Building a Sandbox

[Reynard]

I am not a fan of "adventure paths" or campaigns where there is a set plot/story/goal. IME, players are just too unpredictable to try and lay out any long term plans.

Since my current game is on hiatus due to various players' real lives, I have an opportunity to devise a new campaign and I think I am going to create a sandbox -- a geographical region with enough opportunities for adventure that I essentially don't have to do any prep work at all once play starts (or, to put it another way, I am going to front load all my prep work). I'll probably use 2nd Edition for personal preference reasons, but I don't think the idea is edition-specific. It wil be incumbent upon the players to enagge the setting and the game and create their own adventures.

So, what are the most important elements for building a sandbox? I think some of them are:

1) Multiple levels of play, all at the same time: That is to say, there should be 1st level and 20th level stuff to do, already in place, throughout the sandbox. If the goal is player-motivated adventuring, having the full range of options available is important. if the PCs, at 3rd level, decide to enagge the 10th level villain, they should be allowed to do so. conversely, if at 10th level they decide to obliterate the 3rd level bad guys -- out of revenge, spite or just as a "training exercise" -- they should be allowed to do so. Put simply, the sandbox shouldn't scale with the PCs; it should just be.

2) Interesting Location: I am not sure how many is the right amount, but pre-creating adventure sites -- dungeons and mystic wells and haunted forests and the like -- is an absolute necessity. Given how much I dislike making maps, I'll probably just crib from other sources, and even use pre-statted dungeons from adventures as exiting locations. After all, it isn't much of a sandbox if there's no where to go. This includes the lairs of any powerful monsters like dragons.

3) Politics: Not specifically rulership type politics, but as a general category of existing relationships between power groups and individuals that will inform how those groups and individuals interact with the PCs and respond to the PCs' actions. i think the easiest thing to do here is create those groups and individuals, then map out a web of alliances and antipathies between them, with a few notes about what kinds of things the PCs can get from them and what kins of things the PCs might do will attract their attention.

4) Room for the PCs to Grow: I like it when players want to get involved in the world and change the setting through play. there shouldd efinitely be room for the PCs to start their own organizations, become rulers, take on non-traditional adventuring roles (merchants, mercenaries, etc...) Players should feel that they can do more than just wander the landscape looking for dungeons and lairs. they should feel they can become important in the region. if they don't want to, that's fine, but it is better to have the option.

Anything I'm missing? What do you think? Do you prefer sandbox style gaming, or something else?

 

[S'mon]

I'm a big fan of sandbox, though I don't think you can do *all* your prep before play begins, but you can certainly create a robust framework that allows for easy further development or winging it.

I would not attempt to create a sandbox that covered the entire 1-20 level range. This raises too many problems of versimilitude and would require too large an area map. For a campaign starting at 1st level, anything from 1-6 to around 1-12 should work.

I find a good approach is to:

1. Start with a bunch of site-based adventures, covering a decent range of levels but with an emphasis on lower level play (1-3 or 1-5). I generally use published scenarios, tweaked to taste.

2. Build an area map around these adventures. You need not assign a scale at this stage.

3. Go to the continental map of the campaign world, which may be Toril, Oerth, your homebrew, etc. Find a suitable place to 'slot in' your area map, assigning scale at this stage. Usually something between 5-12 miles per hex or per cm is good.

4. Add a few details on major locales - towns, cities, and such. Perhaps create a few larger-scale maps of areas where you expect a lot of activity, I find 200 yards/hex works well for local maps, between 1200 yards/hex and 2 miles/hex for the next stage out. You may be able to run an entire campaign on a 2 miles/hex map, but some players may find this too cramping. Make a few notes on major NPCs, but only detail any you expect the PCs to be dealing with directly.

5. Decide on a starting locale for the PCs, with rumours to get them started. You can either have a particular starting adventure planned, or let the players choose from 2-3 suitable ones.

6. Begin, & have fun.

 

[S'mon]

A potential problem with sandbox in 3e is that the core system expects very rapid levelling, which requires encounters closely tailored to PC levels. If you create an area with much adventuring suitable for beginning PCs, much will be wasted or require extensive adaptation. 1e/2e, B/X BECMI D&D and C&C, among other systems, do not suffer from this to anything like the same extent, so are more easily used in sandbox play. However 3e can be used for sandbox if you reduce the rate of progression by cutting XP awards. Necromancer's Lost City of Barakus campaign does exactly this, presenting a pre-created sandbox setting suitable for play in the 1-5 level range, with XP awarded at 1/2 rate.

