I’m prone to over-thinking and over-complicating things. This isn’t news. Sometimes, it backfires on me worse than others.
For example, my grand plan of using wandering monster rolls to help re-stock the dungeon has fallen apart. My plan was, I would keep track of which areas the party has explored, and make a wandering monster check for each area, between sessions. If the check came up, I would roll the usual “empty/monster/trap/special” stocking roll; if not, the room would remain in play for the next round of checks. As it worked out, it was just too troublesome to keep track of the list of rooms needing rolls. Too much overhead, too much laziness, not enough return for the investment. And, in hindsight, not enough flexibility.
The way it has turned out, in actual play, the party is moving a lot slower than I had anticipated. Don’t take this the wrong way! I’m not trying to say that the PCs or the players are in any way performing poorly. Quite the opposite. I think it’s a question of dungeon design, more than anything. I feel that the first level, especially, is too tightly constrained. The party keeps passing by that same pit trap. I can’t say why, for sure, but I suspect it might simply be because it’s the only entrance they know. (They actually know of at least five, that I can think of, but a couple of those are just alternate ways on to the same highway.)
My fear is, if I mindlessly apply a simple rule for re-stocking, they’ll have to slog through a never-ending horde of reeks, rust monsters, and half-crazed goblins. Maybe they’ll grow in power and proficiency, until they can just brush off such things without breaking stride, but maybe not. More likely, it seems to me, they’ll just get bogged down.
Therefore, I’m chucking the entirely-random re-stocking idea. Rather than faithfully rolling lots and lots of dice between every session — or, more honestly, rather than coming up with excuses not to roll lots of dice — I’m going to go to an “as seems appropriate” basis. Every couple of sessions, as opportunity and motivation allows, I’ll go through the dungeon and make updates as needed. If the party doesn’t turn over some rock for a while, something might grow under it. If there’s a lot of traffic on a certain hall, it’s unlikely that anyone will set up camp in the middle of it. I’ll still be depending on the dice for decisions, I’ll just be doing it on a more forgiving schedule and with an eye towards plausibility, rather than unseen objective consistency.
On the other hand, my grand plan for hundreds of rumors seems to be working well. Before the last session, I put together a huge master list of 120 rumors: thirty rumors for each of four information-gathering skills. I made the task easier on myself by having several rumors all based on the same root piece of information, but it was still a steep slope to climb. It’s like stocking the dungeon, though: lots of upfront work, then a comparatively small amount of maintenance.
But was it worth it? I think so. One of the goals of this campaign is to take me, the GM, out of the story as much as possible. When I overheard “I can’t wait to see how the GM is going to get us out of this one” remarks, in a previous campaign, I knew I had drank entirely too much of the story-driven Kool-Aid. This game is meant to be a sandbox. The hand of the GM shouldn’t be apparent. If I’m caught for a rumor, if I have to make up something on the spot, there’s a temptation to influence the story, to put my hands directly on the plot. I want to set up systems to avoid that. To keep me at arm’s length, you might say.
In my story-driven games, if the party heard about something twice, they could bet it was going to impact their lives before too long. The rumors were foreshadowing. I chose them for their influence on the story. With the Big Book O’ Rumors, it’s the other way around. If the party hears about something twice, it’s because the dice handed it to them that way. If they follow up on those rumors, they turn into foreshadowing. If the party decides to ignore them, they’re just background noise that improves the simulation of reality. The story comes out of the players’ choices and the PCs’ actions, rather than the action coming from the GM-driven story.
Finally, while I was stocking some of the lower levels, I came to the realization that I had entirely too many numbered areas. For example, I had one area about the size of my living room, surrounded by many smaller spaces, perhaps the size of a walk-in closet. In that one tiny area, for some reason, I had numbered all the individual rooms. By the averages, the “empty/monster/trap/special” rolls would have populated a couple of monsters, and a couple of traps, in a space about the size of a decent apartment. It’s a Dungeon Fantasy sit-com. Visions of the otyugh (literally) hogging the bathroom. The troll and the umber hulk arguing over the bill for the crystal ball.
Really, the whole little knot of rooms should be considered a single location. By crowding my numbered locations, I’ve lost the necessary buffer of empty space between spots of interest. The levels on Diablo 3 are pretty tightly packed, but even so, you find some quiet space between encounters. So, I need to go back and do some erasing. Thankfully, these are all areas far from anywhere the party is likely to reach anytime soon.