Quick Note: You’re reading this on Wednesday because this week’s schedule got a little f$&%ed up. If you’re not reading this when it comes out, of course, ignore this note. There will be a feature article this week. But I had to swap the two due to a real life issue and a computer problem. The Feature article will be live on Friday, 9/18/15. I apologize for the inconvenience.

Welcome back to the Megadungeon…

Last week, we carefully analyzed the experience system behind D&D 5E to come up with an overall plan for the structure of the megadungeon. Currently, we see the megadungeon as playing out over the course of a series of 25 to 30 adventuring days. Each day should occur at a certain experience level and should contain just enough encounters to be challenging to a party at that level. By planning the proper number of days, encounters, and challenges, we should be able to carefully control the pace at which the characters gain experience. That ensures that they will reach the climax of the adventure at 11th level. However, we will also include opportunities for additional experience points such that the PCs may very well be 12th level and could even be as high as 13th level. These opportunities will also be planned so as to avoid the characters gaining too many additional levels to early in the adventure.

That sounds like a great plan doesn’t it?

But there are some serious, SERIOUS flaws in that plan. Generally, there will be four to six serious flaws. We call them players.

Let’s talk about resting and the adventuring day.

Healthy Sleep Cycles

Here’s the issue. One of the reasons the adventuring day works as a pacing and threat mechanic is that the PCs have daily resources. That is, they have resources that dwindle throughout each adventuring day and can only be recovered by resting overnight. These include hit points and hit dice in D&D 5E, but they also include a whole gigantic pile of class abilities. Spell slots, of course, spring to mind. But almost every class has some nifty resource that gets drained through the day and needs a good night’s sleep to recover it.

The result of this? PCs are at their most powerful and most versatile immediately after what D&D 5E calls a long rest. That is, an 8-hour period of sleep or minimal activity. And that creates a sort of weird optimal strategy. In the absence of any external reason not to, the PCs are the most efficient when they handle one encounter or obstacle and then go to sleep and recover all of their abilities.

Once again, this is where I point out my caveat on petty GM arguments about how the GM should handle this or that or whatever. Or complaining about the system. We’re talking about what is. And we’re designing an adventure to do what we want. So arguments about what “should be” are a waste of time. What I’m stating above are realities of the system. And if we don’t want to get f$&%ed over by them, we need to plan around them. End of story.

Now, D&D has an in-built limitation on resting. A character can only benefit from a long rest once every 24 in-game hours. By itself, though, that limitation doesn’t mean anything. After all, there’s nothing in the game that prevents a group of heroes from delving into the dungeon, cutting through the orc guard post, retreating, and then sitting around for a full day, playing poker, drinking looted orc beer, and waiting to fall asleep again so they can clear the next room. And the one thing that really would prevent that – boredom – is not a factor because time passes just as quickly or slowly as the GM narrates it.

In addition to that, the game also provides a short rest mechanic. And some resources get recovered after a short rest. Now, it is clear that these resources are only intended to be available every second or third encounter. Why? Because, otherwise, they would just be available once an encounter. And a short rest is no simple thing. It’s sitting around in one spot for an hour doing nothing. In fact, the DMG specifically notes that the GM should anticipate one to two short rests per adventuring day.

So, we want our heroes to push through six to eight encounters at a time. But the game tells them to retreat after every encounter. What can we do? Well, before we go too far into what we can do, let’s look at what GMs usually do?

Run Them Into The Ground

The whole short rest/long rest thing really became a problem in 4th Edition. That’s when every class had numerous resources that recovered after short rests and after long rests. See, before then, the “15-minute adventuring day” was usually just a theoretical problem. Most parties didn’t do that. And most GMs kept it mostly under control. Occasionally, a party would start pulling out too soon for the GM’s taste and then the GM would start making it harder and harder to get a good rest. “You can’t rest in a dungeon,” they’d warn, “or something will just attack you.” The party would get attacked in the middle of the night, spoiling their rest.

What kept it from being a really big problem was mostly the players. Few players really wanted to do that. They had this vague sense that “fight one fight, then run away, wake up at full health to fight another day” was kind of cheating. They had a sort of drive to push through as many encounters as they could and then retreat when they had to. Occasionally, someone would overextend themselves and there’d be a debate about going back to get some sleep, but that was the exception, not the rule, for most tables.

