppeteering! The protagonist ends up being an empty shell, even if they have motivations, objectives and emotions. Because you ask, what will they do if the story goes X way?.

Protagonists in RPGs aren’t meant to be puppets. They are meant to act, not react. We play these games to be knights vying for glory! wizards trying to find immortality! rebels trying to overthrow galactic empires! These are not everyday people waiting to see what fate has in store for them, they make their own fates! Sure sometimes fate will strike back, but when they defeat the adversities, they get back on their task and find a way to do it.
The knight heads off on her own to find the orc chieftain and challenge them to combat, the wizard goes to the library to find forbidden texts on necromancy, the rebels spread pamphlets to call the workers on strike!

So instead of waiting for the Gamemaster to be a Puppetmaster and be the driver, be a Player and go do what they have to. Have the oracle react to your protagonist, not the other way around.

Protagonist Goals

In Player driven style, it’s important to have protagonist goals that can drive the adventure and are interesting to the player. Both long term or short term goals work, depending on what is the expectation from the player. Overall a short term goal will drive a side quest or a short adventure, like a one-shot, whereas a long term goal has the opportunity of driving adventures and sometimes even entire campaigns.

How to decide such a goal? Sometimes it’s clear as daylight to the player, but not always.

An approach that works is playing stereotype characters, with stereotypical motivations. It helps a lot when what the character wants to do, aligns with who the character is.

Other approaches include random or semi-random motivation generation:

Using an NPC motivation table can help provide some inspiration. Especially considering that NPCs generators usually weigh more on the “gray” scale of good vs evil and law vs chaos, this can give some interesting results. It may help if one takes the best-out-of-three result or something similar, defined beforehand.

A semi-random approach is utilizing a backstory generator. These generators usually allow the player to create backstories, by using some random input upon which the player creates a much expanded output. This creative freedom can help leash any wild results that would instead be generated by a random motivation table. Using the backstory as a starting point, a goal can be more clearly set.

Session Zero

A session zero is a concept well known to social TTRPGs. During session zero, the GM and the Players lay the foundation of their expectations for the upcoming game, as well as defining some limits.
It’s not much different in a solo RPG. Before diving in head first into a game, before character genaration, spend some time to define what you want from the game you’ll play.

What theme is it going to be, which ruleset you’re going to use, which solo engines, oracles and tools. How long will this adventure be, will it use a pre-written module? What character concept have you envisioned. Of course, nothing is set in stone, and as a solo player, you have the advantage of being able to change anything of the above mid-term, without having to answer to anyone but yourself.

Session Zero will save you valuable time going forward and reduce wading into uncharted waters. It provides awareness, so that if the game takes you away from your expectations, you can drive it back, if need be.

A question of balance

Often the question arises as to how many characters should the player run. There is no right or wrong answer, but it comes down to the Session Zero described above.

If the ruleset of choice is pulp, then a single hero, may be expected to wade through waves of foes without problem. In that case, a single hero, or at most aided by sidekick, are enough, and the game can be played as is.

Otherwise, the answer is: don’t do what your protagonist wouldn’t do. If playing a gritty, grim ruleset with a single protagonist, dont rush in blade first to a band of orcs. Choose your fights, or they’ll be short.

If you will be running an entire party so that game balance is proper, then make sure that you haven’t chosen a very crunchy ruleset that will bring down the game to a halt.

To crunch or not to crunch

Some swear by light rulesets for solo RPGs, and some will even claim that heavy rulesets are not soloable.

I’ll say this here once, crunch-heavy rulesets are soloable.

Actually I found out that they can be even easier than light rulesets, since they do the narrative heavy lifting through the use of their mechanics, in exchange for more dice rolls and detailed bookkeeping. It all comes down once more to session zero, and what your expectations are from the game.

Don’t choose a crunch heavy game if you don’t want to count arrows, armour weight, fatigue, or if you expect to finish a combat within 5 real life minutes. Choose it if you’d rather not try to come up with five new ways to describe X hit points worth of damage, and instead have it all described by the ruleset itself.


In sort, don’t. Solo Roleplaying is very personal. So I can only speak as to what brings me joy. If I start cheating on my own dice rolls then the whole immersion structure starts to crumble. I’m no longer playing, I’m writing a story to my own whim. If I don’t follow the rules I set for myself, then it starts becoming pointless.

I might choose different rulesets that have meta currency to avoid such fates (hero points, bennies, luck points or whatever they’re called) or a fail forward mechanic and since this is defined beforehand, it doesn’t count nor feel like cheating.

Getting started

The beginning of session one, can freeze a solo player like a deer in headlights.
There’s several approaches all of which are valid.

