Megadungeon Monday: How to Award XP!

Print Friendly, PDF & Email

September 25, 2015

Welcome back to the Megadungeon!

Last week, we did a lot of high concept dicking around, talking about the structure of the adventuring day. Ultimately, we decided that A) we’re going to need to have random encounters and B) we’re going to have to have a way to restock the dungeon between forays. Those things will keep the party focussed on not sitting still for too long in the dungeon and not spending too much time outside of the dungeon resting. Moreover, we can make our dungeon seem more dynamic and alive. As the PCs clear areas, vermin and nuisance monsters wander around. Moreover, the nuisance monsters can change as the party takes actions in the dungeon. Moreover, we can also have traps and other dungeon elements that interact with the random monster system. We can have alarm traps or alarm monsters that bring random encounters down. And we can reward the party with shortcuts through the dungeon that allow them to bypass restocked areas.

That’s all really cool, that also brought up a tricky problem. We sat down and examined the XP system very carefully to map out how many encounters and days of adventure we want the dungeon to have. But, the primary driver of XP in D&D is monster encounters. So, if we’ve got random encounters running around, that’s going to f$&% up our careful balance, right? Okay, hold on. Let’s take a detour here. I need to because of some of the feedback I’ve gotten.

Imperfect Balance

Let’s talk about this whole game balance deal. Right? The idea behind game balance is to provide a nice, level play experience. The players always experience just the right amount of challenge and spend just the right amount of resources to earn just the right amount of rewards. And that’s the promise that the D&D encounter balancing system TRIES to make. And, if you read the first part of this series, you’d be forgiven for thinking that I’m going to make that same promise.

But, look, perfect balance is bulls$&%. First of all, it’s impossible. The game relies on random elements, AKA, dice, that can really f$&% up the balance system. A series of crappy rolls, especially at the lower levels of the game, can cost the party dearly. A series of great rolls can turn combats into a cakewalk. On top of that, there’s a great deal of player and GM skill involved. Players who build very effective characters and put together combinations of races and classes that work well together and perfect their teamwork will spend far fewer resources than players who are less effective or tactically inclined or just not making the effort. GMs who are tactically weak will present less of a challenge to their players than those who are really effective. Skill and chance together f$&% up that perfect balance.

Which is precisely what we want. Random chance is fun precisely because it’s unpredictable. We could take the dice out of the game and come up with systems of numerical comparisons based solely on the choices the players and the GM make to determine precise mathematical outcomes for every outcome. But we don’t. Because that makes less exciting. Likewise, winning because you came up with just right the plan or high fiving your teammate because the two of you executed a perfect combination of spells and special abilities is also fun. Players LIKE to feel good at the game. We want the skill and chance elements to have a say in the outcome.

Game balance, though, is about precisely the opposite. It’s about minimizing the impact of chance and skill to make the outcome as much a foregone conclusion as possible. At least in D&D. Think about it. The DMG basically swears if you build an adventure that includes precisely six 1st-level encounters, one of which is easy, one of which is hard, and the rest of which are moderate, the party will survive all six in one day and gain enough XP to attain second level. No matter what. Every time. Or at least that’s what it seems. And the reason it seems like that is because the DMG is a book of lies. Well, not really. It’s actually a book of poorly explained ideas.

The real idea behind the DMG is that, if you build that adventuring day, the average GM and the average group of players will, most of the time, experience just enough of a sense of danger and use up just enough resources to have a satisfying play experience. One that is neither frustrating nor a cake-walk. On average. Over the course of many, many play sessions.

And THAT is what a lot of GMs forget. Some GMs whine that their players are having a hard time of things. Others complain things seem too easy. And they complain that the numbers are broken. When, in truth, there’s a lot of wobble. And part of the GMing gig is adjusting for the table you have. If you build encounters as written and they are too easy or too hard, you are EXPECTED to adjust the encounters. You are expected to lean toward hard encounters. Or easy encounters. Imagine if everyone in the world set the volumes on their stereos to five and then half the people complained the stereo was too quiet while the other half complained it was too loud. You’d tell the f$&%wits to twist the dial. Five is just the middle-of-the-road, the average volume. It isn’t the volume for everyone.

Now, apart from trying to provide the average group with a good play experience and empowering GMs to adjust the difficulty knob, the other point of game balance is to prevent extreme experiences. Even if some players find the game easier than average and others find it harder, proper balance should keep the game from going into the so easy it’s boring or so hard it’s frustrating territories. That is, game balance is about trying to keep MOST PEOPLE from blundering into an extremely unsatisfying experience.

