Forum adverts like this one are shown to any user who is not logged in. Join us by filling out a tiny 3 field form and you will get your own, free, dakka user account which gives a good range of benefits to you: No adverts like this in the forums anymore. Times and dates in your local timezone. Full tracking of what you have read so you can skip to your first unread post, easily see what has changed since you last logged in, and easily see what is new at a glance. Email notifications for threads you want to watch closely. Being a part of the oldest wargaming community on the net. If you are already a member then feel free to login now.

I have been spending some time thinking about this the last few days and I figure I would throw it up for discussion.

How would you run a maze in a TTRPG?

Traditionally, these suck for a few reasons. First, continually bringing players to intersections and asking them "left or right" is boring as feth and a decision that the players cannot possibly know the correct answer to so it's a choice they make blindly, which isn't interesting and mostly frustrating.
You COULD draw out an actual maze. But then the players are solving a piece of paper you put in front of them and not playing the game. You COULD lay down a card with the intersections they reach but then there is no chance for them to get turned around and lost (kind of the point of a maze? Is that good? Bad?) and ultimately just becomes trying every path until something works out. Do you just decide to treat it like wandering in the wilderness? But what checks do you make to gain your bearing? How do you pick a direction with nothing to gauge it on?

Keep in mind that I am not talking about spicing up the maze with interesting locations and encounters within the maze. I take those as a given. Instead, the mechanics of the maze itself. What is the best, most fun way to run a maze mechanically that makes it still feel like the players are in a maze without the dull blind choices of actually being in a maze?

Interested in others thoughts.

I guess the question would be, “how is a maze different to a typical dungeon?”

In RPG’s, a dungeon crawl is really just a chain of events. The PC’s might choose path A to start, or B, or C, but in the end, all roads lead to D, the end of the Dungeon.

Being underground with a roof over your head creates physical boundaries that keep the story on track. You don’t have to worry about the PC’s wandering off in any old direction. They will either move forward in the dungeon, or they will move back.

Regarding a maze, if you don’t want each intersection to be blind luck, I’d suggest one of two things.

1st. Have a trail to follow. Maybe a physical trail the Rogue / Ranger can follow. Maybe the Cleric can use their “Turning” attempt to sense the direction the Daemons went. Have a Diviner wizard? Stock up on crystal balls and have at it.

2nd. Clues, hints and Riddles. Each intersection has a hint, clue or riddle. There are several answers, each with a path. Get it right, move to the next intersection / encounter. Wrong? Trap! Or monster, or trap that shoots out monsters!

Otherwise, yeah, there’s a lot of wrong turns and one of the PC’s is going to be rolling the gak out of their cartography skill.

Unless your game has a luck stat. In which case the luckiest PC takes a luck check each intersection.

I think i would categorize a maze by the players ability to become lost coupled with some measure of misdirection.

A maze 1) does not HAVE to look the same on each corridor and whatever but sections of it should look so close to the same that while they know they are walking "forward" the chance of being turned around is possible.

2) I think landmarks probably are the hallmark of a GOOD maze but i am not entirely convinced they are needed to run a maze. By landmarks i mean distinct locations within a maze. A fountain, a statuary, etc etc... Places to find within the maze.

So running with the idea that a maze is a dungeon (and it is) you can say the maze is comprised of x number of locations with traversal sections splitting them up and the traversal sections could be dotted with random encounter tables for traps and minor combats and whatever. You can toss in a classic "minotaur" that hunts the players in the maze too.

So then what mechanics define the traversal that can be entertaining and leave open the potential for getting lost? Lets try not to make it reliant on any class abilities. A lone warrior who enters the maze should have a chance to solve the maze.





Automatically Appended Next Post:
Actually that works well. The simplest but most boring maze has 2 landmarks entrance and exit. Or entrance and center (where whatever the maze contains is hidden). And split by a single extended traversal.

So what ytaversal mechanic makes for interesting game play while allowing for getting lost.

I guess the traversal mechanic can be...

