Bend, dont break.

[Sunseeker]

So if you think about the individual encounter, any trade off of frequency for power is going to look very powerful (which is part of why 4e standardized this across the classes). But if you look at the entire adventure, and balance and pace your game around that, it's a more viable balance, since it forces you to continue even after you've nova'd.

True, but the flaw in this argument has been pointed out many, many times. Not all adventures run to completion, and many take a very, very very long time to come to an end.

For a game that's going to run once a week, for 6 hours a night, and last for 6 months(ie: lots of content over a short period of time), looking at performance over the entire game is fine.

For a game that's going to run once every other month, for 2-3 hours, and last a decade or longer, such thinking can fall apart very quickly.

 

[Umbran]

So if you think about the individual encounter, any trade off of frequency for power is going to look very powerful (which is part of why 4e standardized this across the classes). But if you look at the entire adventure, and balance and pace your game around that, it's a more viable balance, since it forces you to continue even after you've nova'd.

Adventure based design has one paperwork flaw - if your session runs shorter than one adventure, you have to maintain your resource-use tracking between sessions. Some of us only run every couple of weeks, so memory often serves poorly.

Adventure based design also tends to deal poorly with a party that decides to retreat and refuel mid-adventure. That's the "15-minute workday" phenomenon, and when *every* adventure has to have caveats for dealing with that scenario, things often start looking a little railroady.

 

[Rhenny]

Unless you only need to do it once or twice per day. And why would the rogue not try to memorise and prepare a few spells that would help them be a rogue.

If the DM is cool with it, the rogue has high enough intelligence, and in the campaign story there is a plausible way for the rogue to learn a little magic, heck, I'm all for it. The rogue learning how to cast mage hand, spider climb, invisibility (or other spells) will not step on the wizard's toes.

I think that magic use for non-casters should be an option in D&DNext. It doesn't have to be the core.

 

[I'm A Banana]

Not all adventures run to completion, and many take a very, very very long time to come to an end.

And the flaw in this argument is that one assumes you can't define an "adventure." Of course you'd have to define an adventure for adventure-based design, just as you define an encounter for encounter-based design.

You just need to define some parameters. It was the first thing I did, above.

The "adventure" is what takes place between full party recharges. This is what adventure-based means: your resources are dedicated to beating the adventure. If you were to design 4e as adventure-based, you'd design the game over the course of 2 or 3 encounters. Milestones and healing surges and daily powers are already "adventure-based" in this rubrick.

Above, I assume that this holds true for a 5e adventure, too. It's 3 4e-style encounters long, and it involves one "encounter" per pillar. Each "encounter" consists of one obstacle (monster, NPC, hazard, etc.) per PC.

For a game that's going to run once every other month, for 2-3 hours, and last a decade or longer, such thinking can fall apart very quickly.

Adventure based design has one paperwork flaw - if your session runs shorter than one adventure, you have to maintain your resource-use tracking between sessions. Some of us only run every couple of weeks, so memory often serves poorly.

If 5e meets its goal of having a possible adventure in an hour, then everyone's happy.

And because this balance responds well to maths, you can shorten it significantly without compromising the balance, like I pointed out above.

I think shooting for a "one adventure per your gaming session, however long that is" is a reasonable and very realistic goal for 5e.

Adventure based design also tends to deal poorly with a party that decides to retreat and refuel mid-adventure. That's the "15-minute workday" phenomenon, and when *every* adventure has to have caveats for dealing with that scenario, things often start looking a little railroady.

Like I said above, the easiest way to deal with this is to assume the "adventure" recharges when the party does. The party rests for the night, and the goblins call in reinforcements, the kobolds fix their traps (and maybe invent new ones), and the dragon changes lairs. This requires no real bookkeeping, you just run the party through its tasks again, maybe even changing it up. It's a lot like, in an encouter-based game, when the party runs away, the monsters return to full HP. Only you're applying it to the whole adventure, rather than just an encounter.

These "problems" aren't actually problems. And you still manage to gain a vancian spellcaster who prepares a few powerful spells at the beginning of the day and who uses them to great effect, but who cannot complete the adventure with them, and so needs the help of his teammates, just as they will need his help (such as when the rogue fails to unlock the door, or fails to convince the king). Preserving that feel for the wizard is vital, and designing the game around the adventure opens up options like death attacks, charm/diplomacy, exploration, deadly traps, rot grubs, ear seekers, save-or-die effects, hirelings and companions, and a host of other very positive things.