So, in short, you figure out the baseline experience for the average group and you make that your target. And then, you figure out the extremes and you do your best to curb those possibilities. But, you also remember unpredictability is fun. So, once you have that baseline experience, if you really want to do it right, you allow a little wobble.

For example, both World of Warcraft and Magic: the Gathering are pretty much acknowledged as paragons of game balance. They are constantly tweaking and fiddling to keep the game at their desired level of balance. But notice that both WoW and MtG have some things in them that are a little off. MtG releases a certain number of cards that are a little below the baseline and some that are a little above the baseline. Purely numerically, WoW allows some deviation in different abilities, so that if you compare two different ostensibly equal attack powers, you might find one is slightly low in output compared to another. The keyword is slight. Why? Well, part of the fun of the experience is discovering things that feel a little bit better than everything else. Or figuring out things that are a little worse than everything else and avoiding them. And some people thrive on challenging themselves by using subpar abilities that are thematically cool. It lets people set their own experiences and pick handicaps and creates a better play experience.

In simple terms, the best games aren’t the ones that are perfectly balanced, but the ones that have a bit of wobble to the balance. You know, the tires are just a little out of alignment. The car goes mostly straight, but you have to keep your hands on the wheel most of the time.

That wobble doesn’t happen by accident. You can’t just build an unbalanced game and hope.

And THAT is ultimately part of my plan. I’m figuring out the baseline for “balance” in my megadungeon. I’m trying to build a solid experience for the middle of the road group and keep the extremes out of the experience. But that doesn’t mean I’m going to religiously follow the schedules I’ve designed. There may be days with an extra encounter or an extra hard encounter. There may be days where the party retreats early and puts off the climax. Or days where they have terrible rolls for random encounters and get nowhere because fire beetles keep eating their faces every time they turn a corner.

In short, I’m designing the experience that probably no group will have exactly. But most groups, if I do my job right, should get pretty close to the right experience most of the time.

Doing that means that, before I can allow for some wobble, I have to figure out how to make the car go straight. Okay? So calm the f$&% down with your panicked fears that every day will have exactly X encounters and I won’t let the party ever deviate from the path. Relax. Trust me.

How to Award XP

All right, let’s get back to the problem on our hands: the XP system. Currently, the D&D XP system is based purely around killing monsters. You kill a monster, you get the XP. End of story. And honestly, that SHOULD make building a megadungeon easy, right? At least in terms of XP. We know we need 85,000 XP worth of monsters. We can just scatter that many XP worth of monsters around a big ole map and be done with it, right?

Well, honestly, that’s how a lot of dungeons were built in the past. Sort of. More or less.

But, ultimately, remember, our goal is to end the campaign with a cool boss fight at 11th level. And we want the party to definitely be 11th level for that fight. But we also want them to have the option to be 12th level or even 13th level for that fight. At 11th level, it’ll probably kill some of the PCs. Seriously. Ultimately, it should cost them everything they have to win, but they should win. At 13th level, they can probably all walk out alive.

So, we have a narrow mark to hit. We need to make sure the party has earned 85,000 XP (per PC) before they can get into Shub Niggurath’s boss room (or whatever). But we also need to make sure they COULD HAVE earned 120,000 XP (per PC) before Sub Niggurath. And we also need to make sure they COULDN’T HAVE earned much more than that.

Moreover, we need to make sure that the party really can’t get too far beyond the levels we expect at various points during the adventure. We’ve already mapped all of that out, remember? Let’s review. I made some very slight tweaks to the Optional XP thing, but don’t sweat them because they will turn out to be wrong (spoiler alert). This is still pretty much the same final roadmap we settled on.

Experience Progression

Let’s talk about that Optional XP thing. Let’s make a slight tweak for a moment, shall we? To illustrate a point.

Optional Tweak

Notice that if we assume one Optional Encounter per day, every day, we’re pretty nearly on target to where we want to be, right? Isn’t that interesting. That tells us if we assume that, on average, a party will have on random or restocked encounter every day and we only give XP out for combat encounters, our system will probably work. If there’s some wobble in the random encounter system, it’ll allow for an occasional day when there are no random encounters and an occasional one where there are two. Right? Perfect.

We could be done here.

Except this system utterly sucks.

Incentives and Advancement

Okay, let’s say you’re a reasonably smart player. You figure out that this whole damned megadungeon is building toward a final boss fight. That’s your goal. There’s two good ways to handle it depending on the type of player you are.

