Pacing a megadungeon

[Talakeal]

a question that needs to be asked is, what was the correct strategy for players?
because there is the risk of creating a morton's fork. if the players take frequent rests, the monsters will organize and defeat them. but if the players never rest, they will be exhausted and die. what could the players have done to ensure their success?

I have always designed adventures so that if the players go until they are exhausted they will pull through barring extreme bad lack or character mismatch.

The idea of smart reactive monsters shoring up their defenses in opposition to hit and run tactics is not something I am familiar with running, hence why I am asking for advice and consider it a legitimate grievance.

There are plenty of ways for the players to gather intelligence, but their characters don't really have a great build for it and none of the players really have the temperament for it.

[King of Nowhere]

I have always designed adventures so that if the players go until they are exhausted they will pull through barring extreme bad lack or character mismatch.

but a megadungeon is clearly too big to be cleared out in one go, so I guess there are some intermediate objectives that the players can achieve before stopping? Was the thing clearly communicated?

[GloatingSwine]

The idea of smart reactive monsters shoring up their defenses in opposition to hit and run tactics is not something I am familiar with running, hence why I am asking for advice and consider it a legitimate grievance.

The biggest mistake you can make is making the enemy too good at adapting to the players' strategy. Smart and reactive monsters still need to have some limitations based on their situations, and sometimes hit and run should actually work. You're not playing the Borg, you don't have to counter the players every time. If you do they'll just get frustrated, especially if you find a new way to do it every time that they couldn't have predicted in advance.

These groups should not be running with infinte resource cheats on, they should have limits and flaws that the players must find out.

There are plenty of ways for the players to gather intelligence, but their characters don't really have a great build for it and none of the players really have the temperament for it.

For the sort of scenario you were trying to run*, gathering intelligence is so important that it shouldn't ever be build dependant. There might be different ways to get at the information but it's so critical to being able to play the game that everyone always needs it, but your players are going in blind.

*That, I think, your players do not know they are playing. They do not know they are playing a strategy puzzle game where they need to figure out all the possible responses their opponent might have and account for them in advance, but they are.

Edit: Another good rule to stick to is that the first couple of times at least the players try a deliberate strategy, let it work because then they'll actually be motivated to try strategies. And when it's not going to work make sure they know that in advance.

[Talakeal]

The biggest mistake you can make is making the enemy too good at adapting to the players' strategy. Smart and reactive monsters still need to have some limitations based on their situations, and sometimes hit and run should actually work. You're not playing the Borg, you don't have to counter the players every time. If you do they'll just get frustrated, especially if you find a new way to do it every time that they couldn't have predicted in advance.

These groups should not be running with infinte resource cheats on, they should have limits and flaws that the players must find out.



For the sort of scenario you were trying to run*, gathering intelligence is so important that it shouldn't ever be build dependant. There might be different ways to get at the information but it's so critical to being able to play the game that everyone always needs it, but your players are going in blind.

*That, I think, your players do not know they are playing. They do not know they are playing a strategy puzzle game where they need to figure out all the possible responses their opponent might have and account for them in advance, but they are.

Edit: Another good rule to stick to is that the first couple of times at least the players try a deliberate strategy, let it work because then they'll actually be motivated to try strategies. And when it's not going to work make sure they know that in advance.

In this case the strategy was one group of monsters going to the group of monsters across the hall and saying, hey, let's make a pact, if you hear us being attacked, come across and help as out and we will do the same, ok?

Not exactly the Borg there.

[Quertus]

In this case the strategy was one group of monsters going to the group of monsters across the hall and saying, hey, let's make a pact, if you hear us being attacked, come across and help as out and we will do the same, ok?

Not exactly the Borg there.

Yeah, but… look at it from your players’ point of view - how would you, as a player, handle my megadungeon that would do such things? Where I’m roleplaying you, making sure that the monsters are already challenging, and that these two teaming up leads to a TPK (even if by another name of “surrendering”)?

[Talakeal]

but a megadungeon is clearly too big to be cleared out in one go, so I guess there are some intermediate objectives that the players can achieve before stopping? Was the thing clearly communicated?

I suppose not, I probably was playing up the mystery and freedom a bit too much.

It doesn't have clearly defined objectives, but it does have wings. The idea was that the players would explore one wing at a time, but instead they kind of drifted around and fell back when they bit off more than they could chew.

Note, however, that I am not playing the whole dungeon as a hive mind; monsters only react when the party engages an intelligent group and then fails to deal with them, and even that reaction is only directly related to that monster's "territory".




Honestly, from a mechanical standpoint, they aren't doing too bad. Making deals with monsters puts them in a better position than going into wholly hostile territory, but the hit to their pride may not be something they can recover from. Like most of my horror stories, this one comes down to a battle of egos.

Yeah, but… look at it from your players’ point of view - how would you, as a player, handle my megadungeon that would do such things?

You're mega-dungeon specifically? I would need to go back and reread / think about it.

In general, it really depends on the character.

If I was in my players shoes, I would pump the friendly monsters for all the information I could about the dungeon, talk them into working together with us and each other to create a safe area of the dungeon with a defended bulwark against the other denizens, and then manipulate them into believing it is in our best interests to take the magic items and treasure in exchange for trinkets from the human world.

[GloatingSwine]

In this case the strategy was one group of monsters going to the group of monsters across the hall and saying, hey, let's make a pact, if you hear us being attacked, come across and help as out and we will do the same, ok?

Not exactly the Borg there.

But it is. It's a new adaptation to their strategy that stops it from working which is so different from the last adaptation that even if they had countered that they would have failed again in a new way. Because you're the DM and you're in control of the world you can always come up with new adaptations by just inventing new things that could have happened to stop them, like a pact between groups of monsters.

There need to be rules about what you allow yourself to do to counter your players in order to make what you're trying to do satisfying to play, and your players need to be able to discover and understand those rules just as well as you do so that they can predict these outcomes. They should have known an alliance between these groups was not only possible likely before they came into conflict with them. And them knowing it means you needing to build the scenario such that they basically can't avoid finding it out irrespective of their character builds.

Because right now, your players don't know the rules of the game they're trying to play. And so they're failing every time due to things completely beyond their power to account for. They can't have strategies, they can't have plans, because strategies and plans require understanding the rules in which they are expected to operate.

Consider: In the first session they pushed as hard as they could and wore themselves out and had to retreat to rest, then in the second session the same enemy was set up so that if they did it again they would fail (and that's what happened). The enemy had unlimited ability to recover to a position much stronger than they had started and so the players had made negative progress.

In the third session they tried not to wear themselves out by using resource conserving hit and run attacks, but again the situation forced them to fail by trying that due to introducing something they had no way of accounting for (alliances between monster groups).

So the only rule right now is "there are no correct approaches, everything we do gets hard countered".

[Talakeal]

Two generic questions about dungeon crawling:

Most doors open inward so that people can't simply open them by taking them off the hinges, correct? Is there a way to stop such a door from being opened from the outside? Spiking doors work great for stopping monsters getting in, but is there a similar technique for stopping them from getting out?

Second, how easy is it too knock on a stone wall to tell how thick it is? In video games this is a very common tactic for checking for secret rooms, but I have never heard of anyone doing this in a tabletop dungeon crawl.

But it is. It's a new adaptation to their strategy that stops it from working which is so different from the last adaptation that even if they had countered that they would have failed again in a new way. Because you're the DM and you're in control of the world you can always come up with new adaptations by just inventing new things that could have happened to stop them, like a pact between groups of monsters.