Use a stereotypical theme (in a tavern/inn/cantina). That’s an absolutely great place to begin the adventure and ask questions or get supplies and manpower to achieve the protagonists goal.

Or start in the middle of the action; The rebel is in a shootout with some local imperial law enforcement, and must escape in time.
The knight is climbing a steep hill that will lead him to an old nest of the monster he’s after, to get some clues.
The wizard has entered a decaying library in a ruined city of sorcerer-kings.

These are not definitive, just an easy way to start, and of course it could be anything else that has inspired the player or is described in a prewritten adventure module.
In general, try to keep a simple concept. The Oracle and random tables/generators have a habit of throwing curveballs to the story and producing extreme results. Having everything else toned down and rationalised brings a balance.


Secret doors, trapped chests, cursed magic items, enemies patrolling around the corner, betrayal from a close friend.All these -and even more, when presented in a game by the GM provide a surprise element which can be exhilarating for the players. They are crucial for certain game types such as horror and mystery.All is good then for a table full of players and a good GM, but what happens when you play alone?Metagaming cannot be avoided at all cases, but it can be minimized or controlled.
Sometimes you get extra output from the oracle or generator that your character wouldn’t know normally.
There’s several approaches to this depending on what the meta knowledge is and how you want to handle it.

  • Throw the meta knowledge into the bin. Discard it completely and consider it irrelevant. Solves the issue, but not so efficient, as you destroy potentially helpful output that you might need to generate anew in the future).
  • Put it in the subconscious parking lot – to be verified until it’s needed. On a social TTRPG analogy this is similar to the GM pondering on something to be presented later. Until then, it’s like Schrödinger’s cat. It exists, buts it’s nor dead nor alive. Our information has been given, but it’s nor true nor false. The moment we look up into the box, is the moment we ask the oracle for verification. Very resource heavy as you keep information on standby. Similar to flooding your Computers RAM instead of writing the data in the HDD. Could be written down on an index card.
  • Turn it into character belief. The meta knowledge is character knowledge as to what the character believes is true. Call it hunch, cognitive experience, collective unconscious or whatever, but for some reason, the protagonist believes it. Therefore they will act with this prejudice, until it’s verified or not. This could be embarrassing or a lost opportunity if she’s wrong, but it will make the story even more interesting, give a behavioural approach to the protagonist and also give an extra reason for this additional information to exist instead of the player trying to compartmentalise the knowledge away. This is similar on a social TTRPG analogy, to a player having their own strong opinion about something, but the GM hasn’t confirmed it yet.
  • Derive information from this meta knowledge, in a mannerism/inspiration approach. Since derived output can be generated by more than one possible input, it’s a way to use this extra information while providing interesting feedback for the story to progress.
  • Reveal as soon as possible. Thus meta knowledge is true, and becomes character knowledge before becoming a burden.

Rules and Rulings

Rules and Rulings is a common conflict in social TTRPGs, but how does it relate in solo Roleplaying?

In general you can consider that the Rulings are the Oracle answers. In a Traditional Player-driven solo RPG approach, the order of magnitude is as follows.

Ruleset > Oracle > Player

What this means, is that if there’s a rule in the RPG rule book that covers the situation at hand, go with this instead of asking the Oracle. Then proceed with any Oracle questions. Hand-rule it yourself only if nothing else makes sense.
If the ruleset doesn’t cover it, jump ahead to asking the Oracle.
For example, if there is a perception skill, use it to find out if your character can see anything. If there are only classes and levels, ask The Oracle.

Also many times these go hand in hand, as you need both to glean the answer, and you need to use them in a sequential order.
For example, once you have established that your character could have seen something because they are perceptive (successful roll), then ask the Oracle. The fact that your elven scout is eagle-eyed doesn’t mean there is something to see. But if your elf failed to see anything, then you miss your chance to ask the Oracle.

A guideline to the switch between ruleset and oracle resolutions according to the event type, is described below.

Action Events
These are events driven by the player. In those, you roll first the game mechanic. If there is a success then you also ask the Oracle. The actions must be defined specifically. If you search for traps, you can’t ask the Oracle if there is hidden treasure.

  • The dwarf scout searches for secret doors. Roll Success!, Q: Is there a secret door?
  • The halfling thief pickpockets the merchant. Roll Success!, Q: What does the merchant carry?
  • The elf wizard casts a premonition spell. Roll Failure!, Can’t ask the Oracle.

Reaction Events
These are events that are triggered as a reaction to the player. If there is a chance for something to happen you ask the Oracle.

  • The warrior opens the chest hidden below the goblin throne. Q: Is it trapped? A: Yes!, Roll Perception to notice it or Dexterity to avoid it, whichever is higher.

