Thursday 27 October 2011

S is for Sandbox, Part II: Why System Matters (3)

Last time, we continued our look at why system matters, and I left you (still) partway through a list of features that a good sandbox game should have.  We also reiterated the important point that those who say “System doesn’t matter” either have ulterior motives, or have not examined the relationship between system and play experience very closely at all.

We continue with:

6.  Speedy World Creation:  If it is important for the poor, overworked Game Master to be able to create NPCs quickly, the same holds true for vast swathes of world creation.  The more exacting the rules are, the more the Game Master must look up and record in order to create the setting, and the more time it will take.   D&D 3e was the epitome of this, with its statistics for walls, floors, etc.  A game suitable for sandboxing should not require the Game Master to look up information just to put a flight of stairs on a map. 

To put it more directly, if two hours of prep time is all you have, and you can only create a few encounters in that two hours, then you (as a Game Master) are going to be heavily invested in ensuring that the players use those encounters and none other.  OTOH, if you can create several dozen encounters in the same time – far more than the players are likely to need in any given session – you are far more likely to allow them to make choices that determine which encounters get used.

7.  Encourages GM Fiat:  This is related to the speedy creation of characters and setting, as well as speeding play.  It is often better for the game to make a decision and move on.  You can look up the “official ruling” later, if it is important to anyone at the table, but that shouldn’t be the priority during play.

Encouraging GM fiat also allows for customization of NPCs, monsters, and game effects without requiring the GM to do homework.  How much easier to write, “Is blind, but can fight as though sighted” as part of an NPC write-up, than to try to discover the sequence of feats and skills that allows the character to be created “legally”!  How much easier to just an ogre’s stats the way you think they should be to fit a particular concept than to have to look for a particular template (and do the work of applying some of them)!

But encouraging GM fiat isn’t all about speed of play, or even cutting down on the GM’s homework.  The very idea of a sandbox includes within it that the world is worth exploring – it is not simply a generic expression of the rules, but rather a combination of the ruleset and the vision of the Game Master.  Playing in that sandbox allows for real exploration in part because the GM’s vision is as important as the game rules themselves.

This is not to say that the players are unimportant – by exercising choice within the whole, they create the actual focus of play as an amalgam of all participants plus ruleset.  And encouraging GM fiat means that the GM can hold the rules as secondary to the imaginations of all participants.  A cool, and appropriate, character concept need never be set aside because the rules do not account for it.  Likewise, an inappropriate, but “legal”, character concept need never be allowed to drag the game down for all other participants.

Finally, if the Game Master must adhere slavishly to the rules, either there will always be a chance to fall down stairs, break bones, etc. – or such things simply will not, and cannot, happen.

Finally, a good game for a sandbox

8.  Encourages Long-Term Thinking as well as Short-Term Thinking:  If your game is going to last beyond a single session, or a single “adventure path”, the players must be encouraged to consider their characters’ long-term goals.

To be clear, I do not mean “long-term build” here.  I don’t mean how the character will look at various character class levels.  I mean, how the character wants to shape the world around him.

The ability to shape the world around you is a major feature of sandbox gaming.  The game milieu begins life as the domain of the Game Master, but it does not stay so.  It changes in response to PC actions, and wise players can and will learn to make those actions count.  Characters clear wilderness, found towns, create castles, and become lords of the land.  They determine policy, sway kingdoms, and lead men.  In short, they wrest some level of control over the milieu from the Game Master, and make parts of the milieu their own.

And, if the Game Master is actually running a sandbox, this is encouraged.  This is the big reward of the game.  Beyond levels, beyond character power, beyond gold and jewels, is the opportunity to make your choices matter in persistent and important ways.  You may be frustrated trying to do the same in the real world.  You will be frustrated trying to do the same in a railroad.  Your efforts may be resisted and thwarted from time to time in a sandbox, but they should also be rewarded.

After all, that’s one of the biggest draws of the game….and one thing that computer games cannot come close to matching.


The minute you accept that system matters, it then follows that you should have a system that helps meet your goals…or at least avoid systems that work against you!

For my money, the absolute best system for sandbox games available today is Stars Without Numbers, which contains such a plethora of well-made tools targeted at making and running a sandbox that it is simply without peer at the moment.  For science-fiction gaming, the classic Traveller game would be well worth considering as well. 

The original Gamma World game works very well in a sandbox format, as does Mutant Future.

Any early Dungeons & Dragons is good, up to (but not including) the introduction of the Player’s Option books.  WotC-D&D is right out, but the “retro-clones” are right in.  OSRIC, Basic Fantasy, and Labyrinth Lord (among others) are extremely sandbox-friendly.

Not only is WotC-D&D right out, but it is hard to see how either 3e or 4e meet any of the criteria for a good sandboxing game.  At each turn, it seems as though the designers made choices specifically opposed to that playstyle, either through ignorance of the ramifications of their decisions, of because of a different conception of what “fun” or “the story of D&D” is. 

Although the language of 3e (for instance) was inclusive of sandboxing (or “status quo” gaming), the ruleset is not.  Interestingly enough, an examination of WotC modules for 3e and 4e show extremely linear adventures…and I would argue that this is an artefact of the rules as much as of the designers’ conscious decisions.  Perhaps 5e will be better…..?

In any event, I would be interested in hearing the recommendations of others re: good systems for sandbox games.


  1. For a long time my go-to system has been Chaosium's Basic Roleplaying (BRP)... in its various versions and now the Big Gold Book.
    There are a lot of optional bits to the system but at its core it's fast, quick to stat up characters and creatures, not directed at the zero-to-hero mindset of D&D (no levels and abilities only increase if you use them).
    It's not the system for everyone... it's decidedly 'simulationist' and 'gritty'... probably not appropriate for high-powered superheroes... but a I've read your essays every point seems to line up with my experience of it.