 

[Umbran]

If 5e meets its goal of having a possible adventure in an hour, then everyone's happy.

Having what I consider to be an adventure in one hour is not possible. So, I'll withhold judgement until I see what they're calling an adventure before I accept that I'll be happy with it. Their "adventure" may be what I call an "encounter" - and now we are really back to encounter-based design with the names changed.

Like I said above, the easiest way to deal with this is to assume the "adventure" recharges when the party does.

And, like I said above, that starts getting a little railroady. While it makes sense that the opponent can recover some resources (like hit points) as the party does, if there isn't a source of new goblins, actually replacing the dead is not usually particularly plausible.

Here's the basic problem that these discussions often seem to miss - in the real world, conflicts and battles are often won by attrition. The 15 minute workday is the game-equivalent of a real-world tactic. It works, it is intuitive. The problem is that it makes for a pretty lame action adventure story.

4e addressed the problem with encounter-based design, reducing the effectiveness of fighting a battle of attrition, by reducing the number of resources that are long-term issues.

These "problems" aren't actually problems.

Sorry, no. See above. What you've done is sweep the issues into the corners, where they now pile up until you deal with them.

 

[I'm A Banana]

Having what I consider to be an adventure in one hour is not possible.

This is just a semantic point. I defined my term above. An adventure might happen over about 3 4e-style encounters. If you want to quibble about word choice, that's not really relevant to the broader point. Call it a smerp if you want, and then we have smerp-based design. Blah blah blah semantics.

And, like I said above, that starts getting a little railroady. While it makes sense that the opponent can recover some resources (like hit points) as the party does, if there isn't a source of new goblins, actually replacing the dead is not usually particularly plausible.

Which is why I pointed out that a longer time frame makes more in-fiction sense for this. But my focus wasn't on the time frame. If you stipulate than an extended rest is a week or a day or an hour or every 30 seconds or a dynastic generation or whatever, the mechanics function. If the party recharges, so does the threat.

A static threat that doesn't react when the party rests is going to have the "lets rest and recover and nova again" problem regardless of adventure or encounter focus. It's just not very good game design, period. A goblin tribe who just sits in their lair until the PC's show up to kill them and never flees, hires reinforcements, repairs, builds traps, etc., is a gameplay problem, period. "This doesn't work in the face of a static challenge!" is a kind of a strawman argument: a static challenge doesn't work, period.

And thirdly, I would really dispute the "railroading" charge. Railroading happens when there is a lack of meaningful choice, and restoring the challenge when the PC's get restored doesn't remove meaningful choice -- they can still choose other challenges, or employ different tactics next time, or swap out their rogue for an assassin or whatever. Just because they ran away from the dragon (who then switched to a backup lair) doesn't mean they can't go slay the orcs instead.

Here's the basic problem that these discussions often seem to miss - in the real world, conflicts and battles are often won by attrition. The 15 minute workday is the game-equivalent of a real-world tactic. It works, it is intuitive. The problem is that it makes for a pretty lame action adventure story.

Which is why the game-design solution is to restore the equilibrium. You gain all your resources back? So do your enemies. Reset button hit. Try again.

Furthermore, this paradigm allows to you model siege tactics fairly well. If you prevent one side from regaining their resources (prevent the party from resting, prevent the goblins from hiring reinforcements, prevent the dragon from switching lairs), it's easier to employ these rest-nova-rest tactics. That's a bit of an advanced trick, certainly not something I'd expect most D&D games to deal with, but it does become possible in a way that wasn't there before.

Sorry, no. See above. What you've done is sweep the issues into the corners, where they now pile up until you deal with them.

Are we having a conversation, or are you trying to win a debate? If the party retreated from an encounter in an encounter-based design game (like 4e) and then returned, would you keep the same conditions, statuses, damage, and dead monsters as they had when they last left?

If not, then there should be no tremendous problem in accepting that the same principle be applied to adventures in an adventure-based game.

If you accept that principle, then we just need to find a way to justify that explanation in the game world that satisfies you. If you can't think of a reason that the adventure might regain or change its challenges on your own, I've noted a few above (changing the timescale, changing lairs, rebuilding traps, getting reinforcements) and there's many more where they came from (breeding pools, planar gates, zombie reanimation, newly conscripted noncombatants, hired guns, etc.). Finding one or a combination that helps explain it shouldn't be any harder than explaining how the goblin I stabbed five minutes ago in a combat I ran away from is back at full hit points now that we're starting a new combat.

