Incremental Encounters

by Ameron (Derek Myers) on August 17, 2011

A typical encounter has five PCs facing off against five equally powerful monsters and everything happens simultaneously. But what if the encounter was broken down into five incremental steps, each step representing one monster, and the outcome of each step determined the threat level of the step to follow?

Before diving into this scenario let’s not overlook skill challenges. What if, instead of automatically using five monsters, the DM used five complexity 1 skill challenges, or even a combination of five monsters and skill challenges? Assuming that monsters and skill challenges can be used interchangeably to create balanced and satisfying encounters, is this kind of incremental encounter viable? Absolutely.

Think of the encounter as a flowchart. The PCs have to overcome five monsters one by one, or five complexity 1 skill challenges to complete the larger encounter, earn their XP and move closer to their next milestone. They begin by facing a monster of their level or a complexity 1 skill challenge. If the party succeeds then the next monster or skill challenge is the same level or complexity as the one they just completed. If they fail then the next monster or skill challenge increased by one level. This continues until the party faces a total of five individual obstacles.

In the case of skill challenges this rewards diversity, imagination and luck while making things a little more difficult for parties with gaping holes in the skills department.

This kind of encounter works equally well as a series of skill challenges or a series of combat encounters. For example, imagine that the PCs have to move through five consecutive rooms. In each room is one monster (or a complexity 1 skill challenge). If the PCs defeat the monster quickly and quietly then the monster in the next room is caught but surprise and cannot send for reinforcements (success). They go through the door and again only have to deal with one monster. However, if they bungled the first combat (failure), then the alarm is raised and every subsequent room has two monsters in it. This formula continues until they get through the final room.

In an all skill challenge scenario, an exceptionally unfortunate party might end up facing a complexity 5 skill challenge by the end of the encounter, whereas as a party that continues to succeed will face nothing more difficult than five complexity 1 skill challenges. A party of five should have enough variety and versatility that they will eventually succeed on at least a few of the skill challenges along the way. So it’s pretty unlikely that they’ll face a complexity 5 skill challenge at all.

Throwing potentially as many as 15 monsters at the party without allowing them to get a short rest is not a good idea. And neither is making them complete skill challenges ranging from complexity 1-5 before finishing the encounter. So when increasing the complexity of the next skill challenge don’t demand more successes, instead increase the DCs or limit the skills required for success. Looking at the monster example it’s probably better to beef up the monster rather than add more opponents. You could give each monster increased hit points, defenses or attack scores rather than a companion. This way the party is still only facing one monster at a time, but it is harder to defeat if they failed in the preceding room. Perhaps the monster had time to put on his armor, retrieve a magic weapon or drink a potion when he heard the fighting.

Obviously the encounter shouldn’t be anything as boring as just moving through five rooms. Be imaginative, but be sure that there are obstacles worthy of skill challenges or combat along the way. For this kind of encounter to work best, especially if you’re going to use five skill challenges, then you have to ensure that the PCs can use different skills along the way. Five locked doors won’t pose any difficulty to a party with a Rogue or two, but could stop a party of Fighters in their tracks.

This is not to say that you can’t throw a locked door in front of a party of Fighters. In fact forcing the PCs to use their untrained skills or their imagination often makes for more rewarding scenarios. Just don’t go overboard or the players will get frustrated and angry.

When deciding upon each skill challenge ask yourself if a skill challenge is necessary for this situation in much the same way you’d ask yourself would combat be necessary in this situation. If the answer is no then don’t slow things down for what seems like unnecessary checks. If there’s no one in this area of the castle then Stealth checks are not necessary. However, if there are guards patrolling just outside the open window then perhaps a skill challenge makes sense.

It’s important that any skill challenge have real consequences for failure. When using this kind of incremental encounter design, failure means increased difficulty during the next mini skill challenge so be mindful before imposing any additional consequences for failure.

One interesting twist to this kind of encounter set up is that the PCs may decide to intentionally fail one of the incremental steps, be it a skill challenge or a combat encounter. Perhaps they want the monster to flee and alert the guards or maybe they choose to take a more difficult path. Maybe they figure that they’ll have a better chance of getting six successes using their better skills than four using their untrained skills. Since the encounter isn’t talking place in a vacuum the PCs should have a good idea of what skills are going to be required down the road, even if they’re not there yet. It’s up to the DM to decide how much foreknowledge the PCs should actually have, but giving players this option may result in a more interesting and fun encounter overall so be ready to say yes.