  2. That's one I've never played! Thanks for the tip!

  3. Ars Magica, early revisions. You might also want to take a look at Apocalypse World, from Vincent Baker.

  4. The original Ars Magica Faeries supplement is one of the best gaming supplements it has ever been my pleasure to stumble across.

    Thanks for the tip re: Apocalypse World!

  5. Came here via a recent G+ post.

    I agree with much of your reasoning and with the conclusion that some systems fit some play styles better.

    One style feature that needs to match system is the chance of death, or at least the frequency that characters are out of play. In 3.x btb, a status effect (like fear) can put a character out of play for an entire battle. With the length of combat for 3.x, that's a hefty penalty. 4e also has long combats, but status effects are typically rolled for continuation every round, so you at least have something to look forward to. Likewise, with death: if you've got a high chance of character death, you'd be better off with a system that lets you recreate characters quickly. (I don't recall that being covered, but I might have skimmed a section incompletely.) I call it "player death" when players are removed from the game (not Blackleaf!)

    Also, it needs to be said that despite agreeing with you that system should fit style, quite a number of current sandboxing OSR folks (including me) were inspired down that path by the West Marches campaign, which was run with 3.5!

    1. I had a longer reply that got eaten by InterWeb Daemons, but the gist of this is that you are correct. Players sitting bored is a bane to good gaming of any type, and whatever creates "player death" is discouraged in play, whether the designers realize it or not.

      (What level did the West Marches run to? 3e breaks down around 6th level IMHO and IME.)

  6. Everything you say is well-reasoned and makes sense. And yet, I find myself not agreeing on many points. Or at least, knowing it turned out differently for me...

    Clearly there is something in the way I was running 3.5 between 2008 and 2011 that I wasn't doing 'right', i.e. the way I was approaching it fit your sandbox model far more than otherwise. Meanwhile, one-off games I ran from 2000 to 2008 had their ups and downs, and campaigns I played in had their strengths and weaknesses, but likewise did not in all ways conform and leave the impression that the game was at fault.

    I am at a loss to explain this. Could it be that not buying into the premise that every single book published was just as important and crucial and *real* than all the other books published somehow spared me from drinking the kool-aid? Also, could not buying every book they put out factor in?

    I do know that I tend to not care too much what system I'm playing or running, all that much, as long as the GM is adept, or my players trust me. I don't think it's a defensive thing. I mostly care about the characters and what they get to do, story-wise. Was this my armor against the weaknesses of 3rd Edition?

    1. I don't think that there is anything in what I said that excludes running any type of game in any system. However, the argument is that the system either facilitates a particular style of play, or does not.

      Obviously, by limiting the books you use at the table, you limit the impact of the game system that those books represent. If I don't use X, Y, and Z, then it follows that X, Y, and Z will have no impact on my game.

      I ran a 3e game for a long time, and I noted that certain problems didn't arise until mid-levels, and others not until higher levels. Certainly, many problems arose not from the "core books" but from the splats (which made 4e's "everything is core" particularly vexing).

      Being able to say "No" to players is important armour to protect you against the rough corners of any game system. Knowing you what you want is another important piece of armour. For instance, if using minis isn't what you want, and you know it, you can resist the changes moving from 3e to 3.5e that were more mini-centric. Likewise, if you have ideas about how you prefer to handle Initiative (I roll every round, for instance), then you can make those changes smoothly and without disruption.

      System matters. But system is a far cry from being the only thing that matters. Finding a system that matches your gaming style well is a real asset. Having to make due with a system that is designed for something else need not tank your game, unless you let it.

  7. I found your blog while doing a Google search for information about building a campaign around a mega dungeon of all things but that is not what kept me reading this series of articles.

    I am currently running a World of Darkness sandbox game where I have laid some plot hooks but the characters are free to run about and do whatever they wish. In your discussion on systems I feel that the Storytelling system deserves mention for all of the reasons you list. The game I am running has a divergent group of characters with divergent levels of experience, however, they all flow together seamlessly.

    I will be following your blog and look forward to reading the rest of your S is for Sandbox articles.

  8. One system that a lot of people don't seem to be aware of that nonetheless meets all of the criteria you mention is The Fantasy Trip (designed by Steve Jackson and published by Metagaming back around 1980 or so -- though the combat and combat magic system were published earlier as the Microgames Melee and Wizard).

    Talk about quick character generation and simple and fast combat encounters (usually less than half an hour to play out a multi-character combat including magical spells) and that allowed the GM to build a world, or at least an encounter area quickly! It's long out of print, of course, though Dark City Games has published a "rules light" version that allows you to run encounters and some simple role playing events even as a solitaire player. Still, you can find all of the original TFT stuff on line now -- there's actually a web-ring dedicated to it (including documents which translate all the classic D&D monsters and even Empire of the Petal Throne to TFT standards!!), and a yahoo group that has all of the material available for download. The only real "downside" of the system is one that a good GM should easily be able to overcome -- and that is that the games are based on a hex map combat system, which makes it "tough" to depict straight lines such as inns and the like, but again, if you liked Dragonquest from SPI, you can see how they got around that same difficulty.

    I'm in the process of developing a campaign concept right now for a sandbox, and TFT is the system I decided to go with. It's simple enough that an eight-year-old can learn it and play it, but wide open enough that far more experienced players will enjoy the tactical nuances of combat and the role-playing possibilities of a system that doesn't overwhelm you with detail and endless rules....

    1. That's not a system I am familiar with, although I have heard reference to it from time to time.

      That the system is one you like, and one that does what you want it to do, is the important thing.

    2. And here we are nearly five years later with TFT back in print with some updates that address its big issue, stat creep.


Note: only a member of this blog may post a comment.