Luck: Determine a chart. Poor rolls, no advance and potentially lost. Medium rolls, gain more knowledge of maze (bonus to future roll) good roll, find a new significant location.

Include random monsters / traps for some results. If you get lost, you lose your “maze knowledge” and have to start over from scratch. In this case, your maze is (functionally) a web of known locations, and the players roll to see if they successfully find / move between different parts of the maze.

Player Character Agency: Similar to above, except skill checks get involved instead of pure luck. A variety of skills might be beneficial, from tracking, to Cartography, to knowledge of ancient people / arcane / mythology / using chalk. Again, good rolls advance to known spots in the maze, bad rolls get you lost. Random monsters and traps for some rolls.

Player - person - agency: That returns to the hints, clues and riddles. If the players can sort the clues out, they advance. If not, they get lost / come back to the same location / older location. As a GM, I find this is occasionally fun, but the Party will invariably miss the blatantly obvious clue / answer and then you have a TPK when they all starve to death... those glorious imbeciles.

Which is another thing that I think Mazes, mechanically, need. A limited resource. Such as food, torches, air... patience... unlike a standard Dungeon that you can leave and come back to, you have limited resources and that puts tension on your exploration. Getting lost could be deadly if you run out of water and can’t find the only fountain with clean water.

So I guess I just he question would be how do *you* want responsibility for advancing through the maze to fall upon? Luck? Player Character ability? Player-Person ability?

Some of each? My personal preference would be PC ability with modifiers for Player Persons figuring out clues.

I agree that player agency needs to play a major role.

Here is some ideas I had when mulling this over today at work.

Step 1) Build a flow chart of Landmarks. Connecting Landmarks that can be reached from each other with lines and putting a number on the line. The number represents a number of "Progress" needed to reach one landmark from the other.

Example with the simplest version.

[Entrance]-------7-------[Center]

2) DM discretion, the players can use tools/make intelligent decisions to help navigate the maze. Using chalk arrows to mark where they have been. Each of these is worth 1 "Progress". Each instance of this can only be use once per an attempt to navigate between 2 landmarks.

3) Create a table of events/encounters within the maze. These encounters serve a couple purposes. First, they break up the monotony of walking around and describing corridors. Essentially you treat it like traveling through wilderness. You don't need to describe every step. You just describe the interesting bits. Second, the events act as trigger points for the players to gain progress or loose it. For instance in a hallway you might have a combat encounter. At the end of the combat the players make a check to gain their bearings. Success, they remember accurately which way they came and which way they are going and gain 1 progress. Failure they get turned around and go the wrong way and loose 1 progress (if they used the chalk you can indicate this by telling them they eventually run into one of their arrows). Even better, different players pass and or fail and they have to argue about which way to go and the DM just marks progress up or down based on their final decision. All of this is kept hidden from the players btw. This is just tracked by the DM.

If Player progress ever reaches zero they find themselves either at a unintended landmark or the previous landmark or a dead end.

Obviously players won't know about landmarks when first entering. But they will start finding them and using them to kind of build a map of the maze within their heads. Resources as you say are important. You have chalk but how much? You have food and water but how much? Once in can you find the exit again? The point I think is to make the players FEEL like they are trapped in the claustrophobic corridors of the maze and occasionally circling back to places they have already been. Pit them against each other when some get turned around and others do not.


Then you can layer on other complications. What if the "Minotaur" is hunting them? What if the maze rearranges itself in a pattern? None of those are needed to run a maze but they add interesting twists.

I like it. The flow chart / location web seems ideal, with a number of successes allowing the “finding” of significant locations.

If a location is hard to find, or a longer distance, then more successes are needed.

Depending on your “realistic” requirements, you might want to consider what the monsters in the maze *do* when adventurers aren’t looting their corpses. Why do the Orcs live here? What do they eat when tasty PC corpses aren’t available? Is there some form of plant / fungus available that lowly critters can eat, and then larger predators eat them, and so-on up the chain?