Although I haven’t yet had the opportunity to try this kind of encounter in actual play, I’m confident that incremental encounters like the ones I’ve described will work. Players will likely be intrigue the first time the come across this kind of situation because this is not how a typical D&D encounter is usually set up. Play with the design and make sure that each step is challenging, but not deadly. Remember that each step is supposed to represent 1/5 of a normal encounter’s difficulty. The advantage of this kind of encounter design is it really rewards players for playing smart and reminds groups that don’t consider the consequences of their actions that things can always get worse.

Have you ever used this kind of incremental encounter design? How did it work? Is this something that you’re likely to try with your gaming group? How might you tweak an incremental encounter to make it more interesting?

Related reading:

Looking for instant updates? Subscribe to the Dungeon’s Master feed!

1 Sunyaku August 17, 2011 at 7:46 pm

Reading through this made me think of a very cruel maze… something akin to death mountain from Zelda II.

You enter a room and are faced with a challenge of some kind. Failure magically opens one door while success opens a different door. Various successes and failures could cause the party to end up all over the place. Even more fun– you could draw this out LIKE a maze to keep track of where characters are. In this way, you could have special rooms hidden throughout the maze (e.g. containing treasure or special monsters) that players might never discover. And of course, you could limit the numbers of exits or “solutions” to the maze to as many or as few as you want… perhaps after a grueling series of twists and turns the characters find themselves back at the entrance!

You could number the rooms to keep track of the “solutions” players have found for a given puzzle, so that they can choose to fail the second time around if they choose to (although obstacles would reset every time they enter a room). But, as you suggest, I still think it would be good to incorporate a sliding scale of difficult depending on failure or success… but possibly limit how easy or hard the encounters can become (unless you want to make this a meat grinder mod all off on its own).

Perhaps failing the right challenge will get the group back on track to the exit! A dungeon like this has sssoooo many possibilities… and has great potential to be far worse than anything Acerak ever dreamed up…

2 Phantasmavore August 17, 2011 at 11:02 pm

This is a capital idea/proposal! I’d definitely be up to playing through such a dungeon crawl. I wish that D&D Encounters was structured in such a way rather than the somewhat straightforward manner it has been so far, but I’ve only managed to play through the final 2 episodes from last season, and the first 2 of this season, so maybe there’ll be a taste of it soon enough.

I’m going to be joining a Living Forgotten Realms game this weekend, and will have yet another flavor opportunity to try out, so maybe I’ll get a fix like the one you described yet!

3 iserith August 18, 2011 at 11:52 am

I kind of get where you’re going with this, but for whatever reason, I’m not able to see how this could be a universal design concept, per se, outside of the dungeon idea that Sunyaku puts out (which sounds like fun by the way). I read the post yesterday and decided to sleep on it to see if I could come up with other applications before posting. Damn the bad luck, but I can’t.

That may have a lot to do with how I run my games currently. I run dungeons by “Sector.” A sector may be one large room or more likely a series of rooms and hallways. This “Sector” is populated by one encounter’s worth of monsters, usually +2 or more levels xp value than the PCs’ levels. Each sector is organic to the extent that everything makes sense and creates the “old school” feeling of the one trap in the hallway, the one orc in the barracks, the one gelatious cube squishing down the corridor, etc. However, depending on the PCs’ actions and location, they can draw in other monsters to a given fight. I usually do this with a roll of 6 on a d6. It’s my version of “wandering monsters.”

I refined the idea after reading comments by Robert Schwalb and Neogrognard. In order to keep the pressure on the PCs, they can’t take short rests in a sector that is populated. However, they can “rally rest” – which is to say, gain back one encounter or spend one healing surge in one minute outside of combat. Each rally rest comes with a wandering monster check (6 on a d6). This is to simulate that the place is living and not just a bunch of monster closets waiting for the PCs to show up and stomp their guts out. If the sector is clear, they can short rest.