There need to be rules about what you allow yourself to do to counter your players in order to make what you're trying to do satisfying to play, and your players need to be able to discover and understand those rules just as well as you do so that they can predict these outcomes. They should have known an alliance between these groups was not only possible likely before they came into conflict with them. And them knowing it means you needing to build the scenario such that they basically can't avoid finding it out irrespective of their character builds.

Because right now, your players don't know the rules of the game they're trying to play. And so they're failing every time due to things completely beyond their power to account for. They can't have strategies, they can't have plans, because strategies and plans require understanding the rules in which they are expected to operate.

Consider: In the first session they pushed as hard as they could and wore themselves out and had to retreat to rest, then in the second session the same enemy was set up so that if they did it again they would fail (and that's what happened). The enemy had unlimited ability to recover to a position much stronger than they had started and so the players had made negative progress.

In the third session they tried not to wear themselves out by using resource conserving hit and run attacks, but again the situation forced them to fail by trying that due to introducing something they had no way of accounting for (alliances between monster groups).

So the only rule right now is "there are no correct approaches, everything we do gets hard countered".

I am missing something.

The first few pages of this thread seemed to be about how I was playing my monsters too dumb and they need to be smart and reactive. Now that seems to be bad.

These are players with decades of gaming experience, not little kids in their first RPG, why wouldn't they expect humanoid enemies to be able to employ basic tactics like communicating with one another, blocking doors, and keeping watches?

Likewise, what sort of "rules" should I set for myself? And how do those avoid breaking verisimilitude?


Also really curious about how players can get intelligence regardless of their build or proclivities without stripping away all immersion and mystery by just delivering tons of forced exposition.

[King of Nowhere]

I am missing something.

The first few pages of this thread seemed to be about how I was playing my monsters too dumb and they need to be smart and reactive. Now that seems to be bad.

Take everything you read on an internet forum with a grain of salt. You can find useful ideas to try, but that's all. You certsinly csnnot get a miracle solution.

Also really curious about how players can get intelligence regardless of their build or proclivities without stripping away all immersion and mystery by just delivering tons of forced exposition.

By asking questions. Same as you'd do in real life.
Ask people around what they know. Maybe somebody was already in that dungeon, track them down and ask them to draw a map. Pay if necessary.
Hire people. Maybe you don't have divination spells, but somebody else in the world does. Find him, pay him for some spellcasting. Ssme goes for some kind of explorer.
Get allies. You find a goblin tribe inside the dungeon, do not attack on sight unless they are immediately hostile. See if you can negotiate safe passage across their land; should not be hard, they don't want to die and you are not damaging them just passing through. And since you're there, ask them if they can tell you what's in the rest of the dungeon.

If they are intelligent, they can negotiate and there's rarely a need to fight. I bet my players could finish the dungeon in one go, almost without rolling dice.
Of course, this requires the players to not be murderhobos eager to kill sapient beings for a handful of copper

[Satinavian]

I am missing something.

Do you remember being told how your players will inevitably complain if you don't hand them easy wins ?
And that the megadungeon setup might not be a good fit for your group at all in the first place ? And how your attempts at managing pacing might come over as restrictive instead ?

Many people have told you how they would do a megadungeon. I would put intelligent enemies into mine as well. But i wouldn't run one for Bob or Brian.
Of course intelligent enemies are harder than normal enemies. And your players nearly always complain about things being "too hard" whenever get a setback. The only way to make them happy is providing easy wins and disguising that so that they can feel smart.

[Talakeal]

Take everything you read on an internet forum with a grain of salt. You can find useful ideas to try, but that's all. You certainly cannot get a miracle solution.

Good point. I need to stop looking at the forumites as a hive mind, but this thread in particular is giving me whiplash and making me think I am missing something key.

By asking questions. Same as you'd do in real life.

Ask people around what they know. Maybe somebody was already in that dungeon, track them down and ask them to draw a map. Pay if necessary.

Hire people. Maybe you don't have divination spells, but somebody else in the world does. Find him, pay him for some spellcasting. Same goes for some kind of explorer.

Get allies. You find a goblin tribe inside the dungeon, do not attack on sight unless they are immediately hostile. See if you can negotiate safe passage across their land; should not be hard, they don't want to die and you are not damaging them just passing through. And since you're there, ask them if they can tell you what's in the rest of the dungeon.

If they are intelligent, they can negotiate and there's rarely a need to fight. I bet my players could finish the dungeon in one go, almost without rolling dice.
Of course, this requires the players to not be murderhobos eager to kill sapient beings for a handful of copper

So here's the thing.

My players care about two things above all else; wealth and pride.

They care more about this than success, ethics, even power and their character's lives, and certainly more than lore or storyline. Even XP means less to them than gold, which always struck me as very odd.

They would never pay someone for divinations because that is turning treasure into "useless" lore. The same reason they would never play a diviner, it turns combat power into useless lore.


They see having to make deals with monsters as a catastrophic failure because it hurts their wealth and their pride, even though from a utilitarian perspective (meaning their odds of getting to max level and completing the dungeon) they are far better off now than they were before.

[Slipjig]

What do your players think money is for? Money is only good in that it allows you to purchase things. Maybe have a mentor NPC/patron say something about, "you need to spend money to make money". Do they refuse to use consumable items or cast spells that have expensive material components, too?

As for your questions about knocking on walls to find secret doors, that's actually reasonable. A stone door that is as thick as the surrounding wall would be too heavy for a human to push open, so a secret door in a stone wall will either not be made of stone, or will be a thin veneer of stone over another substance. In either case, it may sound different when knocked on. There's a reason you hide secret doors behind much lighter objects like paintings or bookshelves.

[kyoryu]

In this case the strategy was one group of monsters going to the group of monsters across the hall and saying, hey, let's make a pact, if you hear us being attacked, come across and help as out and we will do the same, ok?

Not exactly the Borg there.

You're thinking in terms of "complexity of counter strategy". You should be thinking of "how much more difficult has it become?". An encounter suddenly doubling in numbers and coming from an undefended direction is far, far more than doubling the difficulty of the encounter.

The ideal adaptation increases a bit more gradually, and makes the undesired behaviors less effective while still leaving the desired behaviors viable.

So, having the enemies set up behind the PCs without actually starting the attack might work, as it would basically blow up their easy "get out" route, while encouraging them to go further into the dungeon.

[DrMartin]

Also really curious about how players can get intelligence regardless of their build or proclivities without stripping away all immersion and mystery by just delivering tons of forced exposition.

a couple of ways you could go about this:

- create a framework for downtime actions, one of which is "gather information". You always get a helpful result, without a roll. You can roll to get some extra information. You can spend money to get some extra information as well.
be generous with the information, even giving out-of-character clear words explanations for in-character concepts that your players might not connect to their gaming consequences. As the GM you are your players eyes and ears and they cannot read your mind - even when you think what you are telling them it´s so obvious that anyone would get it.

- allow each skill to be useful in social circumstances and information gathering as well, to represent the network built around a specific activity, or the ability to find relevant contact and information about your field of specialty. Being proficient in riding horses means you can also roll Riding to find someone to buy a horse. Having a good attack skill means you can roll to find the best weapon dealers. Being good at <Dungeon Skill> means you know the right questions to ask about dungeons, even if your social skills sucks, and that you are great at socializing with other people that are also good at <Dungeon Skill> .

combine one and two to taste!

[Quertus]

You're mega-dungeon specifically? I would need to go back and reread / think about it.

In general, it really depends on the character.

