A few interesting factoids about the Blood Shield Bandits:
1. Several of the bandits worship a giant demi-goddess named Herathka. Their cult for this near-forgotten deity stems from the ancient shrine they discovered within a secret chamber hidden in the cave complex at one of their hideouts.
2. Before riding on a raid, Blood Shield bandits will pour a little alcohol on the ground in order to appease the godlings of mischief and mischance.
3. One of the 4th level fighters wears the mummified hand of his dead brother on a strap around his neck. The bandits believe it brings them good luck on their raids.
4. The bandits maintain small stashes of emergency supplies and gold scattered throughout the region. They’re marked by the symbol of a small raven that is carved into nearby tree trunks with small, coded symbols indicating the distance and direction of the stash.
5. Arik the Bold, a bandit lieutenant, has a fascination with all things arcane and magical. He particularly enoys collecting spellbooks, even though he can’t understand them at all.
6. On the night of a blue moon, the bandits burn a taper and watch for the winding sheet: On the morn they’ll ride on a great raid in the direction of the winding sheet’s bent. (The winding bent is an old folk belief. When a fragment of wax stands higher than the candle’s flame it’s known as a winding sheet. When it begins to bend under candle’s heat, the direction it bends is the winding bent.)
I like where you’re going with this; the idea of fleshing out an entire piece of setting around a wandering encounter result is really compelling. It feels very old-school and sandboxy, yet fresh at the same time.
It’s interesting to think that you could treat D&D random encounters (/adventures) as a sort of “group solitaire” activity – the DM/referee is really more of a facilitator than a creator (there’s still plenty of scope for creativity, of course). It’s possible to do this sort of thing with later editions, too — you could do it with 3.0/3.5, but I think it gets harder because there’s significantly more work in generating stats since ability scores, equipment, etc. all figure much more significantly into how they play. With experience, it’s not too hard to wing it and just fudge things, of course.
It’s probably best suited to outdoor adventures, but you can do the same thing for dungeons, too, with a random dungeon generator. Some time last year, no one in my gaming group really had anything prepped for our various campaigns, but we all wanted to play (we’ve been playing more-or-less-heavily houseruled 3.5). So we dusted off our AD&D books, all rolled up 5 or 6 first level AD&D characters (so we’d have replacements to-hand in case of death), and used the dungeon generation and random encounter tables in the back of the DMG. We mapped as we went. So for about 3 sessions we just would root around, fight, kill stuff, and get killed. We had a near-TPK when a flock of stirges came on our first level party — only one character survived, because the stirges all sated themselves on the tasty, tasty blood of the rest of the party. He quickly reached 3rd level, and ended up being the de facto party leader of a few new parties that explored further.
Some nifty things fell out of this — for example at one point we fought a group of bandits, and during the battle they checked morale and the survivors fled. Later, we encountered more bandits in their lair, and figured “aha, this is where those guys fled to!” Future bandit encounters were all considered to be with part of this bandit group.
After a few weeks, this was somewhat less amusing, probably in large part because of the clearly random nature of the dungeon itself. (When there’s a human agent responsible for the design, I think the players will assume there’s SOME method to the madness; but when everyone KNOWS it’s random, there is a limit to how much meaning your brain can impose on the thing.) So we stopped that, and got back to our regular games. It was a fun diversion, though, and refreshing to play something much more “hand-wavey” than what we usually play. (We’re fairly tactically-minded, so in general we prefer the 3.0-ish combat system, and we sort of blended 3.5 combat with AD&D initiative on the fly, but it was nice to have fairly large combats play out in minutes instead of large fractions of an hour.)
Anyway, after that digression, my point is that I think this is really the way to approach sandbox-like gaming: the referee sets some basic parameters (and depending on the system, maybe gets some time-intensive prep out of the way beforehand). Then a lot is up to the rolls of the dice, and the referee’s ability to somehow rationalize the results. I’ve been trying to stretch those DMing muscles recently, and I’m getting better at it. It helps to feed off of your players’ table talk — if you encounter something and you have no idea how to explain it, listen to what your players say about it. If they suggest something that’s plausible, go with it, or at least don’t contradict it until/unless you come up with something better.