So, essentially, this is the same idea as your article above, only simplified (if you consider house ruling a simplication, that is). It keeps the pressure on the PCs while “breaking up” encounters that can escalate depending on the PCs’ actions. I can still (and often do) put plenty of skill challenges in each dungeon or during combats to vary the action, but rarely does it trigger more combat. I find that particular victory or defeat condition to be not as interesting as just about anything else. “You failed the skill challenge, so here are some more orcs to fight.” No thanks. That’ll just encourage my players not to do the skill challenges because failure is not very interesting – they don’t need a skill challenge to find more orcs to fight. They’ll just keep exploring till they find them.

In short, I like the idea, but can’t find additional uses outside of the puzzle dungeon that Sunyaku mentioned. Perhaps you can enlighten me, Ameron!

4 Ameron August 18, 2011 at 12:40 pm

I never really thought about this approach as being like a video game, but now that you mention it you’re absolutely right. I definitely see this kind of encounter design working in a maze. The idea that one of two doors opens depending on the party’s success or failure is a good one – especially if the party doesn’t know which result opens which door. They’ll constantly be second guessing themselves (Is this going to be a good door or a bad door?).

You’d have to had a very clear map, flow-chart, or diagram in order to keep things straight. Numbering the rooms is the most traditional approach. I envisioned the steps as part of a pyramid diagram. The PCs begin and the top and then each step brings them down a level. (See below.)

             1    2
           1    2    3
         1    2    3    4
       1    2    3    4    5

I like the idea of inserting an exit somewhere in the middle. A party facing a complexity 5 skill challenge may decide to exit the maze and start again, taking what they’ve learned and hopefully passing successfully through all the complexity 1 challenges the next time.

The more I think about it the more possibilities I see.

D&D Encounters: March of the Phantom Brigade had a couple of encounter where the actions of the PCs in one week did determine whether or not they got scenario 1, 2 or 3 the following week. It was likely more work for the designers but the players really liked the idea that the adventure so linear. Unfortunately, there’s nothing like that in this season of D&D Encounters.

5 Sunyaku August 18, 2011 at 6:50 pm

@Ameron Indeed! It becomes an exciting architectural endeavor as well. If you connected the rooms together, it almost becomes a little like the B movie “the Cube” ( I could envision room 3:2 (row:room) has a total of 4 doors… 2:1, 2:2, 4:1, and 4:3… but this wouldn’t necessarily mean the design would need to be linear… you could create a “spherical grid” of interconnecting rooms so that the maze bends back in on itself in all directions! For simplicity’s sake in rooms that have four doors, I think I would require a sort of “check valve” system so the party could only flow in one direction at a time (of course you could make special exceptions to this in certain rooms).

Or, if you don’t want a spherical design, you could do a variety of other things at the “edges”, or even in the middle. With trap doors, secret doors, multiple levels, and teleportation/scrying circles, the possibilities to make the maze even more confounding are endless! >:D

Placing the “real” exit in the middle would be fun, but it almost seems prerequisite that there is a least one “fake exit”… perhaps you fight a “mimic door”, and after defeating it you can reverse the direction you were going through the maze? Lolz.

And I completely agree that players wouldn’t necessarily know what they did to open a given door. I would even go so far as to make “failure” the requirement for “success” in certain rooms.

6 Sunyaku August 18, 2011 at 7:10 pm

The more I think about this puzzle dungeon, the more I think this could occupy (or trap, depending on perspective) players for many, many levels of play. Lufia II was one of my favorite games for the SNES, which contained an escalating puzzle dungeon called “The Ancient Cave” which was roughly in the same vein of the ideas we are discussing. I recommend checking it out:

It might be a good idea to consider adding a couple design mechanics from the “Ancient Cave” into the puzzle dungeon. If you die, you are returned to the start unharmed (we could make the dungeon some sort of temporal anomaly to explain this perhaps…) and if you reach a certain “level” (e.g. room 5:1, 5:2, 5:3, 5:4, or 5:5) you receive an item (it was called “Providence” in Lufia II) that allows you to exit the maze along with any ‘special’ items you have obtained. Or you could always be harsher and say that if you die, your soul is forever lost in the weave of time, “providence” is the only means of escape other than the true exit (and finding the true exit is the only way you get to keep the epic loot).

What better place is there to hide some sort of horrible, universe destroying artifact? The key to the chained god perhaps? Or something else equally dangerous? >:D

Comments on this entry are closed.

Previous post:

Next post: