Ask Angry: Unsucking Mazes

Banner Patreon 800 x 100

Category Ask Angry 800 x 450

Do you have a question for The Angry GM! Well, because people can’t get it right, I had to set up a separate page with instructions. About how to send a f$&%ing e-mail. Ready How to Ask Angry to learn how to submit a question.

Logan (@rpglogan from Twitter) asks:

You mentioned before that mazes and labyrinths in a tabletop RPG don’t really work out well. Is there a good way to implement these in your game? How would an Angry maze work if you made one?

On the one hand, I kind of want to beat you unconscious with a copy of Sudoku for Dummies. There’s always someone who does this s$&%. You say “here’s a thing that just won’t work in a table-top RPG because of some very fundamental features of RPGs and that thing,” and there’s always someone who says “well, how would you do it right, then?” You’re missing the point. The point is the thing doesn’t fit into the RPG space. It doesn’t. People try. It doesn’t.

On the other hand, innovation is born of people refusing to accept the impossible and trying to find ways to do things in spaces that don’t really accommodate those things. So, from that respect, this is a good question to ask. “How can I do the impossible” is a nice starting point to build something brilliant. But, the thing is, you’re asking me to do the hard work instead of sitting there and doing it yourself. So, honestly, what you’ve really done is put me on the spot and demanded I invent the impossible.

Fine. Challenge f$&%ing accepting.

But the thing you have to accept, going into this, is that what comes out at the other end is probably not going to resemble the thing that went into it. That is to say, whatever we come up probably won’t be a maze at all in the way you understand it. But it will capture the feel. Or rather, it will capture a similar feel, inspired by mazes.

So, what is a maze. A maze is a collection of branching paths through which the puzzler must plot a route to get from the start point to the goal. Usually, there is only one route that actually leads to the goal. The rest lead to dead ends or double back on themselves. And, in general, there are two varieties of mazes. There are the ones on paper and there are the ones you wander through at Halloween: corn mazes, hedge mazes, and the like. Paper mazes and real life mazes.

When people think of a maze puzzle, they typically think of a paper maze. You have a picture of the maze, basically a map, and you have to find the route through the maze. So, you start tracing paths. Usually, you start with your eyes. And you just start trying a route until you hit a dead end, and then you back up to the last decision point and try again. And again. And again. Until eventually, by brute force, you’ve found the end. A more efficient strategy is to start from the goal and work backwards through the maze. And in order to keep track of your path and not forget which paths you tried, it is often useful to follow a process like only making right turns or whatever.

The thing is, though, that solving a maze isn’t really solving a maze at all. A maze is not something you figure out. There’s no real logical conundrum. It’s just a matter of trying various paths until you find the right one. And solving the maze has no real tension to it. If you’re patient, provided you follow a good heuristic, you’re GOING to solve the paper maze. Now, I am not taking away anything from people who love mazes and it is actually good exercise for your brain for a variety of reasons, but in terms of challenge and engagement, there really isn’t much to a paper maze.

Solving a real life maze is only slightly more difficult. It takes away the most efficient strategy (start from the end) and it relies more on your spatial memory to remember what paths you’ve been on. But, in the end, the maze is still defeated by a brute force, trial and error approach and greatly aided by tricks like always making right hand turns or always keeping your right hand on the wall as you go.

I should point out, by the way, that people have said that the way to SOLVE a maze is the “right turn/right hand” approach. That actually isn’t true. That approach eliminates the chances of you retracing your steps and forgetting what paths you have already explored and eliminated. But it isn’t very efficient. That is to say, because it COULD conceivably require you walk through absolutely every square foot of the maze before you reach the exit. In point of fact, you could PURPOSELY design a maze that using the right hand/right wall solution would take the LONGEST. And you can design mazes with closed loops that utterly CAN’T be solved by following a heuristic. But none of those facts change the essential point: solving a maze is a matter of patience and trial and error.

And THAT brings us to why mazes in RPGs suck. First of all, there’s no real tension to it. By itself, a maze is just a matter of patience. It’s a foregone conclusion that the PCs WILL solve it. It’s only a matter of time. BY ITSELF, that’s no different from any other obstacle that the PCs can’t possibly fail to overcome. And it just isn’t very interesting. Second of all, a maze is full of false choices. Like, chock full. “You come to an intersection, left or right?” “You come to another intersection, left, right, or straight?” “You come to another intersection. Left or straight?” And that is how the maze plays out. FOR F$&%ING HOURS. And the thing is, those choices aren’t really choices. There’s nothing the players can do to discern which choices are good and which are bad. There’s nothing meaningful in the choices. It’s just a random choice. And, generally, the players will follow a heuristic, so there isn’t really a choice at all. They just ALWAYS turn right.

On top of that, we have the issue of mapping. Mapping is interesting because it exists in a sort of meta space. Back in the olden days, players ALWAYS mapped dungeons as they explored. And that was smart because they could easily find their way out. Nowadays, some groups map and some groups don’t and the GM rarely forces the issue. Once upon a time, D&D actually TOLD players AND GMs that it was important to have someone map the dungeon or wilderness as the party explored.