Hey, I see you stopped writing for RPGnet about 6 years ago. Any reason why?
@skeolan: It’s kind of like making salt crystals. You put the ingredients in there and then you can just watch them grow. The Blood Shield bandits, for example, were literally just meant to be a quickie example thrown together while writing the original post. But they stuck around in my subconscious for the next couple of days and accumulated a bunch of random creativity and odd synchronicities from the stuff I was reading. (Including Flawless and The Boy’s Tale.)
@Leland: A human moderator is definitely needed in order to evoke greater depth. The great thing about procedural content generation in tabletop play is that it doesn’t need to actually generate something creative or interesting: It just needs to provide the improv seed for the GM to riff off of.
I’ve also found that there’s a delicate balancing act when it comes to how much the players know about how the “magic trick” behind the screen works. There’s a kind of iceberg-like alchemy: 10% of the experience is what the players see; 10% is the prep work they don’t see (directly anyway); and 80% is the closure they perform by consciously or subconsciously assuming a depth that isn’t really there.
(Or is it there after all? If they went poking, after all, we’d discover it together. BID.)
But if you show ’em how the magic trick works, they stop performing closure and the experience seems shallow. For example, years ago I developed a random dungeon generator customized for a sewer adventure beneath the city of Anyoc . The session went smashing… until I made the mistake of revealing it was based on a random generator. This caused the players to check out of the experience: Once they were aware that there wasn’t “really” a sewer complex for them to explore, they lost interest in the exploration.
@Drohem: That’s a long story I’ve been tempted to tell since I started the Alexandrian, but I haven’t gotten around to it yet. The short version is: RPGnet’s management changed, the place got pretty poisonous, I tried to make a principled stand against the absurd power-egos of the moderation team, and I got banned for it. I also yanked the 200+ reviews I had written for them, during the process of which I learned an important lesson: The point at which a content aggregator believes that their existence is giving value to your creative works (rather than your creative works making the aggregator valuable) is the point where you should probably stop associating with that aggregator.
A common enough tale, I was curious if your experiences there were the same as others. Do you post or write at any other forum or website, then?
“But if you show ‘em how the magic trick works, they stop performing closure and the experience seems shallow.”
Yes, that’s my impression too. A dedicated GM (as opposed to a GM/player) could use a random generator beforehand and just smooth out some of the bigger oddities and the players could project meaning onto it. In our case, we all wanted to just play, so we all bought into the random, essentially meaningless nature of it up front. Which meant that we all rolled with that meaninglessness for a while, but it also meant that we weren’t interested enough to sustain it very long.
Letting this percolate through my thoughts for another day, I think the line of rejection might have something do with associated vs. dissociated mechanics again.
For example, I’ve never had a player reject the use of a random encounter table: The random encounter table is modeling something with a direct reality in the game world (the essentially random movements of creatures).
The random generation of a dungeon, OTOH, doesn’t have that association with the game world: The dungeon was not randomly built.
OTOH, imagine a random dungeon table used while the PCs are exploring a plane of primordial chaos or exploring the dark psyche of a mad god. Suddenly the randomness of the table is associated with the game world and I think the objections would drop away.
A well-constructed building isn’t random, but a classic megadungeon combines natural caves and layers of architechtural revision by varying occupants over centuries that might be just as opaque to the explorers as an orc squad’s patrol route or a giant spider’s mating habits. All those tunnels under Gobwin Knob were originally dug for access to valuable minerals, and only later modified for defense.
What might help is developing a random dungeon generator which somehow automates or internalizes the lessons of [Xandering]. Modify rolls based on what else is nearby, or how many known routes are available back to the surface, or otherwise allow the PCs to make inferences and meaningful decisions about content you haven’t fully generated yet.