Phrasing Questions

In order to minimise metagaming, the questions must be phrased in a way that only knowledge accessible to the protagonist can be delivered.
What they can see, feel, smell or listen. Not what is, thinks or happens elsewhere (time/space).

In a weird noir horror crime mystery, one may go so far as to ask what a person says.
For example, instead of asking:
“Did the neighbour of the victim see anything interesting last night?”
“Does the neighbour say if they saw anything interesting last night?”

This allows the player and protagonist to share the same truths, and allow for room for lies and secrets to exist, unbeknownst to the player until the time they’re revealed.

Interpreting Answers

The Oracle should answer the bare minimum. Imagine real Oracles. Pythia was asked How can the Greeks defeat the Persians? and her answer was With wooden walls!.Of course you don’t have to be cryptic, but don’t turn the answer into You will defeat the Persians in naval battle in Salamis straits.

One of the most common oracle styles to closed questions is the Yes..No, But..And approach. I like the Locked Door question to give an example of how this works.

Q: Is the door locked?

  • Yes, and, it’s barred
  • Yes, it’s locked
  • Yes, but the lock is rusty
  • No, but it’s stuck
  • No, it’s unlocked
  • No, and it’s open ajar

Sometimes it’s tough to interpret a But..And modifier. If you find yourself spending too much time over it, disregard the modifier and go with a simple Yes..No.

Answers to Open questions are sometimes the toughest to interpret. Depending on the tool used and the result, it can either fit perfectly or put the player in a slump. Sometimes it helps to ask a closed question to define it better.

Randomizers and incorrect assumptions

Other engines may implement random factors to make the story more chaotic and take the wheel off the player’s hands. They may be called with any number of terms, but their use is the same. Whenever you ask a question to the oracle, there’s a chance that something will happen to spin the story.

Another less often used approach is for the oracle to have a chance that an assumption set in the question phrased is incorrect.

In the above door example a wrong assumption result would mean that the answer is that there is no door there at all.
This is an excellent way to take GM command away from the player, but at times it can be the most difficult to interpret, especially if there are several contradictory truths established beforehand. In this case even a re-roll may be necessary.


Awarenes of what hat one is wearing, while playing, is very useful to the solo player.
If the question is “what happens next”, then the solo player is wearing the GM hat, and needs to be extra careful.
If the question is “what do I do next”, then the solo player is wearing the player hat. In this case they need to be careful not to do something stupid that might lead to the protagonist’s demise.

Secret Oracles & Clocks

One particular issue related to metagaming is ongoing processes that lead to an event. In case the protagonist is privy to the process and the event, then there’s no problem, but what if they’re secret?

The dragon has swooped down in the village and the elders offered him a maiden as tribute. A young barbarian slayer has come to the rescue, delving into the dungeon to save the fair lady.
Dark cultists have gathered in the ancient cavern. The stars are right and the ritual has begun to summon the thing that should not be into the mortal world. Two detectives and a medium have deciphered the ancient texts and are rushing to stop them.
The bank robbery went awry and it has turned into a hostage situation. The police have surrounded the bank but have no eyes inside. The breach from the special forces is about to begin.
The chaos space troopers have broken into the nuclear fission reactor of the metropolis and are setting it up to meltdown while at the same time they prepare the scientists for ritualistic sacrifice to their deities. Orbital drop from imperial space troopers is imminent.

What will the protagonists encounter once they reach the final scene?

Will the barbarian have slaughtered scores of enemies to find the dragon’s belly full or will he find a maiden ready to fall in his arms?
Will the detective reach the grand chamber to encounter an ancient horror ending all life on earth, or will they find the cultists still chanting?
Will the special forces breach the bank to find the treasury wide open, hostages dead or will they find the robbers still considering their options, surprised from the assault?
Will the imperial space troopers save the scientists and lose everything in a huge meltdown or will they save the city, but find them mutilated? Will they split their forces and risk it all?

All these questions are solved either arbitrarily, by the GM, or in solo RPGs by asking the Oracle, or through another tool, which is the Clocks.

A clock can be fixed, e.g. 10 in-game minutes, perchance, e.g. roll 19+ on 1d20, variable, e.g. roll 20+ on 1d20+number of turns, and/or modifiable based on certain events.

In all cases it suffers from a serious metagaming issue. The moment the clock is triggered, the player knows it and has no motivation to keep going on.
That isn’t a problem when the clock is fixed in-story as well, e.g. the space pirates will ‘space’ one prisoner every 5 minutes unless they are given the set amount of ransom. The player knows and the character knows.
What about cases such as the examples above? Let’s say the barbarian is in the second dungeon room and the event is triggered. He won’t get his reward if the maiden is killed. Why go on? revenge only. Suddenly an interesting story has become a chore.
That’s the issue to tackle here, with the use of Secret Clocks.