But if you're unwilling to accept these explanations and reasons and causes for why the adventure would recharge, I don't know what else I can do for you. If your challenge isn't reactive, encounter or adventure, you have a problem. In 4e, if you unload your dailies and encounters, and then flee, and come back after an extended rest, do the damage and effects remain? If you accept that it doesn't have to happen in an encounter based game, I don't know why you would say that it must happen in an adventure-based game.

 

[Umbran]

This is just a semantic point.

No, it really isn't. "Adventure" has a meaning in and of itself, separate from gaming. It also has a traditional meaning within gaming. If your new "adventure" doesn't match those other meanings well, you're making up new jargon - which would make you the one calling it a smerp, not me.

Which is why I pointed out that a longer time frame makes more in-fiction sense for this. But my focus wasn't on the time frame. If you stipulate than an extended rest is a week or a day or an hour or every 30 seconds or a dynastic generation or whatever, the mechanics function. If the party recharges, so does the threat.

That is a fine solution, if all you care about is arbitrary game rules. My adventures also have to make sense as a piece of fiction. And that's where we have an issue.

And thirdly, I would really dispute the "railroading" charge. Railroading happens when there is a lack of meaningful choice, and restoring the challenge when the PC's get restored doesn't remove meaningful choice

Yes it does. The party makes a tactical choice to retreat and recharge - it is a meaningless choice, because the encounter remains the same no matter how they approach it. It is recharged.

Furthermore, this paradigm allows to you model siege tactics fairly well.

Didn't you just say, just a few sentences before, "if the party recharges, so does the threat"? That's the exact opposite of a siege situation, isn't it? So now sometimes the threat recharges, and sometimes it doesn't? Maybe I missed something.

Are we having a conversation, or are you trying to win a debate?

I am presenting an issue I see with what looked like a too-simple mode of dealing with encounter/adventure design.

If the party retreated from an encounter in an encounter-based design game (like 4e) and then returned, would you keep the same conditions, statuses, damage, and dead monsters as they had when they last left?

No. But what I do with them is entirely situational, not generalized. It is not a simple, "if the party retreats, the threat recharges", which is what I thought you were suggesting.

If you accept that principle, then we just need to find a way to justify that explanation in the game world that satisfies you.

My point is that we cannot generalize this. "If the party retreats, the threat recharges," becomes implausible for the fiction in too many situations to be a root of the game design.

 

[I'm A Banana]

My point is that we cannot generalize this. "If the party retreats, the threat recharges," becomes implausible for the fiction in too many situations to be a root of the game design.

I get the impression that it might help if I step away from mechanical analysis and tell you what this might look like in play, since you seem to be under the misconception that this is somehow a problem that is not easily solved.

[sblock=Gameplay]
So, we have a stereotypical Party of Four. They meet at a tavern, and the shady old guy in the corner tells them that if they recover the MacGuffin that the goblins stole from him, they will be fabulously rewarded. The party goes up to the entrance to the goblin warren.

So, lets say the goblin warren has 4 goblin warriors (1 challenge for each character for Combat), 4 traps (1 challenge for each character for Exploration), and 4 kobold slaves (1 challenge for each character for Interaction: they can be turned into allies!). There's other noncombat goblins in the warren as well -- the warrens are, after all, the goblin's lair. Each goblin has 5 hp. Very simple.

Each character has a "Basic Skill" in each pillar. Basic Fighting (deals 1 hp of damage), Basic Exploring, Basic Talking. This comes from your background.

Fighters get a Power Attack ability they can use whenever they want to deal +3 damage on any attack. Rogues have the Broad Skills ability that gives them a +1 bonus to all their basic skills. Clerics have three spells: Cure Light Wounds, Find Traps, and Divination. Wizards have three spells: Acid Arrow, Knock, and Charm Person. These are vancian spells: single-use, prepared in advance.

So the party wants to get into the warren and get the MacGuffin and get back out. They meet a goblin standing guard, and the spellcasters want to to show off because they assume that as spellcasters they are the only ones necessary in the adventure. The wizard goes in with Charm Person on the goblin. It fails. The goblin gets to attack, pricks the wizard, and the wizard uses an Acid Arrow. Because the Acid Arrow is 5 x the power of a basic ability, it kills the goblin. Blarg. The cleric then heals the wizard. To prepare, the cleric uses Divination to find out what will happen when they go through the front door, and finds out that there's a trap on the other side of the door. So the cleric uses Find Traps to locate the trap, and uses the thief's Basic Exploration skill (5 checks) to try to disarm it.