So, why do I say it exists in a sort of metaspace? Because rarely is any thought given to the CHARACTER in the world mapping the dungeon. Some GMs will allow the party to map even if the PCs have no way on hand to make a map. The GM assumes the physical map represents the character’s memories of the space and leaves it at that. The trouble with that approach is that it means that the PCs have PERFECT recall. And that f$&%s with any challenge that is based on navigation. One of the challenges of navigation, exploration, and spatial layout is that some human brains are better than others at building a mental map. A MAP should require resources from the characters because it overcomes a limitation of BOTH the characters AND the players. A map is a solution to a problem. It should carry a cost.

The cost of a map is pen and paper, sure. But it’s also in time. Every time the PCs have a break, someone has to break out the map and update it. And that isn’t instantaneous. It means that the party is burning five to ten minutes of uncorking ink, unfurling map, updating map, drying ink, capping ink, furling map, stowing map. That’s notwithstanding the time it takes to actually figure out what to draw and write. And some players draw f$&%ing survey maps.

Normally, the metamap problem is not a problem at all. Who cares, right? But when you have a challenge or obstacle based entirely on navigation, the metamap can become an “I win” button. So, if you have any interest in navigation as an obstacle, you’ve got to enforce the idea that a real life map drawn by the players requires an imaginary map drawn by the characters. Otherwise, the players better pay careful attention and remember what the hell is going on.

Of course, there are inbetween solutions like chalk or soap marks to tell you where you’ve been. Dungeon hobo signs, I call them.

So, how can you make a maze interesting? Well, the problem is you can’t make the maze itself interesting. A real life maze is only BARELY interesting. In a fantasy RPG, it just isn’t at all. What you’ve got to do is find ways to add tension to the maze. Or meaningful choices. Or uncertainty to the solution.

Uncertainty to the solution comes down forcing the PCs to solve an actual puzzle to find the exit. What if there is no way to find the find the exit of the maze just by wandering. What if there’s a series of switches that have to be located and pulled in the right order to open the door. Or keys hidden in the maze. A scavenger hunt of sorts. Now, that sort of thing can be pretty dull in its own right. It’s padding out the maze with a scavenger hunt which itself is uninteresting. But if it ends with an actual puzzle (like, which keys get turned in which order), it’s at least a puzzle. Not the best solution. Not by itself.

Tension arises when there is an element of danger to remaining in the maze. This can be accomplished by adding a powerful beast to the maze that is hunting the party and knows the maze and uses hit and run tactics to wear down the party because it knows it can’t beat the party in a straight up fight. Minotaurs, ghoul catcombs, stuff like that. That’s good stuff. It’s especially good if the creature has some way to cheat. Minotaurs know the maze implicitly and can head the party off. Spiders can climb over the walls. Ethereal marauders can pass through the walls. A time limit is also a good choice. What if the maze is slowly flooding. Or the bomb is going to go off. Or the antidote is hidden in the middle of the maze. Or there’s another party also wandering through the maze looking for the artifact and whoever gets to it first wins.

You can combine obfuscated solutions and tension elements. The minotaur is hunting the party who needs to find three keys to escape the maze. That suddenly makes the maze interesting.

As for meaningful choices, there’s a couple of ways you can do that. The first is scatter hints and clues at the intersections that the party can recognize to help them find a way faster. Dwarf stonecunners can keep track of directions and recognize passages that slope upward toward the surface. Elves and alert humans might notice the scent of fresh and stale air. Trackers might notice the trail of creatures passing. If the party thinks to take some time to really examine their surroundings, they can get clues to help them find their way through more efficiently. If they don’t, they are left wandering. You can also use those meaningful choices to allow the party to avoid dangerous paths. A gelatinous cube recently passing leaves a pretty telltale sign. Maybe the party doesn’t want to go in that direction.

Finally, if you want to actually play with the mapping assumption – and I truly encourage you to do so – there’s a couple of ways to use mapping to add various elements to the maze. For example, the classic mapping puzzle is the missing room: a void in an otherwise full map that very clearly SHOULD have a room in it is generally an indication that there’s a secret door. By mapping, the party exposes the likely locations of secret rooms that contain treasures or even contain the exit. It’s something they might not recognize if they didn’t map. A maze filled with teleporters in otherwise identical looking hallways is a challenge that can generally only be solved by mapping as the party recognizes that the space they’ve been teleported to is identical to an area they’ve already been in. It’s especially fun if the teleporters are invisible.

And if you do want to mess with mapping, one thing you can do is the maze with a time limit. Make the party decide whether mapping is worth it or not. Do you really want to stop for five minutes at every intersection while an aquatic minotaur is hunting you through a slowly flooding dungeon? It might be the only way to find the hidden room with the exit.

