Dispelling a myth - Sandbox prep

My One Page Dungeon Contest co-conspirator, ChattyDM, is a self-professed aficionado of the "story-driven" "encounter-driven" adventure type. A few days ago, we had a very funny twitter exchange:

ChattyDM: With my summer D&D season starting this week, I may try to let go of scripted homebrewed adventure and try going with just hooks and ideas..
chgowiz: @ChattyDM *gasp* Why.. that sounds like a *whispered* sandbox.

In this post, Phil goes on to say:
The thing is, I’m not interested in doing a classic exploration-based Sandbox game. This actually requires a lot of preparation as you must detail large areas of your setting in order to give your players something to do wherever they go.

That made me sit back a moment and think - did I really prep that much? I'd have to say, "no". Here's why.

1. Just in time preparation
2. One page dungeon levels
3. Write it down - play it on game day
4. Let the players flesh things out
5. Broad brushstrokes to events and plots
6. Don't overprepare

There are a lot of really well done sandbox articles in FightOn! magazine that outline sandbox creation and yes, if you follow the DMG to the letter and prepare many levels, you will have tons of prep needed - but I think that detailed maps and many hexes comes with play. The info I have on Dark Ages now is the accumulation of many months, but in general, I've followed those 6 principles above. So let me elaborate on them a bit.

1. Just in time preparation

A sandbox is inherently player driven. If we go back to D&D's wargaming roots, preparing a sandbox is much like setting up the terrain and starting points of the pieces in a miniatures battle. You have the troops prepared, but you don't know what's going to happen until play begins. I treat a sandbox the same way. I set up the starting points of the dungeons, monsters, set pieces and villians. I may have a note or two about who they are and what's going on. I know the basic rough few sentence outline of what the campaign setting is, but beyond that, I develop it in play.

When I started my Dark Ages sandbox, I had a general idea of the layout of the lands about 5 days travel out. I had three dungeons. I had one town. I had one set piece. Here's the cool part - one of those dungeons is STILL virgin territory for my players (who are now probably scrabbling through notes trying to figure out where it is... ). I had only level 1 done for the dungeons. I didn't even have a map of the town, and only a few NPCs fleshed out.

It worked. The players didn't need a lot of prep to jump and get started. In fact, the first game, my players went to do something I hadn't even thought they'd do! It didn't matter - with the rough ideas of what was where (broad strokes, see that below) and a random encounter, we had a blast and the session was a success.

As the players have moved into places, I've waited to prepare before they got there. When they started exploring the kobold mines in earnest, that's when I started planning/creating the second level. When the players wanted to know more about the town so they could live there, I finally drew a map. As the players meet NPCs and do things, I take notes and factor them into my game as we go along.

That's not to say all details are improvised. As the players move to do things, I try to stay about 1 or 2 steps ahead of them. If my players do something completely unexpected, I can take 5 minutes, toss some notes together, look at my notes from previous games and go with what I have. My 3 ring binder has plenty of loose leaf.

2. One page dungeon levels

The philosophy behind one page dungeon keys is that the DM adds the flavor and plots and "whys and wherefores" at the table. That neatly fits in with "just in time preparation" because now I can just draw a map, populate it using AD&D or OD&D guidelines and boom, done. When the players get to that point, I can look at what's happened, my broad brush strokes, and go from there.

For instance, my kobold mines. I had cavemen on one side, kobolds on the other. I had no real idea how it would work. I didn't plot things out. However, it turned out the kobolds and cavemen were fighting over the cave. The kobolds were scared of level 2 because of the ancient Dwarf ruins on level 2. One of the cavemen fell down a hole and died and the cavemen buried the hole because evil came out of it. The kobolds made some attempts to recapture the mines but failed. The cavemen fled the caves. This all came out as game-day decisions and play and reactions what the players did. I didn't pre-plot this because I didn't need to. I had broad brush strokes and let the game spin it out.

3. Write it down - play it out on game day

That being said, I write down all my crazy ideas and meanderings and my notes. Again, I don't try to detail things out so much until it has to be detailed out, usually in reaction to what the players do. I think that what has happened is the preparation is substituted for reaction notes and reminders and post-game-details. This reinforces the "players drive it" attitude.

Example - the players found a book. They read it and wanted to know what it mean. I told them, based on vague ideas, but I noted on my notes what I said - now there's an entry for this book in my notes so that if the players come back to it, I know what it is.

I also send myself emails or write things in my wiki when they occur to me. It's a bit like mind-mapping, but without all the buzz words.