So, the party has dealt with 1 goblin and 1 trap (out of 4 each), and has used up all of their spells. A "nova" by any definition of the term.

Because the spellcasters assume they are the only important members of the party, they say, "Hey, lets rest now, there's probably more than one goblin in there." They go back to town, and rest at the inn. The fighter calls them pansies.

While they do that, the goblins discover that their sentry is dead, his face melted off by acid. They say, "We've been found!" Now they "recharge." Possible options include:

• Make your kobold slaves build traps through the night. They can build two traps. This adds two more traps, bringing the total smerp challenge to 5 traps, 3 goblins, and 4 kobold slaves.
• Grab the wolves that've been hanging out near the larder. They tie them down near the entrance and poke them with sticks. This adds two more combat encounters, and brings the smerp to 3 traps, 3 goblins, 4 kobold slaves, and two really angry wolves.
• Ask a noncombat messenger to run over to the next hill and tell the goblin king that they've been attacked. Two more goblins show up by morning, and brings the smerp to 5 goblins, 3 traps, 4 kobold slaves.
• Activate the MacGuffin, which summons and binds a demon into servitude. The demon is an elite creature. This brings the total to 3 goblins, 1 elite demon, 4 kobold slaves, and 3 traps.
• Give eight of the noncombatants swords and put them at the entrance. They are minions. This brings the total to 3 goblins, 8 minions, 4 kobold slaves, and 3 traps.
• Uproot everyone and go to the Backup Lair, where they have a buddy and a few traps already there ready. This brings the total back to 4 goblins, 4 traps, and 4 kobold slaves.

...or really whatever else the DM can imagine. Eight hours or so is not a trivial amount of time.

Now the party comes back the next day, and they're facing...pretty much the same situation they were facing the night before, mechanically. The adventure has recharged, just as the party has.
[/sblock]

The party retreats. The threat recharges. It makes sense in context in a million different ways in a thousand different contexts. No matter what, the nova is not necessarily a winning strategy (though it is a strategy that may change the dynamic -- some of those options locate more in the Combat pillar or more in the Exploration pillar than before...hope that wizard prepares more Acid Arrows...and the demon or the wolves might turn into an Interaction challenge if the party wants to offer some souls or has a druid or something). There are more challenges present than spells can solve.

Does that help dispel your notion that somehow a recharging adventure is absurd or implausible?

If not, I can probably do the same thing for almost any scenario you can derive. I bet you can, too.

 

[Arlough]

[sblock]This is just a semantic point. I defined my term above. An adventure might happen over about 3 4e-style encounters. If you want to quibble about word choice, that's not really relevant to the broader point. Call it a smerp if you want, and then we have smerp-based design. Blah blah blah semantics.



Which is why I pointed out that a longer time frame makes more in-fiction sense for this. But my focus wasn't on the time frame. If you stipulate than an extended rest is a week or a day or an hour or every 30 seconds or a dynastic generation or whatever, the mechanics function. If the party recharges, so does the threat.

A static threat that doesn't react when the party rests is going to have the "lets rest and recover and nova again" problem regardless of adventure or encounter focus. It's just not very good game design, period. A goblin tribe who just sits in their lair until the PC's show up to kill them and never flees, hires reinforcements, repairs, builds traps, etc., is a gameplay problem, period. "This doesn't work in the face of a static challenge!" is a kind of a strawman argument: a static challenge doesn't work, period.

And thirdly, I would really dispute the "railroading" charge. Railroading happens when there is a lack of meaningful choice, and restoring the challenge when the PC's get restored doesn't remove meaningful choice -- they can still choose other challenges, or employ different tactics next time, or swap out their rogue for an assassin or whatever. Just because they ran away from the dragon (who then switched to a backup lair) doesn't mean they can't go slay the orcs instead.



Which is why the game-design solution is to restore the equilibrium. You gain all your resources back? So do your enemies. Reset button hit. Try again.

Furthermore, this paradigm allows to you model siege tactics fairly well. If you prevent one side from regaining their resources (prevent the party from resting, prevent the goblins from hiring reinforcements, prevent the dragon from switching lairs), it's easier to employ these rest-nova-rest tactics. That's a bit of an advanced trick, certainly not something I'd expect most D&D games to deal with, but it does become possible in a way that wasn't there before.



Are we having a conversation, or are you trying to win a debate? If the party retreated from an encounter in an encounter-based design game (like 4e) and then returned, would you keep the same conditions, statuses, damage, and dead monsters as they had when they last left?

