Monday, April 29, 2024

REVIEW: How to Make a Fantasy Sandbox

There were two great obsessions at the dawn of the Old School Renaissance: megadungeons and sandboxes. Each was a distinctive element of many of the foundational roleplaying game campaigns of the 1970s, like Blackmoor, Greyhawk, and Tékumel. Their rediscovery and promotion are among the lasting impacts of the OSR – so much so that both massive dungeons and open-ended hexcrawls are now permanent fixtures of the even wider RPG scene. 

Of the two, I'd say that megadungeons are probably the better understood and more commonly used, thanks in no small part to the many examples of them now available in print. Furthermore, a megadungeon is, in many ways, just a scaled-up version of a dungeon and almost everyone who's ever played a fantasy RPG, whether tabletop or electronic, knows what a dungeon is like and how it's constructed. However, sandboxes are, in my experience, both less common and less well understood. There are no doubt many reasons for this, but a big one, I think, is they require more preparation beforehand by the referee and preparation of a less formulaic sort than what's employed when designing a dungeon, regardless of its size.

Fortunately for those of us who enjoy fantasy sandboxes – my ongoing House of Worms campaign, for example, is something of a sandbox – there are resources out there to aid in their creation. The very best of them has long been Rob Conley's twenty-four part series on "How to Make a Fantasy Sandbox," whose first post appeared in the far-off time of September 2009. It's a terrific collection of blog posts, filled with good ideas and wisdom drawn from years of refereeing sandbox campaigns. I long ago bookmarked many of the posts and refer to them often in my own work, such as designing the Eshkom District for Secrets of sha-Arthan. If you've never read these posts before, I highly recommend you do so.

An equally good – maybe better – option would be to purchase How to Make a Fantasy Sandbox, 180-page compilation of Conley's blog posts, rewritten and expanded with examples, maps, and artwork, available in PDF, softcover, or hardcover formats. In the book's introduction, Conley provides both a nice overview of what a sandbox campaign is and his own reasons for enjoying them:
One of my favorite things to do with Tabletop RPGs is to create interesting places with interesting situations and then let the players trash the setting in pursuit of adventure.

That certainly encapsulates much of the fun my players and I have had with my House of Worms campaign. He goes on:

My focus is not to create any type of narrative. Rather, I focus on helping my players experience living their characters' lives while adventuring. It's called a sandbox campaign because like in life, the players are free to do anything their characters can do within the campaign setting.

This wide-open world with unlimited choices can be very challenging as a Game Master/Referee. The key to dealing with this challenge is organization. A systematic approach is needed to break down the enormous task of dealing with an entire world. Organized into bite-size chunks that one can do in the time they have for a hobby.

Once again, I think Conley has done a fine job here of distilling the essence and unique pleasures of a sandbox campaign, while also recognizing that creating and maintaining such a campaign is not always easy, hence the need for a guide such as this one. 

With that out of the way, he first describes and then elaborates upon thirty-three distinct steps in the process of designing a fantasy sandbox, from creating the map to placing settlements and lairs to choosing a "home base" for a new campaign. It's all presented clearly and methodically, so that it's easy for even a neophyte to follow. Best of all, Conley includes lots of examples throughout, drawn from his own experience of making fantasy sandboxes. Indeed, I'd go so far as to say that these examples are among my favorite parts of How to Make a Fantasy Sandbox. They not only serve as illustrations of design principles, but they also give some insight into Conley's own gaming past, which I found delightful and inspiring.

Throughout the text, the Isle of Pyade serves as the main example of how to implement the thirty-three steps to creating a fantasy sandbox setting. I found this very useful, because it's eminently practical and concrete rather than merely theoretical. If you follow the steps through, one by one, you'll see Pyade grow out of a blank hex map into a fully-fleshed out and complete location. Whether you're a novice or an old hand at this sort of thing, you'll learn a lot from the example of Pyade.

By the conclusion of How to Make a Fantasy Sandbox, the Isle of Pyade is now ready to use. However, many of its details – maps, NPCs, encounter tables, etc. – are scattered across its 180 pages, making it less suitable for use as a reference. Should you wish to make use of Pyade yourself, a better option might be the separate The Isle of Pyade book (available in electronic, softcover, and hardcover format), which takes all the relevant details and consolidates them in one place for greater coherence and ease of use. There's also some additional content in the form of artwork and color reproductions of the original maps Conley made in the late 1980s. If you're fan of RPG "archeology" as I am, this only adds to the value of The Isle of Pyade.

How to Make a Fantasy Sandbox is a very good book, one I am very happy to own and one I am certain I'll refer to and make use of in the years to come. In addition to all the thoughtful insights and clear instructions Conley provides, he "shows his work," which is to say, he lays bare how he works and why, right down to including very useful appendices of resources, hexmapping guidelines, travel and encounter rules, and information about the process behind creating Blackmarsh, his earlier published sandbox setting. Aside from some minor quibbles about editing, I have only praise to offer about this book. If you have even the slightest interest in creating and refereeing a sandbox campaign, consider picking up a copy. You won't regret it.


