D&D 5E How I Sandbox

Zardnaar

Legend
Over the years I have played a little bit of D&D starting way back in 1993 when I was 15. 37 now. For most of that time I have probably used Dungeon Magazine the most and now the equivalent is probably the Quests of Doom adventures from Necromancer games. Outside of Dungeon I generally do not buy adventures apart form the occasional one such as The Night Below, Labyrinth of Madness, or City of the Spider Queen. I do by small PDF ones especially in 2007-2012 or so where I spent almost nothing on D&D.

When I criticize a published adventure for being to railroaded I normally mean that the adventure is very linear, not it is railroaded in the sense that you have the choice to play it or not. Usually rail roaded adventures are plot heavy and have a series of scenes that more or less have to be played in order and you more or less want to stick to the map. Adventure Path type adventures are prime examples although any adventure is rail roaded to some extent. An adventure like Isle of Dread or Kingmaker is more of a sandbox than say HotDQ. A railroaded adventure is not automatically bad (I like The Savage Tide), but some of them are turkeys but might be salvaged with DM work (HotDQ). Not all sand box or smaller adventures are automatically good either.

Now my world. Over the years I have made a rough campaign setting in my head although at times I have done things like design a Pantheon, draw maps, write out a history of the world etc. I have been recycling it since about 1998 although I often change things a bit as the players have changed over the years. Some things have remained the same though.

Z's Sandbox Elements
1. Major themres are the lost past, exploration, naval/water, colonies. Nations generally do not exist except maybe as one or two smaller nations.
2. Late medieval or renaissance levels of technology with lost magitech/modern technology in small doses.

3. Elves are not really nice as such. Some are but the Elven Imperium is in an interregnum with various Elven houses squabbling for influence. Halflings tend to be feral/barbaric (think steppe nomads type), Dwarves are reclusive. Most of the Elven army is mercneary. Roughly based on the East India Company/Carthage.

4. Usually a human nations of knights recently independent from the Elves. Current oneis the Heldannic Knights.

5. City state political system for the most part.

6. Religion is not usually a major theme. I have used the Greyhawk and homebrew pantheons, currently using FR Pantheon.

7. Usually a lost empire/civilization plays a role. There might be one dead for a few hundred years (Rome) and the ancients from 3000 years ago (Atlantis/Magitech equivalent).

8. There is no good and evil as such but a lot of shades of grey. Not all Elven houses are evil, not all of the Human Knights are good.

9. Slavery exists and bad things happen to good people. This can be as arbitrary as a ruler ordering your execution for reasonably mundane offences (or no offence). Partly based on real life as a King of France executed his sisters lover for insulting the King and there have been rulers like Tamerlane, Vlad the Impaler etc. Nazis may have been evil but they were somewhat rational. Not all of my NPCs are rational.

10. Civilization exists in a post imperial collapse (2 empires falling). Technology has been retained, the political system has collapsed.

In my current campaign the PCs are level 9. There is no BBEG as such, I have a rough idea what the campaign is about but the PCs can do what they like. They can get on a boat and sail to another continent for all I care and avoid the minimal metaplot I have put in. Sometimes I prep multiple adventures and depending on where the PCs go it depends on what one I run, or I also ad hoc it as I go along. I often plan a series of encounters which may or may not happen. I ignore the encounter rules in the DMG. Some are very easy or not intended as combat encounters, sometimes PCs attack into overwhelming force and pull it off. Other encounters are fixed though so the PCs will encounter them no matter where they go unless they take extraordinary precautions to avoid them. I do not over do that though as even if I am rarely rail roading something for whatever reason I like the PCs to have illusion of free will. Normally it is repercussions from some group they have annoyed such as assassins dispatched to kill them.

I'm not a brutal DM though as I rarely try to kill PCs. I might attack them when they are unconscious on the ground and in 5E critical hits can kill them. I do run a bit of a hard luck game. I might fudge the dice if I screw up and allow the PCs to survive, I do not fudge the dice if they do something stupid. Some NPCs will go out of their way to kill them though so if they are on the ground on 0hp they will get attacked. Rarely NPCs will attack them at night while they are sleeping or try and lure off individual PCs in town in an effort to split the party up. This is more of the part of the bad things happen to nice people thing. I try not to over do it though but it is more to make the PCs realize there are repercussions to their actions, they can't take short and loing rests for granted and if they nova off for 5 minute work days it can back fire on them.

For example some of the spellcasters like dumping all of their spells early on and then long resting in a safe town. That is fine but for example but I might respond by making the next days encounters a lot tougher as a dungeon is on alert or a key NPC might be sacrificed or they miss out on treasure as a boat leaves port or a dungeons inhabitants flee taking the loot with them. Dumb creature might stand and fight, things like Drow will leave and get reinforcements. PCs can do what they want and play how they want the world might have repercussions (CN PCs tend to hate this). Bad things can happen to nice people in my games, worse things happen to bad people if the PCs are jackasses to the NPCs.
 
Last edited:

log in or register to remove this ad


How I sandbox: Every adventure has two plothooks to two different adventures. Some plothooks might lead to the same adventure (but never have two plothooks that lead to the same adventure in the same adventure). Adventures are designed with rough levels in mind but may not be level appropriate when handed out. Players should have the ability to learn more about the adventures so they're making (somewhat) informed decisions and not relying on the meta game of "John would never throw something at us that could kill us." Fights are largely handled as a series of small fights that can be combined to more deadly fights. Attrition is a major part of the adventures.
 

How I sandbox: Every adventure has two plothooks to two different adventures. Some plothooks might lead to the same adventure (but never have two plothooks that lead to the same adventure in the same adventure). Adventures are designed with rough levels in mind but may not be level appropriate when handed out. Players should have the ability to learn more about the adventures so they're making (somewhat) informed decisions and not relying on the meta game of "John would never throw something at us that could kill us." Fights are largely handled as a series of small fights that can be combined to more deadly fights. Attrition is a major part of the adventures.

I call my multiple-choice campaigns like this "sandboxy" rather than "sandbox". I have one like this currently, an episodic Classic D&D Karameikos campaign.
My 'actual sandbox' Wilderlands campaign does not have 'adventures'; it does have 'sites' - locations such as dungeons where adventure is particularly likely. But the PCs aren't required to investigate all or any of these. It's also notable for the ability of the PCs to go anywhere & do anything adventurous and it will run smoothly, there's never an adventure of the week that the PCs must do - not even a choice of two or three.
Sandboxy games are choice matrices; genuine sandbox games are completely open. Both are distinct from linear games like the Pathfinder AP I just finished, but they're quite distinct I think.
 

How I sandbox:

1) Give the players a premise for the campaign, along the lines of "How about the characters being students at a wizard academy and investigating strange secrets within the academy?", and moving on only once the players are cool with the premise.
2) Have the players all make characters as a group, sharing ideas and talking about what they think might be cool to have happen (taking notes of anything they seem really excited about for later use, as well as anything that comes up that sounds like it would bore them).
3) Start playing - make up on the details that the players need when they actually need them, let them lead the action (decide what to do) so I only have to think of part of what would need to be thought up if I had prepared in advance (the obstacles in the way, the results of actions, and the NPCs interacted with - but never what could/should be done at any particular point, and never any detail that didn't actually come up)
4) take notes of the relevant things made up in the course of play so that players can revisit locations or NPCs and find them to be coherent instead of ever-changing because no one remembers how they were last time.

It is even easier when you have a campaign setting already laid out for you to draw from, rather than having to make up the entire world whole-cloth for every campaign (which I recently quit doing after more than a decade of inventing a world unique to each and every campaign, and have found myself very happy with the results - and also with something fun to do in my free time, reading up on my chosen setting to have more details commited to memory).
 

I usually have at least 2 or 3 different adventures hooks prepped, and a single "travel" adventure that's generic enough to fit into whatever place they are travelling to. No matter what they choose, I can use the "travel" adventure for that session, and follow it up with a fully prepped adventure on the next session (now that I know what they are doing).

I also have a few off-the-wall adventure hooks to introduce as a stalling tactic in the event that players go completely opposite and/or ignore the plot. That way I can roll with it for the session and at least have a better idea of what they are doing for the next one.

Like always, the trick is giving the impression that no matter what they do, it seems like I've planned for it.
 