If I was in my players shoes, I would pump the friendly monsters for all the information I could about the dungeon, talk them into working together with us and each other to create a safe area of the dungeon with a defended bulwark against the other denizens, and then manipulate them into believing it is in our best interests to take the magic items and treasure in exchange for trinkets from the human world.

No, not a megadungeon I mentioned earlier, nor me as a megadungeon, but one I made “in your style”. And your answer… is really good. That last bit would probably bite you more often than help you on most intelligent monsters I’d run, but it’s a clever idea that could be useful if used properly on the correct targets.

However, it’s a really Epimethian answer. It’s “once we’ve failed, gotten in over our heads, and surrendered, here’s how I’d make lemonade out of those lemons”. And maybe that’s the key - maybe the biggest takeaway should be that you know how to make lemonade, whereas your group is allergic to lemons. Food for thought.

But, back to my original question here: how would you avoid getting into a TPK situation in the first place if I made a challenging, reactive dungeon?

Second, how easy is it too knock on a stone wall to tell how thick it is? In video games this is a very common tactic for checking for secret rooms, but I have never heard of anyone doing this in a tabletop dungeon crawl. .

Um… apparently really easy, if you have the skills? I seem to have a racial penalty to the check, as I’m an “anti-dwarf” irl, and can’t hear the things everyone else tells me are obvious.

These are players with decades of gaming experience, not little kids in their first RPG, why wouldn't they expect humanoid enemies to be able to employ basic tactics like communicating with one another, blocking doors, and keeping watches?

You know, my senile mind says this isn’t the first time I’ve been impressed by how good an answer you’ve given me to a “what would you do (as a player)?” question. Whereas, from your stories, no, I don’t expect your players to outperform the children I’ve gamed with.

Which makes me wonder: how are the games different when you’re a player as opposed to the GM? How do the challenges you face vary?

In fact, as usual, I’d suggest using this opportunity to run a series of one-shots; this time, however, I’d have each member of the group run a 1-shot, with the intent to use them as examples and reference points for a conversation to communicate what everyone wants out of the game.

For example, Brian may run a world where all the NPCs are mute.

Also really curious about how players can get intelligence regardless of their build or proclivities without stripping away all immersion and mystery by just delivering tons of forced exposition.

Eh, different builds can gather information differently. Quertus will gather info simply by what he sees, or via Divinations if he’s really interested in knowing things ahead of time. Armus will allow monsters to exit the dungeon, ambush them, and interrogate them. Pidge will read their minds, while invisibly watching them for days if necessary. Another character might investigate their crafts, materials, defenses, diet, fecal matter, myths, language, or any number of other data points in an attempt to gain further information about the dungeon (EDIT: “these defenses are made from Lizard bones, and appear to be designed to stop charging monsters… on the floor, walls, and ceiling.”).

It’s not about having a “one size fits all” Automatic info dump, it’s about accepting (or at least giving a chance to) any reasonable methods your players come up with. And tailoring the information they get to the method they use, and building content such that most any methods will provide enough answers… or, for some methods, enough questions to encourage them to search for more answers.

Thing is, your players don’t seem to want to play the Information game, do they?

Good point. I need to stop looking at the forumites as a hive mind, but this thread in particular is giving me whiplash and making me think I am missing something key.

I mean, I advised some largely non-reactive dungeons (physically incapable of reacting, stuck in Groundhog Day as one, mostly mindless as another), so we were definitely not all on the same page there.

So here's the thing.

My players care about two things above all else; wealth and pride.

They care more about this than success, ethics, even power and their character's lives, and certainly more than lore or storyline. Even XP means less to them than gold, which always struck me as very odd.

So your job as GM should be to give them an adventure high in wealth and pride. Hint: defeating them with Tuckers kobolds is the opposite of that.

So you need foes that are powerful and foolish, that the party (your players) can outsmart. “Average citizens” dumb enough to have not outsmarted your players, where they will ask for help with monsters that can be kited or defeated with simple traps. You need a world dumb enough that your players (and thereby their characters) look smart in comparison. In a word (and I can’t believe I’m suggesting this), you need the Forgotten Realms, as Ed writes it, where everyone wears their pants on their heads, and thinks it is normal to do so. You need the land of the blind, where your one-eyed players can be king. I never thought I’d be praising that facet of FR, or advising anyone to replicate it, but here we are.

I’m thinking that that’s what your players mean when they say that you take the game more seriously than they do. They want a game where they can come across as smart and competent without having to play the “5d Wizard Chess” that they perceive you to do. Or so I suspect.

So, find the most brain-damaged / mindless module you can find, and try running your group through that.

They would never pay someone for divinations because that is turning treasure into "useless" lore. The same reason they would never play a diviner, it turns combat power into useless lore.


They see having to make deals with monsters as a catastrophic failure because it hurts their wealth and their pride, even though from a utilitarian perspective (meaning their odds of getting to max level and completing the dungeon) they are far better off now than they were before.

Eh, would you really want to read the story of Conan being Tuckered? Would that really make Conan feel like a BDH if it were the first Conan story you ever read? Because (unless they’re playing existing characters from old campaigns) this is their characters’ debut, their first story, and they’re not coming across as Conan.

I’m with your players here. Don’t get me wrong, I play a tactically inept Academia mage, and a whole ship of fools, in addition to the more competent characters in my range. Your players don’t. They’re only interested in ego-boosting BDHs. That’s what you need to promise (in your head, not out loud), and you need to keep that promise. If you haven’t let them tell the story of how their characters are at least Conan level BDHs, you’ve failed to write an acceptable scenario for your players.

[Talakeal]

So, having the enemies set up behind the PCs without actually starting the attack might work, as it would basically blow up their easy "get out" route, while encouraging them to go further into the dungeon.

That's actually not too far from what happened. They cut off the PCs escape, and then demanded a tribute for letting them go back. Actually attacking would have been both less interesting and less "fair" from a CaS perspective.

But, back to my original question here: how would you avoid getting into a TPK situation in the first place if I made a challenging, reactive dungeon?

By being smarter than the smarties and tougher than the toughies!

Joking aside, lot's of little stuff. But, admittedly, it is tougher at low levels.

Eh, different builds can gather information differently. Quertus will gather info simply by what he sees, or via Divinations if he’s really interested in knowing things ahead of time. Armus will allow monsters to exit the dungeon, ambush them, and interrogate them. Pidge will read their minds, while invisibly watching them for days if necessary. Another character might investigate their crafts, materials, defenses, diet, fecal matter, myths, language, or any number of other data points in an attempt to gain further information about the dungeon (EDIT: “these defenses are made from Lizard bones, and appear to be designed to stop charging monsters… on the floor, walls, and ceiling.”).

It’s not about having a “one size fits all” Automatic info dump, it’s about accepting (or at least giving a chance to) any reasonable methods your players come up with. And tailoring the information they get to the method they use, and building content such that most any methods will provide enough answers… or, for some methods, enough questions to encourage them to search for more answers.

Thing is, your players don’t seem to want to play the Information game, do they?

No, they don't.

There are lots of ways to do it, none of which much interest this party.

So your job as GM should be to give them an adventure high in wealth and pride. Hint: defeating them with Tuckers kobolds is the opposite of that.

So you need foes that are powerful and foolish, that the party (your players) can outsmart. “Average citizens” dumb enough to have not outsmarted your players, where they will ask for help with monsters that can be kited or defeated with simple traps. You need a world dumb enough that your players (and thereby their characters) look smart in comparison. In a word (and I can’t believe I’m suggesting this), you need the Forgotten Realms, as Ed writes it, where everyone wears their pants on their heads, and thinks it is normal to do so. You need the land of the blind, where your one-eyed players can be king. I never thought I’d be praising that facet of FR, or advising anyone to replicate it, but here we are.

