The Solo Sandbox: What Do We Know and How Do We Know?

Sometimes it can be difficult to solo role-play, it may feel too much like a writing experience instead of a gaming experience. This was a common problem I faced when I began with solo play. You want to explore the game as a player and maintain the uncertainty and surprise which you have in a normal RPG. But as there is no GM and you also fulfill this role (with the help of some tools, i.e. the Mythic GME (aff)) the line can become muddled.

How can we maintain the feeling of surprise and excitement that comes with discovering the world and the NPCs as a solo role-player?

The idea of playing a sandbox appeals to me more and more. The freedom of exploring the setting and of emergent play looks certainly interesting.
My gaming group plays D&D 5e in the Forgotten Realms, Icewind Dale. I’m a player, but as far as I understand it it’s almost a mini sandbox as there are different hooks which you can explore.
Recently I reviewed Yoon-Suin, an oriental sandbox setting for old school D&D. I really like the special snowflake backdrop and want to play in this world.
Yoon-Suin asks you to create the setting yourself. A broad outline exists but the GM basically creates the details with tons and tons of tables: NPC relations, rumors, PCs’ social circle, different factions etc. before beginning play.
What if I do that with solo role-playing? Wouldn’t that feel boring as I as the player already know so much?

How can we make a sandbox interesting for solo RPGs?
I asked this question at G+ and received some very good answers. Here is an excerpt of the ideas, for further reading I recommend the original posts (links below). The ideas are not my own but I know that e.g. John Fiore doesn’t have a public blog, so I hope the collection here makes them more available to the public.

What do the PCs know? Why don’t they know what’s out there?

First, it helps to establish a frame. What is the knowledge level of the PCs? Why do they know the setting or why don’t they?

The adventure begins at some locale, be it a town, a city neighborhood, a dungeon, a small village in the Borderlands or the wilderness. Are the PCs familiar with this place, is it their hometown? Or are they foreigners who freshly arrived at the city? Perhaps they only know their immediate surroundings, this district of town but the rest not so much?

Let’s say that the PCs want to explore the wilderness, completely uncharted territory. In this case, they almost know nothing, so it makes no sense to pregenerate a list of rumors, hooks or hex map locales. Instead, you gradually generate these things with randomizers, setting books etc. in play.

But what if the characters already know a bit more about the setting? Perhaps they are already familiar with this town but want to make a living as daring adventurers? Here you can pregenerate some lists (NPCs, story hooks, rumors etc.). The twist is that some information may be false, so it’s a good idea to have a system which determines if the information is accurate. It can be something simple like a binary yes/no die roll or something more sophisticated like FU‘s yes-no-no-but-etc. system.

To put it in the words of Vincent D. Baker: “What’s at stake?” In the end, every role-playing game becomes interesting by weighing the risks against the rewards. Both are partly determined by the level of certainty the players have, what they know for sure and where they can’t be positive about their knowledge. As a solo player, you may want to keep this mechanic by incorporating the possibility that some information may be false.

It also makes sense to create the PCs first and to think about how they fit into the setting with the questions above. One could create some backstory, relationships and motivations to tie them to the setting. Depending on their level of knowledge or ignorance of the setting, use your generators to create a starting point.

Another idea is to use a whole bunch of different generators, for example the D30 Sandbox Companion (aff), UNE, Mythic GME, the solo tools from Scarlet Heroes, Rory’s Story Cubes, Zero Dice and more (links are here). Don’t decide beforehand what you use. So there is also an element of surprise when you need to consult a table or randomizer.

How do we find out? How do we ask questions?

Like in a “normal” tabletop gaming experience there comes a time where we need more information. To enhance the playing experience one should maintain the distinction between being a player and being a GM/GME.
Solo role-playing is clearly not the same as playing in a group, especially not like traditional RPGs where there is a clear task allocation between the GM (who plays “the world”: she populates the setting, plays the NPCs etc.) and the players who mostly only play one character.
If you constantly cross the line between these two roles, the solo experience can fall flat, as it doesn’t feel like role-playing anymore, only storytelling.


The PCs are looking for a culprit and are questioning a local vendor. Does the vendor know who the culprit is? How do they PCs find out?
There are two ways we can tackle this situation. They are not mutually exclusive, but it is important to be aware of the distinction.

Do we ask from the players’ perspective (gaming in first person)? Then we need to talk to the contact, bribe him, convince him or intimidate him. If your players have some abilities in social situations, these will now come into play. Depending on your gaming system, this might mean a skill roll or ability check. However, we are not “allowed” to ask questions which we couldn’t experience as a player. In this role, we are restricted to the senses and possibilities of a player: what we can feel, see, hear, taste, smell or what we can ask or find out in another way.

As the GM/GM Emulator (third person) the question might be “Who is the culprit”? In this case, it makes sense to use a GM Emulator system like the Mythic GME or something open-ended like Rory’s Story Cubes or other image generators instead of the role-playing system to come up with an idea who this person might be.
So, after the PCs have established a connection to the vendor in some way (skill checks etc.), I ask: “What can the vendor tell me?” or “Who does the vendor think is the culprit?” and then I consult my idea generator.

John Fiore also gives some more tips:

I only ask “bigger” questions than those requiring merely a yes or no answer. For example, “Who am I?” followed by a roll of three RSCs is far more interesting to me than a series of yes/no questions (like “Am I a wizard?” No. OK, then “Am I a cleric?”).


I don’t put a limit on the inspiration. I ask a question, roll, interpret, and then keep on “interpreting” as long as the inspirational fire is hot.

Hopefully, this blog post was helpful. Again, these are not my ideas, they come from the G+ solo RPG community but I wanted to collect them into one source. The original threads can be found below where more information on the topic is available.

The Solo Sandbox
About questioning, scroll down a bit

Big List of Solo RPG resources