4th Edition really changed that dynamic. Even with the “milestone” system that was meant to provide a drive forward, when you had those nuclear option daily powers getting flipped face down and could see how much better they were than anything else, it was hard not to think “well, what if we do go back now? No harm, right? And I can get my nuke back!”

Trouble is, the GM response to this – really, a knee jerk response – was often to ban resting. “You can’t find a safe place to rest in the dungeon,” they would say. “Okay,” the players respond, “we leave the dungeon.” “Fine,” says the GM, “you get attacked in the middle of the night, spoiling your rest.” “We rest again,” say the players. “You can’t,” says the GM.

It became a sort of arms race that ended with the GM saying when the players could rest. And that actually got codified into 13th Age of all things. The GM decides in 13th Age what constitutes a “full heal.”

But there’s a big problem with taking control of the rests away from the players. You’ll eventually kill the PCs. Unfairly.

See, sometimes, the players will abuse the rest mechanic. But sometimes, through a combination of poor choices and poor die rolls and a few wrong turns or whatever, the PCs might end up a little overextended. It happens. And smart PCs will retreat to recover at that point rather than press their luck. It’s a tactical move. And if you ban it at that point, you’re no better than the jockey who rides his horse until it drops dead underneath him. You killed those PCs.

The players are supposed to think about when to rest. And decide to rest when they think it’s the best tactical option.

Incentives and Disincentives

Here’s a simple fact: when you want people to do something, the best thing you can do is offer them an incentive. And when you want people to not do something, the best thing you can do is offer a disincentive. We can’t ban resting. We can’t take control of the decision to rest away from the players. What we need to do is find ways to make the decision to rest unattractive so players won’t want to rest.

Traditionally, GMs fall back on the “time constraint” as a great way to discourage resting. But those GMs are wrong. They don’t understand disincentives. See, if you have three days to complete the adventure and there’s 48 hours worth of tasks to complete, you’ve basically put a hard limit. You’ve said “three long rests maximum.” That’s not an incentive. That’s just a limit. Now, there are ways to turn that back into a disincentive. If there are 24 hours worth of required tasks and 24 hours worth of optional tasks that make success in the final encounter more likely, then it’s less a hard limit. Each rest after the third rest means you have to give up one of the three things that make success more likely. Of course, that only matters if the GM is willing to let the adventurers fail. But that aside.

Beyond that, though, a time constraint in this dungeon is counterproductive. Remember, we want to drive the heroes to explore and uncover. We want them to feel free to wander. And we want to reward them for doing so. If they are under the threat of a time constraint, they will try to find the most direct route to their goal and follow that. In a weird way, we want the adventure to have a leisurely pace. Yeah, I know. That sounds funny, right?

Wandering Monsters

One of the things that traditionally stopped the heroes from resting inside the dungeon was the threat of wandering monsters. In theory, a dungeon is a living place and creatures are wandering around it constantly. If the PCs blunder around noisily or sit in one spot for too long, eventually, something will stumble on them and try to kill them.

Traditionally, wandering monsters work like this. After the PCs spend a certain increment of time in the dungeon (typically ten minutes or one hour), there is a fixed chance that they will have a random encounter. The GM rolls, the encounter leaps out of the shadows, they fight, encounter gets resolved. Often these chances were modified by what the PCs were doing. If they were being loud or sitting still, they would tend to attract more encounters.

Now that approach keeps the PCs from sitting still for too long in the dungeon, but it also encourages them to retreat from the dungeon if they feel they need to rest. It doesn’t prevent them from taking overnight rests whenever they want. And we can add wilderness encounters if we want, but that only convinces the players to go all the way back to town to rest for the night.

The other problem with that approach is it encourages the PCs not to waste time in the dungeon. That is, not to wander too much. Time becomes precious. Now, we don’t want them sitting idle for long periods of time either, so we have to accept a certain amount of “being in the dungeon is dangerous.” But we need to make NOT being in the dungeon dangerous. We need to make leaving the dungeon unattractive.

What’s Unattractive About Random Encounters Anyway

But hold on. Let’s ask ourselves what’s really unattractive about random encounters anyway? Hear me out. I’m exploring a dungeon, I fight through a room of giant rats and get 100 XP and expend 10% of my resources. Then I fight through a second room and get 100 XP and expend 10% of my resources. So far, so good. But then, the GM rolls for a random encounter. I fight through that, earn 100 XP and expend 10% of my resources. After all, XP gain is commensurate with difficulty and difficulty is commensurate with resource expenditure, right?