I’m thinking that that’s what your players mean when they say that you take the game more seriously than they do. They want a game where they can come across as smart and competent without having to play the “5d Wizard Chess” that they perceive you to do. Or so I suspect.

So, find the most brain-damaged / mindless module you can find, and try running your group through that.

That ribbing on FR gave me a good laugh.

That said, I just don't think I could enjoy running a game like that.

Eh, would you really want to read the story of Conan being Tuckered? Would that really make Conan feel like a BDH if it were the first Conan story you ever read? Because (unless they’re playing existing characters from old campaigns) this is their characters’ debut, their first story, and they’re not coming across as Conan.

Conan specifically, not of the top of my head, no.

But plenty of the fantasy movies I grew up on (Willow, Return of the Jedi, The Last Unicorn, Flight of Dragons, etc.) have scenes where the heroes are captured and end up teaming up with their captors against the true villains.

I’m with your players here. Don’t get me wrong, I play a tactically inept Academia mage, and a whole ship of fools, in addition to the more competent characters in my range. Your players don’t. They’re only interested in ego-boosting BDHs. That’s what you need to promise (in your head, not out loud), and you need to keep that promise. If you haven’t let them tell the story of how their characters are at least Conan level BDHs, you’ve failed to write an acceptable scenario for your players.

Yeah, I guess.

But for me, part of being a Big Damn Hero is being tough and rolling with the punches, not throwing up your hands and committing suicide the first time things don't go your way.

I really like characters who never give up, and for me I really like the drama of captured characters, I love movies like Bridge on the River Kwai or the episode of Star Trek where Worf is in the prison camp. Part of what pissed me off so much about Brian's game is that everyone wants to play a character with a legendary resolve score on paper, but don't actually want to RP anything but a whiny little quitter in the game.

[gbaji]

So, IMO the ideal game-play loop is the players play as smart as they can to conserve resources and push as deep into the dungeon as they can, and then fall back when their resources are depleted. HOWEVER, the players don't have great scouting ability (and in a dungeon scouting can be pretty tough at the best of times) so that if they decide to push on, go into a new room, and bite off more than they can chew, and fall back, they will put the monsters on alert and make further progress very difficult.

Yes. The "delve as deep as possible, then retreat" is a logical method for exploring some vast underground complex full of "bad things" to fight. I suspect the problem here is the structure of your dungeon. You have to actually think about how real underground "worlds" would be built. They would not be a massive series of tightly constructed hallways and walls, and rooms (like a lot of "dungeon maps actually use"). They would be a series of tunnels connecting a number of larger caves, with the caves having various features in them (and creatures). Some areas maybe more "worked' (meaning actual walls built, with tools and whatnot), but you need to rationalize why those things are there. Most areas should be natural caves, and those don't tend to cluster together closely.

Also, you need to think in terms of this being an ecosystem. Each set of creatures in each area has to live. They have to eat. They have to be able to build things within their areas (if sentient), and survive potential attacks from other creatures in nearby areas. As a result, I tend to build dungeons like this in sections, with each section being "this is where the <insert creature X> live, where they get their water, where they grow their food, heres where some wildish creatures live nearby which they hunt, etc. These sections are designed to be areas the PCs should be able to handle in a single "go", so to speak. You, as the GM, should make it relatively obvious to the players when this is some small batch of violent monsters living in said set of caves, or an intelligent set of enemies, or maybe even allies (or at least neutral folks who will maybe grant you passage, if you help defeate some monster in the caves "over there", that has been raiding them and causing problems or something).

It realy sounds like your players are not treating the world around them as a "real world" and the NPCS within it as actual creatures/people who actually live in that world. Not sure how you can manage this though, except to make it *really* obvious to them which things they're supposed to be fighting and killing, and which things they are supposed to be talking to instead. You'd think the simple observation that "hey, those critters over there are pointing sharp things at us, but not actually attacking, and are speaking to us. Maybe we should try to communicate instead of just killing them home-invasion style". That usually works. Maybe not with your players though.

Because the other possibility is that they're blundering around expecting the loop of the game to be a kick the doors in and take the treasure dungeon crawl where they don't have to think about that kind of stuff, and you're playing a world simulation back to them they aren't expecting or accounting for.

Yeah. I do think they just expect a classic dungeon hack. And hey. As much as I abhor the concept from a "this makes no sense" game world building pov. If that's honestly what my players really really want, I'd just give it to them. But I'd contrive a reason why this is the case. I literaly introduced some other plane the PCs found a way to a bit a go, where it was (horribly cribbed from various MMORPG game zones) basically a "dungeon hack". Complete with planar manifested "monsters" of various types, that... yeah, literally had bodies that faded away when defeated, leaving only the "treasure" behind, and said treasure could include various things that could be assembled into keys to open up more difficult/powerful/treasure-granting areas of said mega dungeon. And yeah, the bad guys will just hang out in their various rooms waiting to be defeated. Cause that's why they are there.

But yeah. I literally contrived this as "this plane was created by some deity as a testing ground for heroes", and left it at that. High level. High power. Gets tougher as you go. I tend to avoid this like the plague in any location in my game settings where "real people" live though. But yeah, if that's what the players want, then give them that. Sometimes, meaningless dungeon hack stuff is a lot of fun. Not something I'd base a long running campaign on, but sure.

Honestly, I can't think of any way to break that aside from working full time as the PCs "PR team" throwing impossibly easy fights at them but somehow conveying AFTER THE FACT* that this was an incredible victory that only someone as smart and powerful as they were could have possibly won, but they managed to score 110%** victory!

Clearly, they need Sir Robin's bard!

Because right now, your players don't know the rules of the game they're trying to play. And so they're failing every time due to things completely beyond their power to account for. They can't have strategies, they can't have plans, because strategies and plans require understanding the rules in which they are expected to operate.

Consider: In the first session they pushed as hard as they could and wore themselves out and had to retreat to rest, then in the second session the same enemy was set up so that if they did it again they would fail (and that's what happened). The enemy had unlimited ability to recover to a position much stronger than they had started and so the players had made negative progress.

In the third session they tried not to wear themselves out by using resource conserving hit and run attacks, but again the situation forced them to fail by trying that due to introducing something they had no way of accounting for (alliances between monster groups).

So the only rule right now is "there are no correct approaches, everything we do gets hard countered".

Yeah. I'm not sure how much of the "rule" was actually conveyed to the players (or how much "stuck" in their minds anyway). I also don't think that's a correct assessment. The first session, they fought a single fight, didn't come anywhere near exhausting their resources, defeated one room, then retreated. The second session, they returned to the same location, to find that the kobolds they'd attacked (and killed just one room of), had prepared for the party's return, ambused them, and forced their surrender. The lesson there should have been "don't attack a small portion of one group unless you either continue on to defeat the entire group, or don't ever come back".

The third session was a case of them once again attacking and just killing the first room of a group of monsters, and then retreated (so they didn't "learn the lesson at all"). This time, the monsters they attacked, formed a pact with another set of nearby monsters to engaged in mutual defense against the party, so when the party returned they got overwhelmed again.

Seems to me like this isn't a case of "damned if you do, damned if you don't", but the players literally repeating the exact same mistake over and over again and then beisng surprised when it fails in the exact same way. ie: if you attack someone, but don't actually finish them off, they'll take some action to be better able to handle you if you return. Twice in a row, they attacked a group of opponents, just killed the first batch they ran into, and then left. Then, they just came right back expecting to... what? Just kill the next room of opponents and then retreat again and rinse/repeat forever? The first time this "failed" should have been a wake up call that this tactic is a poor one. That they effectively repeated the exact same tactic the very next game session kinda speaks volumes about these players.