If not, then there should be no tremendous problem in accepting that the same principle be applied to adventures in an adventure-based game.

If you accept that principle, then we just need to find a way to justify that explanation in the game world that satisfies you. If you can't think of a reason that the adventure might regain or change its challenges on your own, I've noted a few above (changing the timescale, changing lairs, rebuilding traps, getting reinforcements) and there's many more where they came from (breeding pools, planar gates, zombie reanimation, newly conscripted noncombatants, hired guns, etc.). Finding one or a combination that helps explain it shouldn't be any harder than explaining how the goblin I stabbed five minutes ago in a combat I ran away from is back at full hit points now that we're starting a new combat.

But if you're unwilling to accept these explanations and reasons and causes for why the adventure would recharge, I don't know what else I can do for you. If your challenge isn't reactive, encounter or adventure, you have a problem. In 4e, if you unload your dailies and encounters, and then flee, and come back after an extended rest, do the damage and effects remain? If you accept that it doesn't have to happen in an encounter based game, I don't know why you would say that it must happen in an adventure-based game.[/sblock]

You may be interested in The Angry DM's Project Slaughterhouse.

[sblock]I get the impression that it might help if I step away from mechanical analysis and tell you what this might look like in play, since you seem to be under the misconception that this is somehow a problem that is not easily solved.

[sblock=Gameplay]
So, we have a stereotypical Party of Four. They meet at a tavern, and the shady old guy in the corner tells them that if they recover the MacGuffin that the goblins stole from him, they will be fabulously rewarded. The party goes up to the entrance to the goblin warren.

So, lets say the goblin warren has 4 goblin warriors (1 challenge for each character for Combat), 4 traps (1 challenge for each character for Exploration), and 4 kobold slaves (1 challenge for each character for Interaction: they can be turned into allies!). There's other noncombat goblins in the warren as well -- the warrens are, after all, the goblin's lair. Each goblin has 5 hp. Very simple.

Each character has a "Basic Skill" in each pillar. Basic Fighting (deals 1 hp of damage), Basic Exploring, Basic Talking. This comes from your background.

Fighters get a Power Attack ability they can use whenever they want to deal +3 damage on any attack. Rogues have the Broad Skills ability that gives them a +1 bonus to all their basic skills. Clerics have three spells: Cure Light Wounds, Find Traps, and Divination. Wizards have three spells: Acid Arrow, Knock, and Charm Person. These are vancian spells: single-use, prepared in advance.

So the party wants to get into the warren and get the MacGuffin and get back out. They meet a goblin standing guard, and the spellcasters want to to show off because they assume that as spellcasters they are the only ones necessary in the adventure. The wizard goes in with Charm Person on the goblin. It fails. The goblin gets to attack, pricks the wizard, and the wizard uses an Acid Arrow. Because the Acid Arrow is 5 x the power of a basic ability, it kills the goblin. Blarg. The cleric then heals the wizard. To prepare, the cleric uses Divination to find out what will happen when they go through the front door, and finds out that there's a trap on the other side of the door. So the cleric uses Find Traps to locate the trap, and uses the thief's Basic Exploration skill (5 checks) to try to disarm it.

So, the party has dealt with 1 goblin and 1 trap (out of 4 each), and has used up all of their spells. A "nova" by any definition of the term.

Because the spellcasters assume they are the only important members of the party, they say, "Hey, lets rest now, there's probably more than one goblin in there." They go back to town, and rest at the inn. The fighter calls them pansies.

While they do that, the goblins discover that their sentry is dead, his face melted off by acid. They say, "We've been found!" Now they "recharge." Possible options include:

• Make your kobold slaves build traps through the night. They can build two traps. This adds two more traps, bringing the total smerp challenge to 5 traps, 3 goblins, and 4 kobold slaves.
• Grab the wolves that've been hanging out near the larder. They tie them down near the entrance and poke them with sticks. This adds two more combat encounters, and brings the smerp to 3 traps, 3 goblins, 4 kobold slaves, and two really angry wolves.
• Ask a noncombat messenger to run over to the next hill and tell the goblin king that they've been attacked. Two more goblins show up by morning, and brings the smerp to 5 goblins, 3 traps, 4 kobold slaves.
• Activate the MacGuffin, which summons and binds a demon into servitude. The demon is an elite creature. This brings the total to 3 goblins, 1 elite demon, 4 kobold slaves, and 3 traps.
• Give eight of the noncombatants swords and put them at the entrance. They are minions. This brings the total to 3 goblins, 8 minions, 4 kobold slaves, and 3 traps.
• Uproot everyone and go to the Backup Lair, where they have a buddy and a few traps already there ready. This brings the total back to 4 goblins, 4 traps, and 4 kobold slaves.