4. Let the players flesh things out

When I first created Dark Ages, I had no clue what I wanted to do with demihumans, save minimize their involvement in the world - for now. I had no idea on what kind of history I wanted for them. So I let the players tell me. I came up with broad guidelines. I know what won't work for player ideas - and I let them run with it.

I don't want to give away any spoilers, but the littleling (halfling) backstory that one of my players came up with rocks. It's full of puns, but it's really cool. The elf history and story is continuing to be written, but I riff based on what my players give me.

My deities - all player decided, save for the Church of Light - and I've adjusted that based on what the players saw and how the Church presented in game. In fact, one of the players is now a cleric of the Light and it's going to be interesting to see how she maps out how she interprets the Light. I don't expect her to give me 50 pages of notes, but I'm going to go based on how she goes, within the general idea that I have for the Light.

5. Broad brushstrokes

There's a major, huge area to my game - the city of Irecia. Players are DYING to go there, but they know it's seriously in the whilly-whacks in terms of probably being very dangerous. Do I have the city mapped out and populated? Nope. However, I do know this - Irecia was the "diamond of the East" and a beautiful city for learning. I know it was a fairly civilized place. I know there are probably sewers. I know some of the bad mojo that has happened. That's all I need, though, because that sets the stage for "just in time preparation". I also am not locked into something - if the players do something or something changes in the game, I can take that into account going forward.

My broad brushstrokes also apply to events and plots. Yes, I do have a "ticking clock" of sorts in the campaign, but it's not a detailed flowchart - it's broad strokes that I fill in when needed. Although the players drive the plot, the world does continue to evolve and turn around them. There are some things that may happen as the players do things, and there are some things that will happen no matter what. For instance, the Doom in my Dark Ages game was heralded by meteor showers. I knew I wanted to play on that theme. When I realized that in my game calendar the end of summer was approaching, I decided to have a Summer's End Festival (another just in time creation) - and then I decided to have the shooting stars appear and one of the "sensitive" town NPCs go mad. Now the players have something to chew on - the world is doing stuff. What does it mean?

Another example - the first game. I knew that goblins were in the Darkwoods and they had patrols and various camps. I knew they were sparring with the kobolds. That's it. So the players decided to check out the goblins based on some rumors. Uhh... whut? I was so shocked, but I knew they'd run into a goblin patrol - and they did, and it was a good game. I didn't have the encounter planned out to where exactly the left foot would slide on the fourth goblin to the left in the fifth segment of the second round. Nope... buncha goblins who'd rather be camping than marching run into humans that they haven't seen in 50 years. Battle ensues.

6. Don't over-prepare

I think this is a discipline thing. I mean, you look at Gygax's words himself from 1974:
"First, the referee must draw out a minimum of half a dozen maps of the levels of his “underworld”, people them with monsters of various horrid aspect, distribute treasure accordingly, and note the location of the latter two on keys, each corresponding to the appropriate level."

I know that would make me think I had to know the details down to how many pimples are on my troll-mage's arse. (Note, I have no idea. A lot? My players want to know... as they mount his head on a spike... and that day will come...) I've since learned that while it helps to know the areas surrounding, I don't need to have everything detailed out. Only when I need it.

That being said, you do have some prep. You have to know the area around the "home base". You have to have some places for the players to go. You have to have an idea of what else is out there. I prepared a couple of random encounter charts for the Darkwoods and Dalewoods. That helps me to fill in if needs be (like the Damned attacking Enonia...)

Oh and hooks? I came up with a dozen or so rumors, based on the locations and my broad brush strokes. The players started exploring on their own. As they do things, and the world changes, I provide new rumors, or lay out new hooks (like the battle between cavemen and kobolds, or the sacking of the New Hope settlement, or the Damned attacking, or the strange goings on in the monastery, as a result of the players going there) based on what is going on. That's the nice thing - I can be reactive more than I have to be proactive - because the players are the actors and the drivers, not my story. The campaign is their story.

Summary -

A lot of this might seem like improv and I guess you could call it "inspired improv within a broad guideline" - I fill in the details where needed and I try not to over-reach. Otherwise, I'd never get a campaign off the ground.

You know... maybe we should do this together and prep the Ultima setting for my Ultima rules/sandbox.

[Edited to add - I'd be remiss if I didn't point you to this just-in-time hex map creation technique from the microlite20 Macropedia - a very nice resource and awesome game to boot.)