This is seriously a "light grenade" scenario. If the players can't noodle this out...

Second, how easy is it too knock on a stone wall to tell how thick it is? In video games this is a very common tactic for checking for secret rooms, but I have never heard of anyone doing this in a tabletop dungeon crawl.

Actual secret walls/doors/panels in actual "dungeons" (or castles) are not how they are usually depected in films/tv. They aren't solid stone sections that slide to the side, or out, or in, or whatever (barring some kind of magic thingie, or really good crafting work). A 6" thick actual stone wall large enough to walk through would weigh a ridiculous amount (a little over 80lbs per 1'x1'x6" section, so a 5'x3' section would weigh over 1200lbs) and would never actually support its own weight if it were something that could slide, or open on a hinge or something. Such "real stone" panels would be something that might be opened once maybe, and require significant work to get them back into place. More likely a "secret door" would likely be a woden door, with some kind of veneer on it to make it look like the solid stone wall around it. So some mortar, and fake brick spackled on the door to match the real stone bricks and mortar around it. Actually trying to do this to match a stone wall (and make it look correct) is quite difficult, and often very easy to spot if you take a short amount of time actually looking closely. Hence, hiding them behind other objects in a room anyway or otherwise disquising them. So yeah. You can absolutely tell whether what appears to be a stone wall is just a facade, or is actually a full thickness stone or brick wall. And yeah. If well disquised visually, tapping will tend to work well.

More common (and honestly harder to detect) are actual wooden secret doors, in actual woden structures. Realize that even in stone castles, a heck of a lot of the interior may be wooden floors, walls, etc. I spent a few years as a kid living in a very rural redwood forested area. There was this very interesting guy (apparently a professor at the local college, but I didn't actually know that at the time). He lived a few blocks from a friends house (well, blocks is misleading, more like down the semi paved road that ran behind my friends house that we rode our bikes on). He had built an actual castle on his property. Like three stories tall, stone outer wall (one level high), with wooden structures filling the interior (and overlooking the outer wall itself). The thing was chock full of secret passages and doors. My friend and I used to go over there and play in that castle for hours and hours, just wandering around, climbing the walls, finding secret trap doors (not actual "traps" of course), cllimbing back inside the walls, finding more secret doors that went into various rooms from the back side, finding more passages that went under ground, and other ladders that would take you to up though trap doors into other areas inside the castle, etc. Once you found you way into the secret areas, there were winches or latches that opened up other doors and what not. It was fun! Like ridiculously fun (and probably an insurance nightmare today, but no one cared about that back then). Apparently, it's been built up into even more stuff since then over time, and the castle itself burned down and had to be rebuilt, so who knows if I'd be able to navigate it like I did as a kid.

Wooden "secret doors", if they are built even semi intelligently and only hinge/open on the "other side" are quite difficult to tell from any normal section of wooden wall framed by studs and supports. You can noodle it out if there's some "wiggle" in them. That's about it. Stone wall sections? Either completely impossible if they are as thick as the actual walls (but have some magic, or dwarven crafting to make them or something), or really quite easy if made via more "mortal" methods.

Also really curious about how players can get intelligence regardless of their build or proclivities without stripping away all immersion and mystery by just delivering tons of forced exposition.

I suppose that somewhat depends on how much of a stickler you are going to be as a GM to require specific intelligence gathering skills to be used. On the one hand, if no one has them, then you are condemning your players to constantly stumble around (which maybe should encourage them to learn this stuff). On the other hand, if you do allow PCs to "figure stuff out" even without such skills, then it eliminates the reason for having them in the game, and will effectively punish players for taking them. So you need to decide this, and let your players know during character build if such skills are important, so that they can intelligently choose them (at least make sure someone has some appropriate skills).

Having said this, there's a lot of stuff that should be possible to learn without special skills. Seriously. As the GM, you control the NPCs. Unless the first thing the NPCs do when seeing the PCs is attack, that should be a hint to the PCs to stop and talk to them. And that should lead to simple roleplaying to determine what happens. If the NPCs don't want violence, and the PCs would prefer to not have to fight these guys if there's another option, this should not take special skills to negoatiate.

I guess I'm somewhat flumuxed here, because to me the default condition for any encounter should be "not to fight". So if you aren't having your NPCs automatically attack, then what? Are the PCs just automatically attacking any NPC they encounter, thus forcing this into a combat situation? If that's the case, there's your problem right there.

[Vahnavoi]

The "downward spiral" is completely psychological, not mechanical.

Then what you need is a coaching handbook for how to get a bunch of negative nellies out of a funk. This is a known thing, there are things you can do about it, but they exist outside the realm of game design.

Two generic questions about dungeon crawling:

Most doors open inward so that people can't simply open them by taking them off the hinges, correct? Is there a way to stop such a door from being opened from the outside? Spiking doors work great for stopping monsters getting in, but is there a similar technique for stopping them from getting out?

Second, how easy is it too knock on a stone wall to tell how thick it is? In video games this is a very common tactic for checking for secret rooms, but I have never heard of anyone doing this in a tabletop dungeon crawl.

For the first question, which way doors open is decided by more than one consideration. Especially indoors, where space requirements and direction of common movement are important considerations. For example, where I live, castle gates open inward (so. they'd be easy to defend, similar to what you note), but church doors open outward, so that if there's a fire inside the church during mass gathering, people rushing out will not block the door.
Apartment houses may have double doors for heat or noise control, in which case the outer door opens outwards and the inner door opens inwards.With that additional context out of the way, beams can be used to hold a door closed from either side. If the door opens away from you, some hooks or claws need to be attached to anchor the beam. Depending on make of the door, other solutions are possible. For example, a metal door can be solded or welded shut. Wooden doors can sometimes be frozen shut with enough water. Nailing or gluing planks across a wooden door will at least make pulling it open more difficult. If there's enough time, build another door that opens towards you and the barricade that.

For the second question, different materials will make different sounds when they are struck, and the sounds will also differ based on what's behind. Ultrasound imaging is the high tech version of this. Measuring how thick a wall is, is hard, but almost anyone with ears can, just by trial and error, notice hollow spaces behind thin walls or when material of wall changes from one to another. You can probably test this immediately in the house you live in. If you've never heard of anyone doing this in a tabletop game, you've not played games where architechture is important. It isn't just for finding secret doors. It's also for finding weak wall sections to break or repair, for finding hidden or obscured details in structures, in some cases for finding ore or water veins, etc.

[Quertus]

That ribbing on FR gave me a good laugh.

Glad you enjoyed it.

That said, I just don't think I could enjoy running a game like that.

Give it a try. You might be surprised.

But... let's look at this from another PoV.

Look at just how often your players make boneheaded mistakes. And just how bad those mistakes are.

Let's assume you'll forget or underestimate some of those errors, and so let's set the baseline at "make worse mistakes more often".

Let's call that baseline "genius level" in your new setting.

And let's never have the PCs encounter anything at genius level. Let's always only have them encounter things between "mindless" and "average" for their first campaign in this new setting.