First of all, you might be the sort that realizes that every random encounter chews up resources and drags out the game. And most random encounters are less satisfying than normal encounters. If that’s your basic viewpoint, the best way to play the game is to be as efficient as possible. Stick to the critical path, waste no time resting that you don’t have to, find the final boss, and kill it. Right? Efficient.

Second of all, you might be the sort that recognizes that random encounters give you as much XP as normal encounters. And you can grind those out. Eventually, you can be powerful enough that the random encounters are trivial and the critical path encounters are easy. So, your tactic is to sit in one spot and attract random encounters until you run out of resources. Then you retreat, let the dungeon restock, and do it all again. Keep doing that to get a couple of levels ahead of the curve. And then plow through the adventure looking for the final boss.

Now, I should emphasize that I’m not suggesting all players think like that. But some players do. How many? Who knows. It’s impossible to say. But it is important to note that the XP system and the design of the game creates one of two optimal strategies and neither of those strategies provide the play experience we want. Right?

The thing is, even if most people WON’T exploit a design, it’s still s$&%y design to build a game that encourages a playstyle other than the one you’re trying to create. We want people wandering and exploring. We want them making discovered. Remember? Action archaeology. And this XP system doesn’t support that. And, since we can design any system we want, we might as well design a system that does support the playstyle we want.

That way, if most players just naturally engage with the adventure the way we want them to regardless of optimal strategies, they will get the most rewards possible. And if there ARE players out there who want to break down the system and optimize their game experience, well, we’re encouraging THEM to play along with everyone else.

The problem with D&D is that it thinks about XP only in terms of advancement. Face this many challenges, gain this much power so you can be powerful enough to face more difficult challenges. Then face those, gain this much more power, and so on. But XP doesn’t have to be JUST about advancement. After all, we’ve already decided to build in a ratcheting difficulty system by breaking challenge down by tier instead of by level. That way, we can give players variable challenge experiences. Can we do more with the XP system?

The answer is yes. We just have to decide what we want to emphasize. And what hard decisions we want to drive.

The Play Experience

First and foremost, this is still a D&D game. And we’re still going to be spending most of our time going through encounters. The part finds an obstacle, decides how to overcome it, risks loss or spends resources, ultimately overcomes it, and moves on. We don’t want to change that experience, right? That’s D&D’s bread and butter.

But, does it always have to be combat? And does it always have to be killing? If the party drives a group of kobolds out of the dungeon completely, does it matter that they didn’t kill them? No. The encounter is still defeated. And what about things that aren’t combats? What if the party crosses a ravine? Or disables a trap? Of course those count as encounters. There was either a risk that was overcome or resources were spent. Or both.

Not only is that worth XP, that is the primary driver of XP. I don’t want to change encounters as the driver of XP, just to broaded the definition.

But here’s where things get tricky. Let’s say there’s a room with a group of goblins. And the party sneaks past the goblins. Do they get the XP for that? Maybe you think so. Seems like they should. They took a risk, overcame it, they deserve XP. But the goblins are now (in theory) still there. Does the party have to sneak past them again? Do they get XP the second time they sneak past them? The risk is the same. I’m sure you can see where this would lead to a problem.

So, we differentiate between circumventing an encounter – temporarily bypassing it – and permanently defeating it. That’s something we have to think about.

But let’s talk about random encounters. Are random encounters encounters. Ostensibly, yes. But last week we talked about several reasons why we don’t want them to be as rewarding as non-random encounters. But we still want them to be worth something. Mainly for fairness and because we’re going to build in a certain amount of optional XP, right?

But what else? After “overcoming encounters,” what do we want the PCs spending most of their time doing. Well, we want them to wander. To explore. And, more importantly, to interact with the environment. We want them to discover the story of the dungeon. Action archaeology. How do we do that?

Well, imagine if the PCs find a bas relief that depicts an event from the dungeon’s history. Or they find a scroll they have to translate? Or an amulet they identify as a religious icon devoted to a specific deity?

Looking back at video games once again, lots of games scatter this kind of crap around. Audio logs in Bioshock. The patient recordings in Batman: Arkham Asylum. The scans in Metroid Prime. All of the item descriptions and architecture and level design details in Dark Souls. The memory objects in Shadow of Mordor. All of those games were exploration based. And so, they all reward exploration with tidbits of lore. They let you figure out the world. And we can do the same thing. In fact, we can do it better in an RPG. Because we don’t need one specific mechanic. Scan visors, audio logs, and all that crap? Those specific mechanics serve the general purpose of sharing information, but they are limited because you don’t want to design a thousand unique ways to interact with the environment. You build one system and then cram all of your lore into it.