In the end though, all of these possibilities reveal the secret truth about mazes in D&D: mazes aren’t puzzles or obstacles. They are terrain. They are settings. Essentially, EVERY dungeon is a maze (unless it is completely linear). So, when you ask whether a maze can be an interesting puzzle by itself, what you’re really asking is whether a dungeon can be interesting without anything in it. And the answer is no. No it can’t. This isn’t Farmer Bob’s Spooky Halloween Corn Maze, after all. And, hell, even Farmer Bob eventually gets smart enough to have some scary people running around the maze in costume to scare the crap out of you.

Banner Patreon 800 x 100

8 thoughts on “Ask Angry: Unsucking Mazes

  1. I think another approach is to have the not-actually-a-maze. Capture the feel of a maze (winding corridors, branching paths) but have it actually be quasi-linear. Sort of how a branching story captures some of the feel of a sandbox while providing more direction. The presence of a monster works well for this type of maze as well because it provides the chance and the possibility of going down a wrong path in order to flee, as does the presence of keys/gates.

    I also like how Dungeon Crawl handles mazes- the outermost region is brown rock, the second region is grey stone, and the middle is bluish metal, so you have a stronger sense of being on the right track.

  2. One of my best “mazes” was actually just two corridors, an intersection, and four three exits at the end of each corridor. I basically used the teleporting idea, but it was a Guards and Wards spell: all passages were foggy, and every time they chose an exit or corridor, there was a 50% chance they would go a different direction than intended.

    After they realized they kept coming back to the same 2 rooms, they got really engaged and came up with cool ideas to beat the “maze”.

    Of course, the fact that they were escaping from a magical prison was great to keep the tension up. Guard’s footsteps are scary in foggy corridors…

  3. I’ve honestly never run or encountered anything more maze-like than getting caught in the Maze Spell in d&d. And honestly they do seem sort of dreadful… I think if I made one, it would be small, like it would take less then half a day to get to the centre of it or whatever (the goal). Give it multiple entrances/exits that only open at certain times of the day (hinted by carvings on the doors), and make it incredibly dangerous at night. The PCs would have to spend a few days learning which doors open when, learn where each door leads relative to the complex, and could read hints (inscriptions) on the walls about the horrors that infest the complex at night. Doing this I think you could get a great week or so of adventure as the party slowly learned the secrets of the place while trying to get back outside every night before the last door closed… And you could even have one scene where they actually had to stay down there overnight to find something! Eventually they could find what was at the middle and move on… Uh oh I think I just wrote Dungeon-Quest the Boardgame as a d&d adventure!

  4. An other way of handling it is to mix Angry’s Abstract Dungeoneering with Navigation skill check to see if they manage to find the exit or if they simply fool around and around.

    That way you don’t have to actually handle the maze. You only handle it’s content and the feel of navigating (by setting a difficulty to avoid getting lost or finding something specific).

    Crit fail or normal failure could manage the random encounter side of the game too if the question you try to solve by the roll is : “will the characters be able to SAFELY NAVIGATE the maze ?”

    • That’s what I was thinking, use a intelligence or navigation roll to figure out how long it takes to get through the maze, or use a random roll to determine which room, encounter, or event the party finds. Maybe clever ideas for navigating or taking the time to make a map will modify the roll. What is the dramatic question? What choices are there for the players to make? And, what sources of conflict are there? Minotaurs, ghouls, floods, etc.


  5. Craig Payne of at Exploring Infinity managed to game-ify labyrinths a few years back with a deck of playing cards.

    For folks that can’t string together a google search from that yet, there is no map, you deal out a few cards face down then have the party select one and flip it over to find out where they end up. There are a bunch of patterns in the cards and with a few skill checks (like recognizing some prior party’s dungeon hobo signs,) or trying out some of their own tricks for handling the maze you’d flip over some portion of the cards in advance so that they can start to avoid cards that they think are bad or head straight for cards they think will be good. Dead ends sit around and decrease the total options until you’ve got too many and reshuffle the deck, some of the face cards are more important locations that players can always find their way back to once they’ve been there once, and the entrance and exit/goal cards are the jokers, so they don’t go into the deck until you’ve shuffled once.

    I’ve been thinking about getting some kitschy tarot card deck for something similar.

  6. I love that card idea, and having them figure out which cards they need to shoot for could be a really neat wrinkle, if cards begin getting dealt face up and they have to figure out which suits and values represent what kind of room contents to get closer to their goal.

    Personally, to do a maze or a labyrinth myself, I would ignore the actual solving of the maze. I would give the party some sort of time constraint (like flooding, or a rival evil adventurer team), and then make them decide whether to go carefully or fast, whether they mark the walls, If they’re making a map, to make them try to decide how much they’re willing to sacrifice to try to get there fast. Will they show up first and take the idol uncontested but with a level of exhaustion, or will they end up having to fight for it? Will they have taken the time to make a map so they can retreat via the most direct path, or will they have to follow their soap and chalk marks and be stuck spending longer to leave the dungeon? Will they have paid enough attention that they can throw up a wall of stone to fool their pursuers into going down a wrong turn to run into that nest of spiders they heard on the way through, or will they forget their path and blunder into it themselves and take pot shots from the warlock while spiders web and poison them?

Leave a Reply