Now, assuming I – the player – can freely choose when to retreat and rest, and I choose to retreat and rest when I have spent 80% of my resources, does it matter if those resources were spent on random encounters or on planned encounters? No, no it does not.

And, even worse, if my GM is the clever sort of GM who carefully uses the XP system to plan the pace of the adventuring day, I might have stopped short of a major climax that I was supposed to reach that day. Worse, I might have a gained a level that makes that climax a little bit less climaxy.

From a player standpoint, there’s no difference between random encounters and planned encounters unless the player is actually forced through a certain number of planned encounters. And the whole point of random encounters, or one of them, is to obviate the need for forcing the players through certain numbers of planned encounters.

So, what if random encounters didn’t provide XP at all? What if they only sucked up resources and gave nothing in return? That’d certainly make them painful. The players would quickly learn to avoid random encounters and to conserve as many resources as they could.

Now, that’s just hypothetical. Removing XP from random encounters is pretty heavy handed. And we have built some wiggle room into our leveling scheme, so we can tolerate XP from random encounters. Just not that much, right?

What if Combat is Not the Driver of XP?

Now, I know I said that we’re going to assume we’re writing a game for publication and we don’t f$&% with too many of the established rules and teach too many new systems. That we wanted to choose carefully what things we would play with. But we already decided that we were going to play with the XP system, right? Remember that whole “tiered XP” business? Well, it’s becoming pretty clear that the XP system for D&D is going to be an obstacle to what we want to do. Is it okay for us to rewrite it?

Now, I don’t mean change everything. But, can we change what things earn XP? Hear me out.

We’ve got this concept of an “encounter” as being a thing that presents a certain amount of risk or expends a certain number of resources. When the PCs overcome an encounter, they gain a certain amount of XP. Now, the PHB and DMG tie encounter to combat pretty expressly. But we’ve already noted we want to get beyond that definition a little. We want obstacles like traps and hazards and action scenes. Right?

In addition, there’s other behaviors we want to incentivize, right? Exploration and discovery, primarily. So, what if our megadungeon offers XP primarily for overcoming specific obstacles. We’re just broadening our definition of encounter here. If you fight five goblins, you gain 250 XP, just like you would normally. Or if you cross the pit of razorthorn, you gain 250 XP. Right? But on top of that, we have bonus XP. What if gaining a piece of lore or a specific artifact is worth 50 XP? That puts a tangible reward on top of “learning about the dungeon” or “adventuring archaeology.”

Now, on top of that, if you kill monsters that aren’t already part of another encounter or obstacle, we’ll give you a little bonus XP, but it’s piddling. In fact, it’s so tiny that it drives home the fact wandering encounters are a waste of time. What if killing monsters is only worth one tenth of the XP listed in the monster’s stat block. If you overcome the goblin guard post, it’s worth 250 XP because that was an obstacle. But if you get ambushed by five goblins and fight them off, that’s only 25 XP.

That might even encourage the players to seek ways to avoid combat.

Now, we’re going to come back to this. Because we are going to write (and tweak) our own alternative XP system. But let’s go back to wandering monsters, because that’s where this started.

While the Hero’s Away…

Having digressed to XP systems, we now have a mechanic to prevent the players from stopping and sleeping too frequently IN the dungeon. But what keeps them from leaving? What makes them want to spend as much time as they can in the dungeon?

Imagine this: you’ve got this chamber and it’s a goblin camp. The goblins defend it, they live there, they hunt from there, and so on. And then, one day, the goblins are dead. The heroes came it, murdered the goblins, took their stuff, and left. What happens?

Well, maybe nothing at first. The other creatures in the dungeon have probably learned that that room is goblin territory. But eventually, something is going to notice the goblins have left and there are juicy goblin bodies laying in there just begging to be eaten. Right? Something is going to move in.

And there’s our answer. As the heroes “clear” rooms – that is, as they remove entrenched creatures and obstacles – other wandering beasties gradually backfill those rooms. The longer the heroes are out of the dungeon, the more backfilling happens. So, when they return, rooms they cleared will have other threats in them. Now, these should be lesser threats. Nuisance threats, for the most part, but enough to chew up resources.

That means the heroes won’t want to be away from the dungeon for very long. They don’t want to give it a chance to restock. But they have to rest eventually.