  1. Didn't realize there was a compiled and revised publication of these old posts. And I have a birthday coming next month. Guess I'll be updating my wish list. :)

    While I love the concept, I've always had an issue with using "sandbox" to describe this sort of game. Bad childhood memories of the actual sandbox my parents foolishly built for us when we were young. Do not mix farm cats with sandboxes, people. It's just plain gross.

    1. Even just suburban cats + sand-pit = bad news! As for the OP, I really don't have time to work up a new campaign setting ... really ... oh, who am I kidding...!

  2. Not thirty-three, thirty-four steps. He forgot about random encounters, and when he gets around to it he admits it's not his strong suit and provides some rather lacklustre examples.

    I am not a notably clever or creative person. I have only a few years experience of running regular hexcrawl play. Mr. Conley's wisdom must surely dwarf my blind fumblings. So why do I feel like everyone in the OSR is sleeping on wilderness encounter tables? If the majority of hexes are unkeyed, then the random encounters should not be an afterthought: they are a critical aspect of gameplay.

    In a dungeon one can get away with minimalist wandering monsters, since there's a strong context and environment in which to place them. Not so much in a hex crawl. Moreover most people designing a megadungeon wouldn't leave 80% of the rooms unkeyed as they would for a hex crawl. With spreadsheets, it's trivially easy for anyone to create their own nested tables and generate results instantly - or, if not using a computer to aid play, pregenerate them and keep them at the back of your folder. Yet so often encounters amount to the likes of "2D6 wolves - wing it". Improv is a necessary skill but I can't regard leaving such a significant component of play so totally improvised as being good design.

    I second-guess myself about this, because I can't be the only person who's thought along these lines, so why is something that seems self-evident from my experience not common knowledge? Have I got the wrong end of the stick? What aspect of hexcrawling have I missed that makes a highly developed encounter table undesirable or superfluous?

    1. Yes. And I’d argue also that having a good random encounter chart for a wilderness is critical to describing what that wilderness is. In a book one could get away with a flowery bit of flavor text, but in a game the important element is what lives there and what happens there.

      I tend to break my wilderness encounter tables into two parts: the kinds of encounter, and the actual encounters. The result is that I can tell what this area is like with only a quick glance at a simple table:

      01-55 Normal animals
      56-75 Fantastic creatures
      76-90 Natural encounters
      91-98 Humanoid creatures
      99-00 Civilized folk

    2. Two answers, first the problem I have with random encounter tables is that they are too general. Particularly, if the group is in a urban or wilderness/rural environment. A tailored encounters would fix the issue but in a sandbox campaign with the party always on the move trashing things across the setting. It hard to keep coming up with tailored charts for every new locale and situation.

      This brings me to number two, which is the travel rules I do include in the book. Out of all the random encounter systems I used, the journey system in Adventures in Middle Earth was the best. So, I made my own take that is more suited for D&D in general. But knowing that there are folks like you who have a set of beloved tables (their own or purchased), I made room for that in my Travel rules by having a random encounter result. A feature that the AiME journey rules did not have.

      For me, the travel rules provide enough structure to create encounters on the fly that are tailored to where the PCs are. It also takes a lot less work to manage throughout a campaign.

    3. I agree that specificity in random encounter tables is highly desirable, which is part of why I love Greyhawk’s national and regional WM tables from the 1983 Glassography booklet. They’re not perfect (omitting Hardby, and the oceanic bodies of water got very short shrift), but provide an excellent model for how to structure such tables for AD&D hex-based travel, as well as to help a DM see how such tables reinforce locale-specific flavor (Rhenne in the City of Greyhawk lake and river systems, blue-furred bugbears in the Land of Black Ice) and at the setting level (where the human-centric design approach is foregrounded strongly).

      I’m very pleased that Black Blade will stock copies of Rob’s new books at our booth next month at the North Texas RPGCon, and will buy a copy myself if we haven’t sold out by Sunday!


  3. I backed the Robert Conley's Kickstarter out of loyalty for his bailing out of the City State of the Invincible Overlord Kickstarter and got an even better resource! Who knew!

    1. Thanks for the compliment. I'm glad you like it!

  4. I have nowhere near the sandbox experience required to contribute here, but I applaud the discussion. We actually departed the D&D realm to prowl across the Piedmont region of Northern Virginia, and random encounters were many, day and night. By comparison, while crawling around the sprawling storm-sewer networks underground, encounters were few (although you do find yourself face-to-face with snakes in soggy leaf clumps) and could have been anticipated if I had greater facility. I can only imagine the difficulties in tailoring randoms for a hexcrawl. It's like walking in raw off the street and sitting-in with a jazz quartet. I prefer the copperhead fever.