Of course the (secret) name of this setting is "The Land of the Blind". But now the secret tagline is, "Like most GMs, Talakeal is terrible at roleplaying just how stupid real people really are", since the "This is how dumb the setting has to be to make your players feel good" tagline makes you feel unmotivated. (Is that fair? No. But it is your fault if the PCs come off as dumber than the players. So ask yourself: are your players really that dumb? If not (and I assume they're not), until you fix that gap, make the other denizens of your world just as dumb as you make the PCs. Maybe that'll encourage you to figure out how to fix that gap. Consider it your punishment, the one your players probably want to give you when they say that you care more than they do.)

Does this alternate perspective, of self-blame (AKA "This is an Opportunity, because it's something *I* can fix!"), and caring more about your players and less about your setting, make you feel any more motivated to create The Land of the Blind, Forgotten Realms rip off #980q391037?

Conan specifically, not of the top of my head, no.

But plenty of the fantasy movies I grew up on (Willow, Return of the Jedi, The Last Unicorn, Flight of Dragons, etc.) have scenes where the heroes are captured and end up teaming up with their captors against the true villains.

And in how many of those did the protagonists start out with a desire to genocidally murder and loot these future allies based on their race / species? I think there's a certain shared something in most of those movies, that contrasts with a (very important to maintain) complete lack of respect for their very existence in most "murder other sentient beings" Fantasy settings.

Yeah, I guess.

But for me, part of being a Big Damn Hero is being tough and rolling with the punches, not throwing up your hands and committing suicide the first time things don't go your way.

I really like characters who never give up, and for me I really like the drama of captured characters, I love movies like Bridge on the River Kwai or the episode of Star Trek where Worf is in the prison camp. Part of what pissed me off so much about Brian's game is that everyone wants to play a character with a legendary resolve score on paper, but don't actually want to RP anything but a whiny little quitter in the game.

That's tough.

So, there's lots of components to this. And I don't give your players credit for understanding even as many of them as I'll bother to list.

For some, like "prison camp" (where I don't remember the Worf reference, I only remember "There. Are. Four. Lights!" or whatever)... I have negative interest in roleplaying through anything like that, regardless of my character. SO, to repurpose some words, despite the Simulation saying that "imprisoned + strong willed" is a good combination, Gamist concerns of "the players find that anti-fun" suggest that's a bad target to aim for.

"Having a high ____" (saving throw, pain tolerance, torture resistance skill, whatever) is just a die roll, not something that is itself directly roleplayed. IRL, I had a root canal without anesthetic. But very few people who know me casually would suspect I have a high tolerance for pain. The correlation between these stats and roleplaying... aren't exactly something I'd expect from your group, so let's call them "unrelated". (On an unrelated note, I would love to be able to birth a child. But unless medical science has advanced further than I'm aware, XY says that's impossible, so I could never even call it a goal or bucket list item.)

"Rolling with punches", making lemonade out of lemons... is an acquired taste, and not the only form of being an underdog, a state which is itself not exactly essential to gaming.

I suspect that your players are "allergic" to Setbacks - to events in game leading to a game state that is worse than where they started. And that's a perfectly valid preference, even for people who are the picture of mental health, who don't suffer from depression.

I'm going to interrupt myself, and suggest you try a 1-shot of Shadowrun (or CP2020, or some other "Heist" game), where the PCs a) are the underdogs (compared to the Megacorp or whatever); b) absolutely and obviously HAVE to do recon / gather intel / come up with a plan b2) *before* the action (so none of this retcon powers ** from RPGs designed to emulate Heist movies (that would, in addition to being bad on its own merits, would give your players even more bad habits)). This will provide data on how your players handle other forms of being the underdog, how they handle planning / recon / whatever when it's obviously a required part of the game (as opposed to D&D, where "kick in the door" is a valid playstyle), and... lost my train of thought.

I'm not sure how your players handle other forms of "underdogism", to know if that form of BDH is for them.

That said, if you look at many (OP) Isekai characters, or the party I call my "BDH Party", they aren't BDHs because they're some weak underdog who has to roll with punches (often or ever), but explicitly because they don't (usually) even feel such punches, and just wade through such obstacles as though they were humans, trivially defeating things that lesser beings would struggle with. So being a BDH does not, in point of fact, require rolling with the punches, and they are certainly not required to actively have done so while "on screen" to be a BDH.

(Yes, yes, it almost certainly makes for a better movie or whatever if they do, but you're not making a movie, you're playing an RPG.)

Then, in the case of my BDH Party, after wading through the intended challenges like they were humans, struggling with things like "convincing the 'supposed to be friendly' village that we're the good guys allies a lesser evil than the one we're here to end".

Your players probably don't have your preferences of liking characters who never give up, and, even if they did, their depression doesn't let them roleplay through the "not giving up when bad things happen". At best, they can have high stats that let them roll "I don't give up" rolls, so that they never have to deal with bad things happening. You can make lemonade; they want to roll to not get lemons. Same core concept, completely opposite implementation in that one key area. Good luck wrapping your noodle around that one (unless, you know, you are built to have an easier time of it than I am, because people are different, dagnabbit!).

[Lacco]

Likewise, what sort of "rules" should I set for myself? And how do those avoid breaking verisimilitude?

I think I saw a post regarding this topic some time ago, but I'm not able to trace it back. I remember the main point: it's easy for a GM to forget that he plays effectively a hivemind (all individual NPCs can act in 100% organized way and communicate their intents telepathically and without errors), while the rest of the players are somewhere on the scale between bunch of pre-schoolers to a SWAT team.

If we are talking about rules to play the enemies - not only monsters - smart, but not too smart, the basic rule is to play the enemies according to their mentality and abilities. I tend to underplay their efficiency, mainly because I deal with projects and organizations in RL and therefore I know that most of the time, even the well-oiled machines break down because of petty bathroom squabbles and a night of a bad sleep.

I always attempt to play the enemy as if they were actual beings - unless we are talking uber-enemies (e.g. genius level tacticians). So if I have a bunch of kobolds, they will use basic tactics, but they will pile them up in several layers. Elite dwarven legionnaires will be highly organized and will never run - even their retreats will be organized. Bandits will mainly focus on causing chaos and mayhem, and if their ambush fails, they will run away.

So: smart play depends on the attributes of the actors. So a goblin group (my world, my goblins) will have certain 'smarts' - they will be savage and will try to overwhelm, but may use some hit & run tactics. However, they will not keep rotating guard posts, won't go for complex traps and their idea of ambush is 'Groik yells "ATTTAAAACK" and we run and kill them". They will also run if their attack does not work well - and if more than half starts to run, they all run. Except for the one-eyed guy deep in the front who did not notice.

Monsters that are closer to animals are a different thing: they will mostly rely on instinct and learned behaviour.

Applying that to my GMing leads to a non-written set of rules:

• Enemy is as smart as his character sheet tells me (even if I do extrapolate it).
• A smart NPC may always make a stupid decision when in high-stress situation such as battle.
• Allies may easily become enemies on daily basis.
• There are no walkie-talkies in a dungeon.
• If in doubt, check the ability to notice/hear stuff.
• Morale is a thing.
• Even if the enemy has a plan, one cog will fail.
• Elite enemies have a backup plan even if one cog fails.

So some of the fights look like real mess: the enemies shout at each other, bicker, don't relentlessly attack (because they are waiting for their friends to jump in) and don't always make tactical or sensible choices.

On the other side, a smart enemies will play to their strengths, will plan, will communicate, will even stop to think about the situation, and will - sometimes - mess up.

Uber-tacticians will be ready for anything the players throw at them unless it genuinely surprises me - and even then they may have a response at hand unless they run out of resources or are unable to foil the players.

These also usually help avoid breaking verisimilitude at my table.

Also really curious about how players can get intelligence regardless of their build or proclivities without stripping away all immersion and mystery by just delivering tons of forced exposition.

