Some conversations today on reddit got me thinking about secret doors in dungeons - or, more specifically, about the mechanical procedures used to determine whether the PCs find and open such doors. In theory, at least, secret doors are one of the cooler elements of an adventure setting. In practice, however, I've often found their mechanical execution either disappointing or unwieldy. On the one hand, having a predetermined X-in-6 chance of detection is simple to adjudicate, but it seems a bit lame to me - much of the time, the door simply won't be discovered, and whatever cool prep lies behind it goes unseen. A different approach could be to make the existence of doors more readily apparent, but present them to players as puzzles. Or, still alternately, one could just richly describe the layout and dressing of a dungeon room, sprinkling detail-seeds that astute or curious players can respond to, hopefully manipulating their way into secret rooms. THAT, however, requires a lot of work to do well consistently, made much worse by the fact that even published modules (in my experience) don't bother to explain how one gets into a secret room, just the fact that it's there.
Could methods for traps offer useful ideas? For traps, I really admire the Chris M Into the Odd approach - finding the trap is easy, but figuring out how to handle it is the challenge. Also, in the afore-mentioned reddit discussions, some folks mentioned the Angry GM "Click" approach to traps. If a trap does get set off, say "click" and briefly describe what a PC sees, then ask how they deal with it. They could (for example) brace themselves, jump, duck, leap to the side, etc. For each kind of trap, there's an optimal and a least-optimal way to deal with it, so the player has to make a choice that could have consequences.
Hmmm, I thought...could there be a way to do something similar to handle secret doors effectively? I scratched my head and came up with the little system below, and revised it a little bit across the day in light of the reddit conversation. I doubt very much that no-one has ever come up with something similar - this seems a pretty straight-forward approach. I haven't tested it at the table yet, but I'd like to. I think it sacrifices certain things, but gains even more on the flip side.
A METHOD FOR HANDLING SECRET DOORS
Key concepts:
+ recognizing that the door exists is the easy part; figuring out how it opens is the catch
+ given unlimited time and safety, PCs will eventually get through the door
+ it's a dungeon: PCs don't have unlimited time and safety
So, the party enters a room with a secret door. What happens next?
Except for specific, planned edge-cases, the PCs automatically detect the secret door's existence (maybe they feel a suspicious draft; they notice a recess or bulge in the masonry; they spot varied discoloration in plaster or mortar; they see little holes leading into a hollow space beyond; etc.).
Although the players always detect the door, they must choose whether to try to open it, and how long to spend on the attempt. Secret doors will open using one of (let's say, for now) five different opening methods, detailed below. A party may spend one turn attempting to use one of these methods to open a door; for each attempt, they must tell the GM which of the five methods they are checking for. However, the mythic dungeon will not give up its secrets lightly; every attempt to open a secret door triggers a random encounter check, with an encounter in 1 on 1d6 (for epic mode, make it 2 in 6). Alternately, the players can say "forget this, let's use brute force and break down the door" - but that either automatically triggers a random encounter, or triggers a risky 4-in-6 random encounter roll.
Some characters (like elves in B/X D&D and its clones) have a special bonus when looking for secret doors. They still get this bonus, but instead of detection, a successful roll means they immediately notice the correct way to open the door. However, this still triggers a random encounter check. So B/X elves, for example, are still useful for this niche, but they speed the party up and reduce risk rather than decreasing the chance of missing a secret room entirely.
What are the five (suggested) methods for passing a secret door? Here's my working list, though I imagine these could be refined.
1 DIRECT PRESSURE applied to the door will open it - perhaps it slides back on hinges, glides sideways on rails, or can be leveraged up from the floor.
2 MANIPULATION WITH TOOLS is necessary to open the door. Maybe inserting a narrow dagger-blade into a crack in the mortar will depress a lever or button that opens the door. Maybe a needle poked through a small aperture triggers the catch.
3 A SPECIFIC KEY is needed. Some object, somewhere else in this dungeon, must be inserted or applied in order to open the door. This requires somewhat special handling. If the players choose to test for this method, adjudicate as follows. If the players are already in possession of the needed key, they open the door (tell them why). If they have encountered but did not take the required key, tell them that it looks like they need to go back and get it (triggering a decision to spend more time and random encounter rolls to get into this room). If they have not yet encountered the needed key, tell the players that they realize what kind of key is needed, and that they should be on the lookout for an object of the specific description.
4 AN ENVIRONMENTAL EFFECT must be triggered. Remember how breathing on the little key was necessary in The Fifth Element? The door's trigger must be warmed up, cooled down, licked, wetted, dried, exposed to light or dark, etc.
5 INTERACTION ELSEWHERE IN THE ROOM opens the door. The interaction might involve one of the methods above, but not applied to the door directly. Sitting on a chair, pulling a torch-sconce, yanking a cord - across the room - will open the door.
In each case, remember that with this method, if the players choose to try one of the 5 options, and they make the correct choice, then they get it; they don't have to start playing 20 questions to see whether it's heat, water, sweat, urine, light, baking soda, cheese, etc. that triggers the door...they just have to risk the random encounter check. Sounds easy, right? Well, not if it takes them 4 or 5 tries to open the door...
Now why didn't I choose 6 methods, so you could roll on a 1d6 table? Hold your horses, read on...
Each dungeon, or dungeon level - or maybe each dungeon-crafting faction - has its own preferred signature way of opening secret doors. That preferred method is twice as likely to appear, so that a specific dungeon can have 1d6 door-mechanism tables like this:
Svirfneblin dungeon doors open by:
1 Direct pressure
2 Manipulation with tools
3 Manipulation with tools
4 A specific key
5 An environmental effect
6 Interaction elsewhere in the room
PROS/CONS?
This approach certainly abandons some of the, well, secrecy of secret doors, but it makes some positive changes too:
+ entering secret rooms no longer involves a binary find/fail, but rather is about the resource-management game. Given enough time, the party WILL get into a hidden room.
+ yet time is precious, and paying the enforced encounter check for each attempt to enter makes every attempt to enter a hidden room a calculated risk. Player agency becomes an important part of the process.
+ this may save harried GMs a fair bit of stress and prep time, as they just don't have to come up with amazing dressing details for every hidden door - but they don't have to just say "uh, um, so you find a door and go through it..."
+ although it may still be optimal to provide player-facing maps that omit secrets, this system means it is no disaster if, in a pinch, a GM must share maps that show where all the secret doors are. Of course, knowing to where the doors lead offers powerful intelligence, so hiding that information is still best practice. When time is tight, however, this means a GM can get away with sharing almost any map, if need be.
+ making different types of dungeons or even different dungeon levels just slightly more likely to feature specific methods of opening secret doors lets players study patterns across those dungeons and make more informed choices in light of what they notice. This boosts coherent dungeon design and player agency.
I think I like this approach. It could be combined with "special" doors that don't follow these rules, of course, but for most dungeon secret rooms, I think this would solve most of my problems without actually causing any new headaches. What do you think?