I create a general world map with three main "starting areas". Usually the outskirts or borderlands of some major kingdoms in the world. Basically: a place where poverty and unrest is going to lead to poor people deciding its better to venture into the woods and kill things than it is to farm crops. The are at least two major campaign-level plots taking place in each of these areas generally, there are always a variety of going-ons in every town. These plots move and resolve or worsen in the background. Player intervention can change them dramatically, but lack of involvement won't stop them either. Generally speaking the plots will try to involve the party in order to get them interested.

Beyond that the world is much like our world was thousands of years ago: mostly empty. There aren't roving bands of orcs or ancient ruins all over the place. You could go days with nothing interesting happened at all. The interesting areas of the world are mostly known about either commonly or at least through legend. That's where you're going to find stuff. That doesn't mean there's nothing out in the world. Just mostly nothing.
 


Sandboxing is fun. I'm in league with a few other posters on this thread - AaronofBarbaria and Discosoc in particular.

Basically, I drop three hooks, give the players (not pcs) a very rough idea of the adventure, and then let the PCs choose where and what they want to do.

I keep my setting small, so that PC actions can have a ripple effect on the world. I make sure most NPCs have a method for travelling, so I can re-use the ones that "click". And I drop references to previous adventures into the current adventures to allow for greater continuity.

PCs join organizations - and I have those organizations ask for favours relating to whatever adventure they pick, so that there's a metaplot that begins to form that is shaped by player choices. It's maybe a bit sandboxy (whatever adventure you pick, you know your guild contact is going to ask for you to do something hard). But I don't require players to make certain choices (you don't have to LISTEN to your guild contact - whichever way you choose, there will be repercussions).

I plan ahead, and foreshadow adventure hooks that may show up in the future. I particularly highlight adventures I'd personally like to run as a GM by making those hooks really ring out. I've done this for a bunch of dungeon modules that will be coming up - particularly "Chadranthar's Bane" and "Vesicant", which I'm pretty sure will get picked up on by the PCs after they leave the Forbidden City.

I don't rely on tables unless I've got the luxury of having one in front of me. Usually, when players ask a question on choices of action, I just stop and ask myself a few questions:

1. What would fit with the world?
2. What is a good result for the PC's roll or ability that is being used?
3. What would be open to new ideas and isn't closed off?
4. What is FUN?

And go from there. Players will have incomplete information, will start to speculate, and their speculations will inform me on what they're worried about and (usually) problems I haven't seem coming - it used to be called "Chumming the dungeon". While they're planning, I'm throwing together mini-encounters I can run while I try to figure out which way the adventure is coming.

All of this, by the way, requires world knowledge. Because of this, I keep my world small. There are nine deities, and that is IT. I know them inside and out. Every time I need a deity, it's tied to one of the nine. There are only a few nations, and they're small and relatively weak - I can use them when needed, and know who they are and how they run (and their goals). There are few power sources, nine organizations that can really move the PCs around, and a few metaplots that I can draw upon when I need to.

The world is relatively loosely defined, though, so if I need a magical forest, ruined lighthouse, or haunted swamp, I can place one on the map with little trouble. And because it seems that I always pick adventures with cults, it's lead to a setting that's rife with the blighters. Which is kind of fun, too.
 

There was an old White Dwarf article that discussed "Scenario Design" vs "Environment Design". Sandboxing involves "Environment Design" - you create an environment for PCs to explore, not scenarios for them to play. This is why my Karameikos game is not a true sandbox - the PCs are presented with adventures & adventure options each week. Whereas in my Wilderlands sandbox there is the environment, in which exist such locations such as the Halls of Tizun Thane and the Caverns of Thracia, but the campaign is not 'adventure driven' - NPC quest givers don't, as a rule, come over to the PCs and ask them to do X. Rather the PCs roam the world encountering interesting stuff, and there are dynamic NPCs doing likewise. A lot of my sandbox effort goes into detailing tons of motivated NPCs whose behaviour interacts with the PCs and with each other. The world is a web of interaction; more detailed in the immediate campaign area, but thanks to the Wilderlands of High Fantasy Box I can look anywhere in the world, see who's there, and have an idea what they're up to.
 

Remove ads

Top