Intelligence in megadungeon? Assuming we are talking about collecting clues about the enemy, there are some clues that can be provided without players using skills.

If there are several dozens of human skulls (with helmets) on pikes in front of the goblin encampement, they may assume the goblins were able to take out some folks. If there is a long hallway filled with human, dwarven and elven skulls, with some relatively fresh heads in the end, covered in goblin markings, they may take it as more serious warning.

Bloodstains, broken arrows and scorches may show them this spot is used for ambush and may even give them clue what to expect.

Hearing shouting, bickering and drunk singing may tell them the discipline is not the main focus.

Dead bodies tell very detailed tales about the thing that happened to them.

Do you mean intelligence like this? Or did you mean something different?

[Vahnavoi]

@Lacco: sounds like you're referring to a post by Icefractal, I forgot the thread. There's been discussion on the same topic in this thread. Anyways, out of your rules, there is one that needs amending: "there are no walkie talkies in a dungeon". There are in fact several fictional conceits that could give either players or enemies a functionally equivalent, from telepathy to messenger familiar, even if you're not giving anyone a literal radio.

A better rule would be, "keep in mind how communication is done". This cuts both ways. If player characters are given sufficient communication channels, that will go a long way to answer how they can gain intelligence of a dungeon. It allows for new tactics, like splitting to scout ahead, without it being an automatic fail on their part.

[gbaji]

If we are talking about rules to play the enemies - not only monsters - smart, but not too smart, the basic rule is to play the enemies according to their mentality and abilities. I tend to underplay their efficiency, mainly because I deal with projects and organizations in RL and therefore I know that most of the time, even the well-oiled machines break down because of petty bathroom squabbles and a night of a bad sleep.

I always attempt to play the enemy as if they were actual beings - unless we are talking uber-enemies (e.g. genius level tacticians). So if I have a bunch of kobolds, they will use basic tactics, but they will pile them up in several layers. Elite dwarven legionnaires will be highly organized and will never run - even their retreats will be organized. Bandits will mainly focus on causing chaos and mayhem, and if their ambush fails, they will run away.

So: smart play depends on the attributes of the actors. So a goblin group (my world, my goblins) will have certain 'smarts' - they will be savage and will try to overwhelm, but may use some hit & run tactics. However, they will not keep rotating guard posts, won't go for complex traps and their idea of ambush is 'Groik yells "ATTTAAAACK" and we run and kill them". They will also run if their attack does not work well - and if more than half starts to run, they all run. Except for the one-eyed guy deep in the front who did not notice.

Yup. Important for the GM to actually "roleplay" the NPCs based on their actual capabilities. It's probably one of the most difficult things for GMs to learn to do though, since it almost always involves the GM actively *not* acting on information he knows, so as to lower the capabilities of the NPCs. But one of the worse things for a game is if/when the players notice the NPCs acting in ways that they should not, based on them having information they should not. If you have players regularly asking "how did that NPC know to do X?", you are probably failing at this to some degree.

I also vary the amount of coordination of my NPCs based on training. My players can often tell, just from how I'm running the NPCs in a combat, whether this is a "random group of enemies" or "well trained squad who fight/train together regularly", just by how they move and coordinate their actions. It's one of those aspects of an NPC encounter that isn't going to appear on a stat block, but adds a ton of realism to the encounter itself. And also a way to make things that maybe look extremely diffuclt turn out to be a lot easier (or at least less difficult) when the PCs actually engage.

Discovering part way in a fight that the bad guys aren't covering eachother properly, and/or sometimes leave gaps undefended, or seem confused about who is supposed to be defending where, makes a huge difference in the actual difficulty of the fight for the PCs. Sometimes, you can include humorous bits in there as well. PCs are advancing, but there's a pit trap in front of them. Some of the enemies charge forward to attack the PCs, while one of them runs back and pulls the lever for the pit trap, and.... a few of their own guys fall because in his panic, he didn't look to see who was actually on the stretch of flooring, and the other guys in their panic didn't realize that they were charging across one of their own defenses to attack the party. This sort of thing serves both to whittle down the defenders a bit *and* allow the GM to present something that would have been a logical and very nasty defense (and perhaps warn the players to keep an eye out for stuff like this further in), without actually springing it on them.

[GloatingSwine]

Likewise, what sort of "rules" should I set for myself? And how do those avoid breaking verisimilitude?

The rules need to be things like:

What will group A try doing in response to an attack, how good are they at it, how long will can they try and keep it up, what will they do then? And you need to have that all prewritten and never change it no matter what your players do, because this is the character of the NPCs you're playing and the material limitations of their situation.

And remember, your job isn't to beat your players it's to challenge them. Instead of working as hard as you can to make their strategy fail, make them work harder to make it succeed.

If their responses are going to include things like co-operation with other groups every group you introduce needs a prewritten diplomatic situation, who it likes, who it doesn't like, who it will and won't fight with, and all of these need to be written down before the players meet them and not change until the players interact in a way that changes them. This needs to be relatively common knowledge among the denizens and you need to at least offer the basics without the players asking (because they're not going to and you know it.) So when your players were at least talking to the Kobolds you need one to tell them what the basic diplomatic state of the area is.

Also also: This is the start of the dungeon, you shouldn't be pulling out all the stops (fortifications and ambushes and alliances) at the start. These groups should have simpler responses that they're worse at doing. Like have your kobolds throw up some basic fortifications, and if the players continue to hit and run have them eventually run out of the ability to do so. They don't have infinite resources, they can't build back stronger every time. If you want your players to suffer from not being decisive enough have them come back one day and find that the Kobolds have cleared out and they get less for winning than they would have otherwise, but clearly describe that something is changing in their behaviour. 2 or so encounters before this make it clear to the players that the enemy seems to now be buying time instead of fighting to repel them. (You may need to make it even more explicit on the first few "they look like they're clearing out, if you don't push through now you'll lose your chance". This is because your players have +100 to save vs. hints)

[Talakeal]

So my players have decided not to quit, but they did come up with a new plan.

Rather than going into the dungeon themselves, they are going to sit in safe areas and summon incorporeal creatures to clear out the dungeon for them. This is smart (I have seen plenty of solutions for everything from the Tarrasque to Tucker's kobolds that boil down to "Summon Allip. Wait. Loot.") but isn't exactly fun game play.

Clearly, there are many things in the dungeon that can deal with incorporeal foes, but the majority have no way to harm them.

Any idea how I should arbitrate this?

[Amidus Drexel]

So my players have decided not to quit, but they did come up with a new plan.

Rather than going into the dungeon themselves, they are going to sit in safe areas and summon incorporeal creatures to clear out the dungeon for them. This is smart (I have seen plenty of solutions for everything from the Tarrasque to Tucker's kobolds that boil down to "Summon Allip. Wait. Loot.") but isn't exactly fun game play.

Clearly, there are many things in the dungeon that can deal with incorporeal foes, but the majority have no way to harm them.

Any idea how I should arbitrate this?

The tricky thing here is balancing "rewarding your players for a smart solution" and "keeping the game fun for all involved". It should definitely work for the enemies they've already scouted that have no defenses against this kind of attack - that's the reward for their smart plan. (Ones that haven't been scouted might be taken by surprise or might have adequate defenses - the players have no way of knowing ahead of time. It's definitely a cool callback to occasionally have them find something that's already been killed by their advance force of summons).