But we don’t need to be that limited. We can just have a list of “discoveries” that are all the clues – artifacts, scripts, environmental details, and so on – that add up to the story we want to reveal. Some of them can be a simple matter of finding the thing. Others might require skill checks, examination, research, whatever. Still others might be hidden behind encounters or obstacles or puzzles. In the abstract, though, they are all the same thing: lore. Or discoveries.

Now, the thing is, the random encounter system dissuades players from exploring every nook and cranny, from going back to previously explored areas once they’ve found the way forward, and sitting for a half hour making a rubbing of a bit of ancient text so they can bring it to the library to research it. So, in return, we offer XP for those discoveries. It’s a nice tradeoff. Especially when some of the discoveries can also hold useful clues. For example, a discovery could warn about a particular type of monster. Or reveal directions to a secret room with treasure in it. Or a weakness of a particular monster. Some discoveries might also be valuable treasures. A longsword inscribed with details about a hero of the dungeon is also a magic longsword. Neat, eh?

What about treasure? Is treasure worth XP? In some games, that’s a useful idea. But in this one, I’m not sure it fits. After all, wealth allows the PCs to buy supplies. And magic items are their own reward. In truth, magic items and equipment are their own advancement path. And most players will seek out treasure whether we encourage it or not. And those that don’t, we really don’t need to encourage them to go money grubbing. This adventure isn’t about building wealth. But individual players can make it about building wealth if they want.

So, we have fixed encounters, random encounters, and discoveries. Those are worth XP. Is there anything we’re missing?

The answer is yes. Definitely. Remember one of the other things we talked about as a design goal. As much as we want the game to be about exploration, exploration without a sense of progress and accomplishment doesn’t satisfy everyone. We need to build in a sense of progress. And we talked about pacing and building toward climaxes.

When we look again at our list of video games that we want to emulate because we see them building the play experience we want, we notice that most of them have definite moments of victory, of accomplishment. Often boss fights. Sometimes opening up a new area. Sometimes unlocking a new power. In fact, you can sort of map out your progress through most of those games by the milestones, by the high points. Super Metroid for example. Without going back and playing it, you can probably remember key moments in the game. Encountering Ridley on the Ceres Space Station. Finding the Morph Ball and bringing the ruins of Zebes to life. Spore Spawn. Kraid. Varia Suit. Speed Booster. Return to the Surface. Crocomire. Wrecked ship. Phantoon. Gravity Suit. Maridia. Draygon. Space Jump. Lower Norfair. Screw Attack. Ridley. Broken Metroid Cannister. Baby Metroid Attack. Mother Brain. Escape. That’s not an exhaustive list. And it might not even be the list you think of. But those are the high points for me. The milestones. Some were treasures. Some were story events. Some were boss fights.

Now, this will become doubly important when we talk about gating and managing traffic flow in our dungeon, but for now, let’s just acknowledge that we want to build some milestones into the adventure. Big events that change the game. The high points. They won’t include every discovery and every climax and every treasure. But we’re going to build a story progression into our dungeon. We know that because we have a final boss. And that demands a whole bunch of other stuff to come before it. So, we’ll have milestones.

Fixed encounters, random encounters, discoveries, and milestones. Those are the four ways to gain experience in our dungeon.

Invisible Systems

Now, let me remind you once again that I was strongly against changing too many of the rules of D&D because we want this module to be approachable to all D&D fans. And yet, we’ve spent thousands of words rethinking everything about encounter building and experience rewards. What gives?

Experience and encounter building in modules are a good example of an invisible system. If you look at WotC’s modules, they dole out XP awards at set intervals. Apart from combat experience, that is. They tell them GM “if the PCs finish this chapter, give them 500 XP. Huzzah!”

When and how to dole out experience is something we don’t need to leave up to the GM. We write it into the module and give the GM very specific instructions. The only thing we really need to tell the GM is “hey, don’t give out experience for killing monsters, we’ve already factored that in.” We should also tell the GM to tell the players “oh, hey, here’s the things you get XP for.”

Some GMs do like the information though. So, it’s the sort of thing we could cram into a sidebar. Otherwise, the entire system for designing and pacing the dungeon and how and why we’re giving out XP here for this and not that? We don’t need to explain it. This is a background. It’s a design system, not one we use for table play.