That might not be relevant to your style of game, but there could be factions vying for power within the structure of the maze. Maybe a powerful wizard has a ravenous sport, in which each faction “volunteers” a member to engage in gladiatorial combat for the right to the limited food supplies the wizard dumps in.

Maybe the other factions are as much victims as the PC’s...

Or maybe it’s just a fun romp to change up a dungeon crawler. No wrong answer, of course. I was my group’s most frequent DM so thinking about the wider world was always in the back of my head, chasing out new plot threads and possibilities.

 greatbigtree wrote:
I like it. The flow chart / location web seems ideal, with a number of successes allowing the “finding” of significant locations.

If a location is hard to find, or a longer distance, then more successes are needed.

Depending on your “realistic” requirements, you might want to consider what the monsters in the maze *do* when adventurers aren’t looting their corpses. Why do the Orcs live here? What do they eat when tasty PC corpses aren’t available? Is there some form of plant / fungus available that lowly critters can eat, and then larger predators eat them, and so-on up the chain?

Agree. I feel like oozes, mimics, and undead make for the best Labyrinth ecology unless it's like the Minotaur legend and people are making sacrifices to it.

That being said, these mechanics work for any contained environment where the players can get lost and you want that to play a part. Traversing a sewer network under a city and trying to arrive at a specific location works with this. You just make the different Landmarks different sewer exits along with a few neat odd things to stick in the tunnels. A mystic wood legend of zelda style that traps the players so they cannot find their way out etc etc... The maze is only really a maze when it is in fact a maze. The core mechanics have more uses then that. I will need to run some tests on the numbers to find a optimal range of progress needed. That will of course be coupled with the tools available to the players. If certain landmarks have puzzles that can be solved to offer clues and those clues create permanent +1 Progress when trying to navigate. Progress requirements then have to be set in such a way to accommodate the collection and use of all clues. But you also don't want to set them too high or else it becomes a slog.

That might not be relevant to your style of game, but there could be factions vying for power within the structure of the maze. Maybe a powerful wizard has a ravenous sport, in which each faction “volunteers” a member to engage in gladiatorial combat for the right to the limited food supplies the wizard dumps in.

Maybe the other factions are as much victims as the PC’s...

Or maybe it’s just a fun romp to change up a dungeon crawler. No wrong answer, of course. I was my group’s most frequent DM so thinking about the wider world was always in the back of my head, chasing out new plot threads and possibilities.

Absolutely yeah. I think I am gong to try to write up a short adventure doc utilizing the mechanics and post it up so others can play test it if they see fit. I think this idea has real legs for simulating the feel of a maze without any of the slog that a maze actually is.

I think a "maze" as a puzzle for part of a session or whatever is never really going to be very satisfying. For a real "maze" experience, with the players having the chance to get lost, or meaningfully use their knowledge of the layout to their advantage, you need a really big maze dungeon that will take many, many sessions to explore fully.

I use the Barrowmaze megadungeon for a lot of my introductory games, and it has a very mazelike map as the name might suggest. It is all one one "dungeon level", and I have definitely had my players get lost in the dungeon and be really stressed trying to find an exit or the way back to somewhere they know, and mapping was super important. Once they began to use the layout to their advantage they were really excited.

And as suggested above different areas of the maze are home to different factions - a bunch of Orcus cultists, some rogue wizards, some mongrelmen, a tribe of gargoyles etc.
Here is my map:


And here is the player map of the sections they had explored:

There are discrepancies between the players map and the reality of the dungeon. I think real frustration can come from spending lots of time drawing a real map based on a description of the place that ends up flawed and potentially unusable when the descrepencies come into play.

Thats not good game play because it just happens naturally from the disconnect in description to player interpretation.

I don't have any issues with the idea of making a mega dungeon in which each "landmark" is in fact a series of connected rooms or a dungeon in and of itself with more labyrithian corridors connecting them.

But i cannot justify players attempting to draw their own map or spending the time making one yourself for something that complex. Its not good game play to keep describing cross roads or tee sections and dead ends where players have to make choices they have no way to choose correctly with.

I think the discrepancies are part of the gameplay, myself. People get turned around in the dark, underground, and the discrepancies represent that to some degree. The largest problem is the diagonal corridor that is going in the wrong direction, and the players were kinda obstinate about misinterpreting that, so I let them.

But for sure, that playstyle is not for everyone, at it does not produce it's rewards in any time frame shorter than a dozen sessions or so, which might be totally inappropriate for your game.

I also would not draw a dungeon like that myself, I prefer to purchase large dungeons. Made one myself once, and that was enough for me.

And with crossroads, where possible, it is important to give the players some kind of clue. Alternatively, that is where spells like Locate Item become really useful for the players.

The roots of D&D are much closer to the BarrowMaze, then the nodal web of locations.

That said, modern players have a different expectation than old school D&D players. What you should do is very group dependent. I know what I would do for my group.....

Yeah and the roots of dnd are 50 years old. Game design has come a long way. Nobody in their right mind thinks THAC0 actually makes for a better game play experience.

 Lance845 wrote:
Yeah and the roots of dnd are 50 years old. Game design has come a long way. Nobody in their right mind thinks THAC0 actually makes for a better game play experience.

Sure, unless THAC0 is what your players WANT.

Ain’t nobody got time for thac0.

 Easy E wrote:

 Lance845 wrote:
Yeah and the roots of dnd are 50 years old. Game design has come a long way. Nobody in their right mind thinks THAC0 actually makes for a better game play experience.

Sure, unless THAC0 is what your players WANT.

There is no argument against players wanting a bad thing. Sure. Some people actually enjoy the Transformers movies for some reason. Them liking a terrible thing doesn't make it any less terrible.

Personally, I am always more interested in making things better.

I think the node design is totally valid, I just prefer the "reality" of the actual map design. Played both, enjoyed both, but from the GM side there is something more satisfying about not abstracting that stuff for me.

YMMV, not appropriate for all players (although my game is a sandbox and the players do not have to interact with Barrowmaze if it frustrates them, they can go do some more linear dungeons, or whatever else they want).

Approaching it from a totally node based POV at least has consistency of design and will greatly aid game flow. Lots of professionally released dungeons I have played have done something similar, and I didn't mind it, though I was aware of the abstraction and therefore did not take the "maze" as seriously.

Btw just to be clear, i am super happy for everyone to have fun however they have fun. I am just super opposed to the argument "well traditionally...".

What was has no bearing on what is or could be. And it carries no weight.

It really depends on what you're trying to accomplish. If you want to give your players a puzzle to solve you need to have them work through the maze mechanically. If you want to give them the feeling of being lost in the maze, you can have them do stat checks and as they fail them repeat descriptions of the environment to give them the sense of doubling back on where they've been and going in circles.

 Lance845 wrote:
Btw just to be clear, i am super happy for everyone to have fun however they have fun. I am just super opposed to the argument "well traditionally...".

What was has no bearing on what is or could be. And it carries no weight.

I am also opposed to that argument. My argument was, what do your players like? Do that.....

I know what my players would want to do, and that is the "Nodal" approach. They would want to have it abstracted with a few rolls and descriptions as they navigate the Maze, that then leads up to capstone events or encounters to role-play/solve/combat.

I do not know what your players would want though.

I would add that it is also about what you as a GM like. I am excited by the idea of a "real" maze and wanted to try running it, rather than an abstraction. It tickles my fancy. As long as I know my players won't hate it, I am gonna go with it, because I don't believe that you should have to sacrifice stuff you think is cool to please your players.

I'd run it as a maze, with the 'dull blind choices' since that's what makes it a maze.

Abstracting simple and straightforward things because they might be 'boring' takes me right out of the game. If you feel you can't do a maze well or make it interesting for the players, then don't do one.

But don't turn it into some wacky flowchart or 'skill challenge' instead. At that point you're promising the meat of exploration, but hiding it away and offering a quick nothingburger in its place. At that point you might as well skip everything but boss fights.

Voss wrote:
I'd run it as a maze, with the 'dull blind choices' since that's what makes it a maze.

Abstracting simple and straightforward things because they might be 'boring' takes me right out of the game. If you feel you can't do a maze well or make it interesting for the players, then don't do one.

But don't turn it into some wacky flowchart or 'skill challenge' instead. At that point you're promising the meat of exploration, but hiding it away and offering a quick nothingburger in its place. At that point you might as well skip everything but boss fights.

So when you do overland travel in a TTRPG you do the whole thing in real time? No random encounter tables? Whats the difference to you in abstracting literally everything else in a TTRPG and a Maze?

I, for one, abstract "maze"-type encounters. Let's assume we're in D&D 5E. I'd have the players make a Survival check every [interval of time], with a total number of successful rolls needed equaling [whatever difficulty I want to set]. If it's a magic maze, I might give them disadvantage on the check. Success means they make progress toward their goal, whether that be the center of the maze, an exit, or what-have-you. Failure means they wander around lost for a while and make no progress. Time keeping is important in this situation, as it means more chances for random encounters, and, depending on what interval you set, determines whether or not the group gets lost and starves to death.

For a basic, non-magical maze, I'd have every roll take half an hour, with random encounters rolled every hour, and needing three successful rolls to get through. For more complex or magical mazes, make each roll take an hour, needing five successful rolls to get through. For truly labyrinthine mazes, make each roll take an hour and need ten successful rolls. Flavor advantage if the group maps things well, and disadvantage if there's confusion-style effects, moving walls, etc.

 Lance845 wrote:

Voss wrote:
I'd run it as a maze, with the 'dull blind choices' since that's what makes it a maze.

Abstracting simple and straightforward things because they might be 'boring' takes me right out of the game. If you feel you can't do a maze well or make it interesting for the players, then don't do one.

But don't turn it into some wacky flowchart or 'skill challenge' instead. At that point you're promising the meat of exploration, but hiding it away and offering a quick nothingburger in its place. At that point you might as well skip everything but boss fights.

So when you do overland travel in a TTRPG you do the whole thing in real time? No random encounter tables? Whats the difference to you in abstracting literally everything else in a TTRPG and a Maze?

Immediacy. If the group is in a maze, there's generally something afoot there. The point of it is to explore the maze, not pass time between points of actual interest.

Again, if there isn't anything of interest in a maze, and/or you can't make it interesting, why in the world do you have one?

For a basic, non-magical maze, I'd have every roll take half an hour, with random encounters rolled every hour, and needing three successful rolls to get through. For more complex or magical mazes, make each roll take an hour, needing five successful rolls to get through. For truly labyrinthine mazes, make each roll take an hour and need ten successful rolls. Flavor advantage if the group maps things well, and disadvantage if there's confusion-style effects, moving walls, etc.

Ok, I have to ask, checks on the half hour and hour? How big do you folks think mazes are? That sounds appropriate for areas that are 5+ miles across, not anything ever described as a maze built by people.

Voss wrote:

 Lance845 wrote:

Voss wrote:
I'd run it as a maze, with the 'dull blind choices' since that's what makes it a maze.

Abstracting simple and straightforward things because they might be 'boring' takes me right out of the game. If you feel you can't do a maze well or make it interesting for the players, then don't do one.

But don't turn it into some wacky flowchart or 'skill challenge' instead. At that point you're promising the meat of exploration, but hiding it away and offering a quick nothingburger in its place. At that point you might as well skip everything but boss fights.

So when you do overland travel in a TTRPG you do the whole thing in real time? No random encounter tables? Whats the difference to you in abstracting literally everything else in a TTRPG and a Maze?

Immediacy. If the group is in a maze, there's generally something afoot there. The point of it is to explore the maze, not pass time between points of actual interest.

Again, if there isn't anything of interest in a maze, and/or you can't make it interesting, why in the world do you have one?

For a basic, non-magical maze, I'd have every roll take half an hour, with random encounters rolled every hour, and needing three successful rolls to get through. For more complex or magical mazes, make each roll take an hour, needing five successful rolls to get through. For truly labyrinthine mazes, make each roll take an hour and need ten successful rolls. Flavor advantage if the group maps things well, and disadvantage if there's confusion-style effects, moving walls, etc.

Ok, I have to ask, checks on the half hour and hour? How big do you folks think mazes are? That sounds appropriate for areas that are 5+ miles across, not anything ever described as a maze built by people.

Consider the generally beloved movie Labyrinth. Do you watch all 13 hours of Sarah's exploration of the maze or do you hit the highlights? When I would put the players in a "maze" I would do it for something like that. Or the classic Minotaurs labyrinth. Or a sewer system that sprawls the full area of a city.

A tiny corn maze at a local farm that can be solved in 10 minutes? No. Why would that be an interesting encounter for the players?

Voss wrote:
Ok, I have to ask, checks on the half hour and hour? How big do you folks think mazes are? That sounds appropriate for areas that are 5+ miles across, not anything ever described as a maze built by people.

Well, it depends on the maze, but I'm presuming careful, methodical evaluation and a rogue checking for traps every so often. If the group wanted to make it potentially faster, I'd let them make checks twice as often, but I'd also check for encounters twice as often. Also, I'm presuming a basic maze level in a dungeon takes about three hours to solve, assuming the above. Feel free to flavor to taste.

I'm planning an ever-changing maze which the players will have to navigate.

The plan is to have every room be rectangular for easy mapping. The rooms are randomly added as they open doors, and will be removed once they are far enough away from an inactive room. Rooms will be empty, or have encounters or traps in them.

The "Map" I will have will simply show a fixed location of where they need to get to, and show where they have gotten to. They will be being talked through the maze by an Artificer with a Sending stone and this magical map, showing only their location and the exit. So it will rely on their descriptions for him and his relayed information of which way to go to get closer. Once they open onto the location of the exit, they have made it through the maze.

This allows me, as the DM, to relay to them how they are doing, through the Artificer. The traps have a lot of features to force moves for haste or to disorient the party. There is also potential for them to get split up, which I will counter by them finding previous adventurers the Artificer tried to help, and taking their sending stone.

I'm really looking forward to it; it's a shame that it's a long way into an as-yet unstarted campaign

So I think you should consider how your mechanics look from the point of view of the player who does not have your information.

They will enter the maze. Go through a few room descriptions and whatever... and then what? Will they have a reason to back track? If they do how does the randomization look to them? Whats fun? Whats interesting?

If this is being presented to them as a puzzle what is provided to them to solve it?

Consider how you are spending your prep time. Is drawing out all the cards a good use of your time if the mechanics of it will never come into play? If the players will never notice?

 Lance845 wrote:
So I think you should consider how your mechanics look from the point of view of the player who does not have your information.

They will enter the maze. Go through a few room descriptions and whatever... and then what? Will they have a reason to back track? If they do how does the randomization look to them? Whats fun? Whats interesting?

If this is being presented to them as a puzzle what is provided to them to solve it?

Consider how you are spending your prep time. Is drawing out all the cards a good use of your time if the mechanics of it will never come into play? If the players will never notice?

So they will enter the maze and be presented with a random room - let's say it's a hallway with a door to the right halfway along.

Their artificer helper will tell them they need to get to the right to get closer to their goal. Lets say this yields a T-juntion. They will have to choose a direction. They travel for 6-7 rooms and find that the route has spun around and is going the wrong way. They backtrack, and find that whilst the last few rooms are still there, further back has now changed. They have to pick a route through random rooms, guided by their artificer (who functions like a broadly scottish, swearing sat-nav with no roads to follow). So they know if they are going the right way, but there could be a dead end at any point. Plus there are challenges all the way, to add flavour, but not in every room (or it will become dull).

So the only reason for them to backtrack would be if the maze had spun them around, and they would know, thanks to the artificer helping them.

My only objection would be the Deus Ex Machina of it all. The party succeeds only because the god of the machine (artificer) literally directs them along.

It is disempowering to the players.

It might be fun, but not really different to being pulled by the nose through any old dungeon. *shrug*