As I’ve been putting together a new link roundup for the last couple of weeks, I read with great interest Hack & Slash’s new series on his player agency theories. He deals with what he calls “the Demon of Design” in three posts: On the Virtual Matrix of the Demon of Design, On Binding the Design Demon, and On Banishing the Design Demon and Having a Cup of Tea. Most of the meat is in the second post, but the TL;DR is that he believes all design should be predicated on giving your players interesting choices, not on creating a realistic or verisimilitudinous (whee!) gameworld.

He in turn links back to two posts by Ben Robbins (of Westmarches fame) from 2008: Bad Trap Syndrome and Curing the Bad Trap Blues. They’re worth reading in their entirety (like everything Ben Robbins writes) but the takeaway is that the venerable arrow trap is a bummer, what he calls “a damage tax for walking down the hallway.” There’s no player choice so there’s no real fun. He then describes two alternate solutions: one, create a complex mechanical language for traps that lets rogue skills aid the party but not instantly overcome the trap, or two, eliminate hidden traps entirely.

It’s hard not to read option one as something of a dig at the upcoming 4th Edition of D&D, as the post was written in March 2008, but I can’t remember if WotC had released any mechanical info about the new system at that point (I know their worldbuilding preview books were already out). Either way it pretty accurately describes how 4e handles traps, essentially using its Skill Challenge mechanism to make most traps multi-part hazards with various levels of success.

But this is clearly not the path Robbins prefers. He prefers what Hack & Slash describes when he says “all hallway traps should be puzzles to solve”– basically, do away with the hidden surprise trap entirely and move the action from dice-rolling (“Make a Reflex save” or “Does a 17 hit your AC?” or whatever) to player involvement. How? By making every trap easy to spot. He writes:

Yep, no surprise traps. You might not know exactly _what_ the trap is, but it is always pretty clear there is something dangerous. It might be the remnants of past victims (a litter of half-melted bones scattered in front of one unusual door…) or some particularly suspicious detail (why is there an open spiked pit at the bottom of that staircase?).

…

The interesting part is that removing the surprise and basically announcing there’s a trap (for anyone who’s paying attention) completely changes the dynamic of play. Instead of being a hit point tax for walking down the hall, it becomes, well, a game. The players huddle, they have their characters look around, they brainstorm possible dangers and ways to get around them.

Interesting stuff. Seems like madness, right? But I think it may actually be the solution I’ve been looking for.

Confession time: I’m not great with traps. It’s a challenge area for me. I love them with all my heart but I don’t run them well. Shameful example: in my 4e campaign I designed a cool dungeon dedicated (or so it seemed) to the snake god. It was in the shape of two intertwining snakes, and there were secret rooms in the spaces between the coils. The players entered armed with a mysterious line found on a scrap of parchment: “The Heart of the Snake is in the heart of the snakes.”

All in all a successful mini-dungeon– they found the Heart of the Snake (a giant, sickly-green jewel) in the central secret chamber between the two twining snake-shapes. There was just one problem: the arrow traps.

The party marched boldly down a hallway. The monk, in front, got hit with an arrow and took some damage. And from then on out, he and the ranger rolled spot checks for every five-foot step they took. Again and again with the dice, to the point that another player, growing understandably bored, grabbed their minis and dropped them at the end of a long hallway and grumpily said “You walk down the hallway and roll a bunch of spot checks” or something like that. I was not in control of the table any more because frankly, it sucked. I hadn’t given the players anything interesting to do.

Eventually I only made them roll when there actually was a trap, to see if they spotted it or not, because frankly, they were going down these hallways one way or another, so the only question that needed to be resolved was whether they stumbled blindly into each arrow trap. This was the best solution I could come up with on the fly, because by that point I was red-faced with shame at my poor dungeon design. It stunk.

Now let’s reimagine this scenario using Robbins’s concept of no hidden traps. The party is moving down a lengthy hallway when they notice some dark spots on the wall ahead of them to their left. A closer examination (leaning carefully forward from a few paces back, of course, because in this fantasy I have cautious players) reveals that there are slight pits in the wall, as though it’s been struck by something repeatedly in the same few places. Maybe there’s even a smudge of blood.

In a new-school game (one that uses Spot checks, or Find Traps if you like) this is where I’d probably stop giving out free information. Clearly something is up in this stretch of hallway, and the party can probably figure out what it is. They know there’s some kind of trap here, probably an arrow trap. At this point I would let them make Spot/Find/Search/Thievery/whatever checks to find the cunningly-crafted pressure plate that if depressed, sets off the trap, then a Disarm/Thievery/whatever check to deal with it. If nobody succeeds on that roll to find the pressure plate, they still know there’s a trap. They just don’t know how it’s set up or activated. They’re welcome to poke around with ten-foot poles and so on, or even just say “Well, who knows?” and take their chances continuing down the hall, but they don’t get the easy success of making the roll.

I think this method would allow for a decent blending of old- and new-school styles; it doesn’t make the rogue’s trap skills useless, but doesn’t make them into automatic victories either. Of course, for a totally old-school experience you’d eliminate all the checks. Once they’ve noticed the pitting on the wall, it’s up to them how to proceed.

You can also combine approaches: I’ve been tending towards successful Spot checks giving them things like “You notice that one of the floor stones is slightly raised” and letting them work from there. It’s not exactly a deep secret as to what that raised stone does, but it’s never, never ”You see a trap” because then the game is just about the dice. This lets the rogue or ranger’s skills get some advantage, but not give the whole game away.

Two final thoughts about this. First, it means that traps should be varied, or at the very least clues should. If you have ten arrow traps with pitting on the walls and a raised stone pressure plate, the party will very quickly cotton to what’s up and then we’re back to “Are there little holes in the wall? Okay, I press around on the stones with my pole” or even worse, just pressing every stone as they walk down the hallway. It becomes mechanical again. Have all sorts of different traps in your dungeon, and when you reuse a trap, leave a different clue. Then switch up the clues and leave red herrings so that orc corpse full of little arrows is actually balanced right on the edge of a hidden pit of spikes, placed there by those darned evil cultists to trick do-gooders.

Second, I do believe I will be using this method going forward. I’m working on a small dungeon for a game of Midnight Magic Mutant Madness (my gonzo, slightly adult post-apocalyptic Labyrinth Lord/Mutant Future thing) at GMX in October and I’m going to have lots of traps– but none of them will be hidden. I’ll let you know how it goes.