Since you're playing a custom system (iirc) - depending on how your system works - if incorporeal creatures multiply when they kill things, there might be more dangerous (but less organized) spawn creatures left in the area. Presumably if your players can summon them, they can handle some number of them in a fight. For your players in particular, you might want to remind them that created spawn won't necessarily be under their personal control, and about any lore or setting-level consequences for these kinds of actions (corruption, alignment shift, bad reputation, attention of evil entities, etc.).

If incorporeal creatures in your system behave like incorporeal undead in older D&D editions, you should probably avoid ambushes with more than 1 or 2 enemies - any stat-draining or paralyzing creature in multiples quickly creates a death spiral once it succeeds on even a single attack (assuming it survives a round of combat). Let the party spot them well ahead of time at least twice before they get jumped by a single shadow/wraith/etc. later on, and do your best to telegraph ambushes.

[GloatingSwine]

Any idea how I should arbitrate this?

Have most enemies run away from it until the players get bored of doing it.

[Quertus]

So my players have decided not to quit, but they did come up with a new plan.

Rather than going into the dungeon themselves, they are going to sit in safe areas and summon incorporeal creatures to clear out the dungeon for them. This is smart (I have seen plenty of solutions for everything from the Tarrasque to Tucker's kobolds that boil down to "Summon Allip. Wait. Loot.") but isn't exactly fun game play.

Clearly, there are many things in the dungeon that can deal with incorporeal foes, but the majority have no way to harm them.

Any idea how I should arbitrate this?

Have the incorporeal creatures kill everything that has no defense against them. Done. Having discouraged brave play, reap what you sowed, and give your players a well-deserved win.

Now, note that "running away" technically is a defense against incorporeal creatures... but that may well just result in "100 goblin corpses were found next to 50 gnoll corpses at the edge of the gnoll camp, then the remaining 50 gnolls were killed by the incorporeal creatures".

Actually roleplay (gasp!) the magadungeon's inhabitants, have them do what they've already done, and ally with one another against this threat... only to still be killed by it.

Let your players learn all about how the magedungeon operates by all the corpses they see as they loot their way through the megadunegon.

Then - after you've beaten these lessons home over and over with detailed reports of corpses they've looted - make them actually use that knowledge when they come across foes that were able to kill the incorporeal creatures. Or hide from them (darn Mimics). Or whatever.

[Talakeal]

The tricky thing here is balancing "rewarding your players for a smart solution" and "keeping the game fun for all involved". It should definitely work for the enemies they've already scouted that have no defenses against this kind of attack - that's the reward for their smart plan. (Ones that haven't been scouted might be taken by surprise or might have adequate defenses - the players have no way of knowing ahead of time. It's definitely a cool callback to occasionally have them find something that's already been killed by their advance force of summons).

Since you're playing a custom system (iirc) - depending on how your system works - if incorporeal creatures multiply when they kill things, there might be more dangerous (but less organized) spawn creatures left in the area. Presumably if your players can summon them, they can handle some number of them in a fight. For your players in particular, you might want to remind them that created spawn won't necessarily be under their personal control, and about any lore or setting-level consequences for these kinds of actions (corruption, alignment shift, bad reputation, attention of evil entities, etc.).

If incorporeal creatures in your system behave like incorporeal undead in older D&D editions, you should probably avoid ambushes with more than 1 or 2 enemies - any stat-draining or paralyzing creature in multiples quickly creates a death spiral once it succeeds on even a single attack (assuming it survives a round of combat). Let the party spot them well ahead of time at least twice before they get jumped by a single shadow/wraith/etc. later on, and do your best to telegraph ambushes.

Elementals, not self replicating undead.

Have the incorporeal creatures kill everything that has no defense against them. Done. Having discouraged brave play, reap what you sowed, and give your players a well-deserved win.

Now, note that "running away" technically is a defense against incorporeal creatures... but that may well just result in "100 goblin corpses were found next to 50 gnoll corpses at the edge of the gnoll camp, then the remaining 50 gnolls were killed by the incorporeal creatures".

Actually roleplay (gasp!) the magadungeon's inhabitants, have them do what they've already done, and ally with one another against this threat... only to still be killed by it.

Let your players learn all about how the mage-dungeon operates by all the corpses they see as they loot their way through the megadunegon.

Then - after you've beaten these lessons home over and over with detailed reports of corpses they've looted - make them actually use that knowledge when they come across foes that were able to kill the incorporeal creatures. Or hide from them (darn Mimics). Or whatever.

You lost me.

And yes, RPing the monsters is the core of this. But this requires a long solo game, I was wondering if there is any way to do it reasonably without having an entire series of solo sessions to myself.

What discouraged brave play? All of their failures are (so far) caused by running away from a threat rather than dealing with it.

What did they do to make them "well-deserving" of a win? And why do their previous forty odd victories not count?

What lessons are there to be learned from the corpses? I mean, subtle hints sure, but nothing that the PCs are likely to pick up on.

You really think everything should just die? I mean, a single incorporeal creature is going to take a LOOOOOOONG time to hunt down and kill everything in the dungeon, and there is plenty of stuff in there that can kill or banish it. Likewise, won't other monsters loot all of the corpses?

[GloatingSwine]

What discouraged brave play? All of their failures are (so far) caused by running away from a threat rather than dealing with it.

Are they?

In your campaign diary thread the first and second expeditions they returned to town when they were, in your words, pretty beat up and in need of rest, and the third and fourth both seem to have gone south due to a miscast/wild surge from the wizard gooning someone unexpectedly. So that's 2 where they seem to have done as you wanted, 2 where they lost because of bad dice when the wizard TKOd herself, one where they assessed they were outmatched (and the wizard gooned someone with a miscast) and one where they got the two prong ambush (that apparently you know they are bad at dealing with so should probably tone down).

From the campaign diary I'm having a hard time tracking whether they're making progress when they score victories. They win encounters, but they only seem to win the ones that don't matter. Any time they encounter anything that seems significant in the diary they get krumped. Twice by their own wizard.

It very much seems from the campaign diary that the party is simply outmatched quite a lot of the time, and their retreats are happening because the fights actually are going badly for them.

[Talakeal]

Are they?

In your campaign diary thread the first and second expeditions they returned to town when they were, in your words, pretty beat up and in need of rest, and the third and fourth both seem to have gone south due to a miscast/wild surge from the wizard gooning someone unexpectedly. So that's 2 where they seem to have done as you wanted, 2 where they lost because of bad dice when the wizard TKOd herself, one where they assessed they were outmatched (and the wizard gooned someone with a miscast) and one where they got the two prong ambush (that apparently you know they are bad at dealing with so should probably tone down).

From the campaign diary I'm having a hard time tracking whether they're making progress when they score victories. They win encounters, but they only seem to win the ones that don't matter. Any time they encounter anything that seems significant in the diary they get krumped. Twice by their own wizard.

It very much seems from the campaign diary that the party is simply outmatched quite a lot of the time, and their retreats are happening because the fights actually are going badly for them.

Honestly, I think the game is going fine.

The game is balanced around ~5 encounters per day, and so far they have cleared 35 encounters and had seven adventuring days, so they are right on track.

They aren't really "making progress" yet as they haven't really gotten their feet under them and are just kind of wandering around the dungeon rather than committing to one direction, but at this point they are still earning XP and treasure, and have mapped ~80% of the first floor, and they certainly aren't struggling to survive.

I don't really think they are being cowardly (this time), but I also wouldn't say they are being discouraged from being brave.

As you point out, the defeat at the hands of the olags was pretty random, a chase of wild magic being wild magic, and I would have probably fallen back at this point as well.

The only real failure on their part was not thinking to block the doors to the ghasts or adjust their tactics in the rematch.