How to Make a Mechanic More Interesting

So, we have a neat mechanic now, right? When the heroes explore the dungeon, their behavior may draw random encounters. And when the heroes leave the dungeon, encounters fill in the cleared areas. These encounters are less dangerous than standard encounters, but they are also worth far less experience points, making them painful and annoying.

That’s neat and all, but we’re asking the GM to do extra work now, right? After every visit, the GM has to “restock” the dungeon. If we’re going to ask the GM to do that work, we want to get as much bang for our buck out of this as possible.

For example, the wandering monster mechanic can also be a penalty for failure in certain encounters. A simple example is the alarm trap or the classic shrieker, a harmless fungus that emits a loud noise when disturbed. When the PCs blunder into one of these things, they bring wandering monsters. Maybe decisions the players make throughout the game will change the chance of wandering monsters.

On the other hand, this makes shortcuts through the dungeon more valuable. If the PCs discover passages that allow them to bypass long stretches of the dungeon they’ve cleared, there’s a real value to them. And rooms that are somehow free of or protected from wandering encounters become a great way to give the PCs a chance to rest. We’ll keep these possibilities on the back burner for now. There’s lots of ways we can interact with this mechanic.

But there’s another interesting interaction that bears mentioning. Suppose, for example, somewhere in the dungeon is a sealed tomb. Until the seal is broken, there are no wandering undead in the dungeon. But after it’s broken, undead start to spread through the dungeon. Suppose there is a tribe of goblins. Once the tribe’s leaders are wiped out, goblins no long appear in the dungeon.

In the past, random monster tables have usually been boring static affairs based on the level of the dungeon and nothing more interesting than that. What if our wandering monster tables are factional? And the actions the PCs take in the dungeon affect what kinds of wandering monsters appear? That’s a pretty cool way to tell the story of the dungeon.

Making the Metagame Work for You

We’re not ready to start banging out some hard and fast numbers here. Before we can figure out this wandering monster system, we’ve discovered we’ve got a whole XP system to figure out. And we’ll look at that in more detail next week. But I want to address an important design concept right here: making the metagame work for you.

Here’s the deal. Like it or not, there is a metagame. That is, there is a structure underlying the game. And there are always going to be players who are good at analyzing it and using it to their advantage. And you can piss and moan all you want that this shouldn’t be. But you don’t get to talk about should when you’re designing something like this. You need to worry about what is, not what should be.

Players should not metagame. Sure, fine. Whatever. I don’t agree, but it also doesn’t matter. Because some players will. And if you don’t want those players breaking your system, you make a system in which metagaming leads to the decisions you want the players to make.

Real human beings don’t do one tiny portion of a task and then sleep for 23 hours. Real human beings have inertia. The longer we work at something, the more driven we are to keep working at it. At least until fatigue and biological needs say otherwise. Real human beings are also driven by a sense of urgency even when no urgency actually exists. We are hardwired to treat lost time like a lost resource.

And that’s how characters in the game should behave. But D&D isn’t built to provide any incentives for that behavior. It relies entirely on the players to bring their own incentives, the GM to provide incentives and disincentives, and adventure designers to build incentives and disincentives.

Which is precisely what we’re doing. We’re creating a system whereby, if you understand the system, it leads you to make the decisions that most human characters would make in the game. That is, push forward, don’t waste 23 hours out of every sleeping, and treat everything as urgent.

But that only works if the players understand it. Which means, ultimately, these rules have to be a part of what the players know about the adventure. They have to understand that every hour spent outside the dungeon increases the chance of wasted resources for minimal rewards. That suggests our final product might actually need a player resource. A few introductory pages the GM can photocopy or download and handout. Which is fine anyway. Most adventures would do well with something like that.

We’re also going to need to spend time telling the GM how to restock the dungeon. Which means we also have to design the dungeon in such a way as to make it quick and easy to restock. And we’re talking about a dungeon that may have hundreds of encounter areas and will take 25 to 30 adventure days. This dungeon could conceivably give a table nine months to a year of fun game play if they stretch it out.

And THAT is why we’re spending so much time deciding how to present things and how to use the systems in the game. And I feel like I’d be cheating you if I didn’t explain the thought processes.

Next week, we’re going to work out solid XP numbers and a basic framework for the wandering monster mechanic, though that will probably change over time.