 

[Woas]

Perhaps you are blessed with super-inquisitive players that eat up fantasy world history and anthropology, but for me sandbox games end up being 100% wasted resources on my part (the DM) and quickly sputtered out.
What's important though is that what your saying is good for a sandbox game is actually good for any type of game. Which is basically variety and a strong foundation that plants the characters in the world.

 

[shilsen]

Personally, I really like sandbox-style games, and have been running my Eberron campaign for over 2 years with mostly that approach. From a logistical standpoint, here's one suggestion I have about the following:

2) Interesting Location: I am not sure how many is the right amount, but pre-creating adventure sites -- dungeons and mystic wells and haunted forests and the like -- is an absolute necessity. Given how much I dislike making maps, I'll probably just crib from other sources, and even use pre-statted dungeons from adventures as exiting locations. After all, it isn't much of a sandbox if there's no where to go. This includes the lairs of any powerful monsters like dragons.

Since there is only a finite amount of time and energy any of us can devote to DMing and preparing for a game, I prefer to minimize preparatory work and ensure that everything I prep will see use. So when it comes to adventure sites and/or potential plot hooks, something which I do is to not bother actually creating them beforehand. I will give the PCs enough information for them to know the place/plot/option exists, but I won't sit down and actually stat out things unless I know they're heading there. I highly recommend doing the same. For example, let the PCs know that there's a haunted dungeon below the ruins of some castle, that the great wyrm Whatzisnehm lairs on mount Overthere, and that there are job openings for guards for a caravan heading across the Hotanddry Desert, but don't bother actually putting together the dungeon or the dragon's lair or the desert until the PCs actually get there. Just have some preparatory ideas about what's there and ad lib the rest if they start digging up info. It's worked great for me and I wouldn't do it any other way.

Anything I'm missing? What do you think?

I think you've got most of the stuff you need already covered. One thing I recommend (which you've probably already planned on) is having some explicit indicators and hooks for things the PCs can do. Most players/PCs like to have at least some potential directions indicated for them, especially right at the start of a campaign. I always run campaigns without a pre-planned plot and the way I start them off is by dropping the PCs into the middle of an action-packed moment. Then, by the end of the first session and into the second or third, I'll drop half a dozen plot hooks on the PCs, making sure they differ drastically and that some have links to the PCs' backgrounds. Depending on which one the PCs pick up on and what they do with it, new plot hooks will open up (usually multiple ones at once), and more will appear due to their backgrounds, interactions with NPCs, choices (both what they do and what they choose not to do), etc. At the same time, I make it clear (both IC and OOC) that they are free to ignore all the plot hooks and go do something of their own.

Do you prefer sandbox style gaming, or something else?

Much prefer the former.

 

[Jack7]

A potential problem with sandbox in 3e is that the core system expects very rapid leveling, which requires encounters closely tailored to PC levels. If you create an area with much adventuring suitable for beginning PCs, much will be wasted or require extensive adaptation. 1e/2e, BE/X BECMI D&D and C&C, among other systems, do not suffer from this to anything like the same extent, so are more easily used in sandbox play. However 3e can be used for sandbox if you reduce the rate of progression by cutting XP awards. Necromancer's Lost City of Barakus campaign does exactly this, presenting a pre-created sandbox setting suitable for play in the 1-5 level range, with XP awarded at 1/2 rate.

That's a very good point and the reason I do not assign experience points based on conquering opponents according to the level of the opponent, but rather to the difficulty of over coming that opponent in the game situation, modified by other factors of course.
See my answer to 1) below.

1) Multiple levels of play, all at the same time: That is to say, there should be 1st level and 20th level stuff to do, already in place, throughout the sandbox. If the goal is player-motivated adventuring, having the full range of options available is important. if the PCs, at 3rd level, decide to engage the 10th level villain, they should be allowed to do so. conversely, if at 10th level they decide to obliterate the 3rd level bad guys -- out of revenge, spite or just as a "training exercise" -- they should be allowed to do so. Put simply, the sandbox shouldn't scale with the PCs; it should just be.

Real life doesn't scale "in game terms" and to tell you the truth, I don't think the game should either.

And of course I know nobody in real life runs around with a silly "level number" attached to them either, they reach a certain level of competency through personal social, group, and political maneuvers. But I have encountered, in real life, men far lower on the social order in any organized group, let's say an organized criminal gang in this example, who are far more dangerous in reality than a team boss. Small cell terrorists who are personally far more dangerous than their organizational leaders.

So level should be no real consideration (other than necessary in-game mechanical ones) for how dangerous an opponent is. Brutality, psychological abnormality, cunning, drive, will, determination, ambition should be. Sneaky, crafty, brutal and vicious low level enforcers can be far more dangerous, and even highly organized and efficient in their own ways and methods of operation, than bosses and body-thugs. As a matter of fact, many times, the higher in level (in respect to any given group association through which they operate) most individuals become the less personally dangerous and willing to engage opponents directly, or even directly involve themselves in dangerous and vicious operations. So, I'm with you, dangerous should definitely not equate to level, even at a mechanical level, and, as in real life, people should be able to engage any particular opponent as to need, and as circumstances warrant, not by some artificial contrivance of "game need" or challenge rating.

The same can be said to a certain extent about situational and geopgrhaic encounters (dungeons, adventures, mission scenarios, even training exercises). Some locals have a reputation for being especially dangerous to operate within, but in every locale you have dangerous opponents and harmless opponents and well organized and efficient opponents and incompetent and ineffective opponents. So in truth there are no 20th level dungeons, or first level dungeons, just places where you have especially effective and vicious opponents on average, and places where you have not so effective and not so well-prepared enemies on average. Players and characters should be able to assess such situations and adapt themselves accordingly, no matter what level is artificially affixed to either individual or locale, if they have any experience and foresight at all. And so no place should necessarily be too dangerous (unless they are just completely and helplessly overclassed and/or outgunned) and no opponent too weak if that opponent is properly played, instead of just played as a preconceived and static automaton.

3) Politics: Not specifically rulership type politics, but as a general category of existing relationships between power groups and individuals that will inform how those groups and individuals interact with the PCs and respond to the PCs' actions. I think the easiest thing to do here is create those groups and individuals, then map out a web of alliances and antipathies between them, with a few notes about what kinds of things the PCs can get from them and what kinds of things the PCs might do will attract their attention.

As with the interesting location idea I think this is a good start. However, do not in my opinion, map out alliances and working relationships to such a degree that they are not open to flexible variation and open to "situational and interactive awareness."

Because A hates B does not mean that a working relationship cannot develop between these two because both have personal reasons to kill X, or promote W.

Plus circumstances, rather than just preconceptions, often determine realistic relationships between individuals and groups. As information and intelligence develops over time someone who thought Z was their adversary discovers that Z would actually make a valuable associate, partner, or even potential ally.

So open all relationships to adaptation over time. You might want to keep a flow-chart, as if you were tracking cell or organizational links, and then allow it to alter over time in interesting ways in order to track changes in developmental progression.

You might also have sub-categorical notations on political, financial, social, religious professional, and personal relations in order to establish interesting sub-plots and themes and adventure/mission probabilities. You might also want to encourage various players to keep such "apparent relationship charts" so that they can take note son developing circumstances and even devise their own operations and scenarios based upon what they know, or have discovered. My players often develop their own infiltration operations, undercover ops, or recon missions based upon what they have discovered through their own Intel gathering. And of course Intel gathering is really the art of establishing, both overt and covert, relationships with those who have information of a valuable nature to share.

But in any case good Intel is a constantly shifting set of relationship parameters. For instance A might be allied to C for financial and political reasons, but because of a personal and/or professional matter, thus links become overtly or covertly strained or even shattered. Setting off a whole new chain of interesting in-game events.

4) Room for the PCs to Grow: I like it when players want to get involved in the world and change the setting through play. there should definitely be room for the PCs to start their own organizations, become rulers, take on non-traditional adventuring roles (merchants, mercenaries, etc...) Players should feel that they can do more than just wander the landscape looking for dungeons and lairs. they should feel they can become important in the region. if they don't want to, that's fine, but it is better to have the option.

I completely agree here. Much of what is lacking in many settings is the sense of realism and of real causes and of real relationships and enterprises, rather than mechanical and artificially contrived ones based upon mere "structural" character and milieu considerations. In other words the character should be far more than just an wooden and even petty "magical mask" that the player assumes, but rather the way in which the character interacts with his imaginary world should allow "player seepage" into that world, so that the player can explore things about their own strengths and weaknesses through the character.


We play adventure paths too by the way, but, they are not static and dissociated from the world in which they exist and operate, and we also play "sandbox" as you have basically outlined the parameters here. With us it is pretty easy because, first comes the world) - which is basically our real world, so it is easy to create both semi-historical and mythological campaign lines, secondly the player) - because to me the game is about the player first and foremost, the character secondly, and not the other way around, thirdly the campaign or quest) - as a focal point for individual or goal activity and to set a common objective or set of objectives, and lastly the adventure or mission scenario) - the actual specific objective to be met.

Some players are free to move so far through a particular scenario or adventure, then leave off of that and go on to something else, or finish what they started initially, and whereas not all paths converge nor are all adventures related one to another, many are. And sometimes players are forced off an adventure, through circumstances beyond their control, because in the middle of an operation something more important develops on another battlefield or at another location.

I also allow villains, NPCs, characters from other parties, monsters and what not, that might normally be associated with another set of circumstances or another location, more or less open movement between scenarios and locales. Monsters and NPCs and others are not necessarily tied to static locations or particular treasures or singular causes, but are themselves fluid in the way they relate to the world, as are the players and the characters they play.

So in my opinion fluidity and realism enhance in-game experiences and make for better role play and situational development.
That style of play is also more like real life and far more interesting, useful, and fun than mere imaginary escapism.

Anywho, good luck.

 

[Rallek]

Unless I'm running a 1-3 month "mini-campaign" to fill a gap between "real" games, I always run sandbox style games. My players expect (read as: DEMAND) that kind of freedom and depth. They want to know where the coins are minted, where the mines that produce the gold are, where the grain in the bread they're eating came from, and why The Three-Legged Dog seems to have a better wine cellar than it should. Of course you have a number of plot-hooks to dangle and see if the party wants to go that way, but having them make their own fun (if that's what interests them) is just fine too.


When it comes to Sandbox games, world-building that actually makes sense, and (in my opinion) 3e in general, one has to give E6 a try. My group recently started using E6 rules, and we're never going back to "normal" 3.Xe.


If you're not using E6, what I would advise is that you build for the first 10 levels of play or so, and then increase downtime. Once the PCs can comfortably handle the biggest "regular" threat in the region, they spend more time building up their resources and spreading their influence. They may still spend 90%+ of every session adventuring if that's their thing, but the adventures that they take on now are the kind of threat/quest/opportunity that may only come along every year or so. Once they hit lvl 15 or so, they may only go "Adventuring" every 3 in-game years or so. If they keep playing past that, maybe they only get lvl 20+ adventures ever 5-10 years. In my opinion it's time to think about retiring the characters and starting some new ones. The new ones can be retainers/hirelings/apprentices of the old ones, or completely unrelated. This lets them keep building the history of the setting through progressive generations, and see the long-term effects of things that earlier characters have set into motion. It also allows you to pull some treasured past characters out of "retirement" to deal with a major threat to the setting every once in awhile. That can be a blast.



As always with this hobby, YMMV.



EDIT: I once had a group become titled landowners who started an organization that later became the dominant alchemical guild in the region. They later played "Robin Hood" types dedicated to the overthrow of the "repressive" guild structure that their former characters had put in place earlier... they actually met their former characters as antagonists several times. That was a fun series of games.

 

[Reynard]

Perhaps you are blessed with super-inquisitive players that eat up fantasy world history and anthropology, but for me sandbox games end up being 100% wasted resources on my part (the DM) and quickly sputtered out.

See, I don't think it is wasted effort because if the goal is to be able to run on the fly, without having to fudge stats or make up stuff that is better thought out well beforehand, all that work helps you do so. Even if the players never decide to delve into the intricacies of the relationship betwen the Assassin's Guild and the Temple of Vengeance, the fact that you, as DM, know those intracacies means you can do a better job when the players' choices bring them into contact with those two organizations.

What's important though is that what your saying is good for a sandbox game is actually good for any type of game. Which is basically variety and a strong foundation that plants the characters in the world.

True, but again the point is to avoid pre-planned gaming and let the players decide what it is they are going to do. I think the broad foundation is much more important, as is versimilitude and established "rules" of the setting.

 

[Reynard]

When it comes to Sandbox games, world-building that actually makes sense, and (in my opinion) 3e in general, one has to give E6 a try. My group recently started using E6 rules, and we're never going back to "normal" 3.Xe.

The reason I am leabing heavily toward 2E is that it does a lot of things I like, I can use 1E resources with little to know extra work and I happen to be lucky enough to have access to players that don't mind playing an older edition. 1E/2E actually solve a lot of the same problems that E6 solves -- the power curve is smoother and the high end is much closer to the middle than in 3E. Plus, once PCs "cap out" and advancement slows, there's lots of motivation to do other stuff than just try and level. Characters started getting followers and were suggested to start building keeps, etc... at 9th level for a reason.

 

[Rallek]

Apologies, I somehow missed that you were using 2e in original post.


I too played/DM'd 2e for many years, and to be honest, I probably prefer it to even E6 3e. If all of my players still had their books, I'd probably steal a couple of things from 3e and go back to running modified 2e.


Good luck with the new campaign.