A Secret Clock & Oracle is one that is rolled, but you don’t know the result until such time as when you reveal the event. What is important is that a timetick (a round, turn, minute, week or whatever is suitable) is predefined, and each timetick that passes triggers a question to the oracle.

There are three approaches:

  • Use a card-driven oracle, where the cards are drawn face down, at every timetick
  • Roll the oracle dice under an opaque cup or into a box, every timetick
  • Roll the oracle dice at the revelation of the event, once for each timetick

A fair amount of balance is needed to not make it overly difficult, especially when rolling dice. The cards-drawn approach as it keeps things balanced, and there’s this tangible feeling of the hidden answer in front of you. The hidden dice approach works as well, but depending on the case, one might need a fair amount of dice. The roll when revealed approach misses the tangible feeling of the hidden answer.


Chekhov’s gun (Chekhov’s rifle, Russian: Чеховское ружьё) is a dramatic principle that states that every element in a story must be necessary, and irrelevant elements should be removed. Elements should not appear to make “false promises” by never coming into play.

But still, we’re playing a game, so there is need of some ‘false promises’ to keep the suspense and possibility of failure in play.
That’s where ‘Hooks’ come in. Places where the story threads can hook into each other to create the overarching plot. But since they’re hooks, instead of knots, they may fail to grab onto each other and become dead ends.
Until the time that the Oracle makes the Hooks grab onto the Threads, it’s all fluid.

How does it work? It mainly has to do with inactive, but not dead, story threads.
Say the protagonist fought a thug and let them live. Perhaps this thug can appear again in the future and may even evolve to be a nemesis. But they will never be, unless the Player puts them back again in the story.
But to keep this from going against the concept of Player driven play, what needs to happen is that at the next time when the thug could be reintroduced, the Oracle is asked.

If the Oracle verifies the existence of the thug, then the Hook, has caught to the thread, if not, it flows away until there’s an occasion when they can be reintroduced.

Prewritten adventurer modules

The main approach of playing a prewritten adventure module solo, is to read it first. Enjoy it, leave nothing hidden, and then let the oracle take over.

Each time that you have a new scene, encounter or location, ask the oracle if it is modified. Depending on the answer, alter it accordingly, using the tools in your disposal.

This approach allows you both to enjoy reading the work, immersing yourself in the story, but still, allowing yourself be surprised.

As a side note, this method can be used to almost everything you consider “fixed” in your world. A myth, a historical event, a typical tavern, a town guard. Using this, you can spice up a relatively bland day in your adventurer’s life, but without losing the player perspective.

Zoom in/out, Speed up/down

This is one of the points where you have to put on your GM hat and ask yourself how much detail you want.

You can take command of the game and zoom out or speed up, and then ask a simple question “does anything interesting happen?”
If it’s a positive answer, then drill down on the details, otherwise, skip through the boring stuff.

This can be combined with skill rolls of the ruleset as appropriate.

For example if nothing interesting happens, then you can make a simple Investigation skill roll to see if your character learns anything new in town. Or if you just arrived in a city and want to buy some things from the merchant, and do not wish to haggle, just buy them, without needing to play out the scene.

Solo engines and tools

What follows is a list, which is by no means exhaustive.

MUNE, a simple complete Oracle, and Engine, that is a great way to get started in solo Roleplaying.

Mythic GME, a full-fledge Oracle and Engine, complete with examples. May be the most popular of the engines at the moment. A great read, if not used as is. I prefer the Mythic Variations 2 for running Mythic, and also, for expanded reading, the first Mythic Variations.

CRGE, an Oracle and Engine, while a bit cumbersome to use, it’s a great read.

UNE, an NPC generator. Can generate NPCs, motivations, and helps to set up discussions.

BOLD, a waylays and background story generator, and a connections mechanic that can be used to generate intra-party banter.

Recluse a simple Oracle that uses the mechanic of false presuppositions.

The Terrain Randomizer, a tool that generates interesting terrain for your scenes.

Scarlet Heroes, a full fledged RPG and solo Engine, combined in one. Its intuitive mini-game mechanics are borderline on the RPG/boardgame style.

Ironsworn, a full fledged RPG and solo Engine, combined in one. An excellent approach to solo roleplaying.

Motif Story Engine, another solo engine, with very interesting concepts. Is designed with ‘patches’ that you choose, for a solo engine to your liking.

Motif Character Engine, apart from character emulation, also provides player emulation, for a solo GM experience. Also with patches.

Blade and Lockpick, a solo game engine and ruleset in one. Allows easy running of a group of characters.

Freeform Universal, and RPG so simple and lightweight, using a mechanic that feels almost like a solo Oracle, that is just ready to be soloed.