...or really whatever else the DM can imagine. Eight hours or so is not a trivial amount of time.

Now the party comes back the next day, and they're facing...pretty much the same situation they were facing the night before, mechanically. The adventure has recharged, just as the party has.
[/sblock]

The party retreats. The threat recharges. It makes sense in context in a million different ways in a thousand different contexts. No matter what, the nova is not necessarily a winning strategy (though it is a strategy that may change the dynamic -- some of those options locate more in the Combat pillar or more in the Exploration pillar than before...hope that wizard prepares more Acid Arrows...and the demon or the wolves might turn into an Interaction challenge if the party wants to offer some souls or has a druid or something). There are more challenges present than spells can solve.

Does that help dispel your notion that somehow a recharging adventure is absurd or implausible?

If not, I can probably do the same thing for almost any scenario you can derive. I bet you can, too.[/sblock]

The way you are saying things, it seems that you are just changing "Encounter" to "Adventure", but I will assume this discussion is not so hung up over a simple name change, so I am going to take a stab at another idea here. Let me know if I am on the right track.

Encounter Powers refresh in between encounters (short rest), but encounters themselves are similar to Daily Powers and will refresh with every extended rest. So if you want to complete the Adventure, you must complete it in 1 day (or at least, before you take an extended rest.)

The problem with this is that Adventures (as defined above) are very limited in possible scope. I can only see this working if adventures are built in dungeon/castle assault style.
When I say that, I accept that the dungeon/castle you may be assaulting could be a social network of the criminal-underworld/upper-class-politics as much as it could literally be a dungeon crawl or castle invasion.
But basically, you are always going around attacking established and/or stationary enemy encampments and societies. Basically, your only choice is to be marauders.

P.S. Another Angry DM article that seems relevant to what we are talking about.

 

[I'm A Banana]

Encounter Powers refresh in between encounters (short rest), but encounters themselves are similar to Daily Powers and will refresh with every extended rest. So if you want to complete the Adventure, you must complete it in 1 day (or at least, before you take an extended rest.)

This sounds about right, by way of analogy. For this spellcasting, you basically have 3 powers, each of which can be used once per day. If you have 3 encounters, then they are equal to "encounter" powers (since you are going to use one in each encounter) that you just choose at the start.

There's one point I want to tease out, though.

Like was talked about above, an "extended rest" need not mean at the end of the day. Mechanically, we just want it to be completed before you take an extended rest. If that extended rest is a night's sleep, then that day. If that extended rest is a week's vacation, then before you take that week off. If the extended rest is "the winter" or "between aeons," then that. If the extended rest is 5 minutes, then before you take that break.

The period of time is pretty arbitrary. It doesn't matter for the mechanics. You can set it at whatever you want. Extended Rests happen at the end of the day in 4e, and I can see a case for keeping it and it gives us some shared ground, but if a night's sleep was a short rest and a week's rest was an extended rest, then that would give us some more leeway in the fiction, and I like having that leeway, so I think it's worth exploring that. Not necessary, though.

The problem with this is that Adventures (as defined above) are very limited in possible scope. I can only see this working if adventures are built in dungeon/castle assault style.
When I say that, I accept that the dungeon/castle you may be assaulting could be a social network of the criminal-underworld/upper-class-politics as much as it could literally be a dungeon crawl or castle invasion.
But basically, you are always going around attacking established and/or stationary enemy encampments and societies. Basically, your only choice is to be marauders.

Well, that's just the sort of "default mode" of D&D. You go somewhere where monster/evil is, mess up its day, take its stuff, and come back to town.

It's also the default arc hero's journey (which is why I adopted the term "adventure" to begin with): Go out, do something heroic, come back.

But I think you could invert it pretty successfully in an invasion scenario, too. When orcs come to attack the town, or an earthquake strikes, or whatever, you simply don't have the option to take an extended rest. Which makes sense: you're not going to bed down for the night in the middle of an orc invasion. Or if your party is the target of an assassination plot. Or whatever.

So I am not sure what kinds of games this would not work for, off hand. If you'd like, give me an example of something you think might not work, and I'll see if it does or not, and how it looks if it does! I'd like to try that out!