In point of fact, we’re actually doing the GM a favor. Because we won’t tell the GM “hey, figure out how many monsters the PCs beat and add up the XP and divide it by PC.” We can just say “each PC in this encounter gets 100 XP if they beat the monsters.” Done and done.

So, we’re fine on that. Just wanted to throw that out there.

How Much XP to Give for Encounters

Alright, kids, time for some more spreadsheets. But these ones are actually kind of simple. We want to figure out how much XP to make everything worth, so that we can plan out progression. So, again, we’re going to start with the big spreadsheety map thing we came up with in the first article. Because that represents our overall level progression. For the most part.

Experience Progression

Now, let’s start looking seriously at it.

First of all, we’ve got a mistake. Look at the line for level 11? Notice that level 11 has two days of adventure, right? Why? Because we built that chart trying to map out the XP progression. So, we have that line that assumes the heroes start at level 11, have two days of adventure, face 7 encounters in each of those days, and then they hit level 12.

But that’s not what’s going to happen. That LAST day of adventure? Level 11, Day 2? That should end with the boss fight? The last encounter. The PCs aren’t going to see the XP for that day. I mean, they will, but it won’t matter for this adventure. So when we do our math, we can’t factor that day in. I mean, there’s a chance the PCs will go partway through that day and then retreat before they face the last boss. If we let them. But that we can figure out when we actually figure out how to handle that on the last day. The last day doesn’t matter.

The other thing is that that entire part about the optional XP? That’s only sort of a guideline now. We based it on the assumption that there would be a number of optional encounters and that would add up to optional XP. Right? And we wanted to make sure there was just enough to give the PCs a chance to overlevel.

So, the Optional XP Column tells us how much optional XP we want to make available at each level. But we no longer care about the number of encounters. We just care about ballparking the XP gain.

Okay, so there’s two types of XP in our game. Fixed XP and Optional XP. Fixed XP is the XP we’re going to do our best to make sure every player gains. And it’s going to be made up of two facets. Encounter XP that comes from specifically placed encounters along the critical path from the beginning to the end of the dungeon and Milestone XP that comes from specific story accomplishments.

Now, how much XP is a Fixed Encounter worth? Well, we don’t want to f$&% with the system because we’ve already assumed how much XP a Fixed Encounter is worth. Except that we already HAVE f$&%ed with the system. We’ve decided challenge is tiered, not linear. And therefore, to some extent, XP is tiered, not linear. Beyond that, though, we’ve also assumed that every encounter is worth the same XP. We haven’t accounted for difficult or easy encounters. In theory, about a quarter of the encounters will be easier than normal and give less XP. In theory, about a quarter of the encounters will be harder than normal and give more XP. However, in addition to that, every time the PCs fight multiple monsters, because of the way encounters are built, they get less XP than if they fought a single of an appropriate challenge. Why? Because when there are multiple monsters, you increase the XP for calculating the challenge of the encounter, but you give the PCs only the base XP for each monster.

How much does that s$&% matter? Well, that “multiple monster” business, which I call Phantom XP, is bulls$&%. We can ignore that rule altogether. GMs will give as much XP as we tell them to. So, in our world, as long as the encounter is designed to the right level of difficulty, the XP value should be the same whether it’s multiple monsters or one monster. That is, a moderate level 5 encounter should be worth 500 XP per PC regardless of whether it’s one umber hulk or two wights or four ghouls. Those are all, in the eyes of the DMG, roughly equal challenges. Therefore, the PCs should gain 2,000 XP. Done and done.

What about the issue of Easy and Hard encounters. Well, allow me to show you a little something I whipped up, just to test a theory.

Encounter Breakdown Test

What is it? Well, you probably recognize most of it as the XP Thresholds by Character Level table with the “deadly” column removed because that column has no use. Next is the XP needed to gain a level. You know what that is too. At 2nd level, you need to gain 600 XP to get you to 3rd level (for a total of 900 XP). Divide that by the XP for medium encounters and you get the number of medium encounters you need in a day to get you up to the next level. Follow?

Now, if you take that number of encounters and assume one quarter of all encounters are easy and one quarter of all encounters are hard (the rest being medium) and reward XP accordingly, the XP Gained column shows you how much XP you’d actually gain.

For example, at 3rd level, the assumption is you will face 12 encounters. If three are easy, you get 225 XP. If three are hard, you get 675 XP. The remaining six are medium and you get 900 XP for those. All told, that adds up to 1,800 XP. That’s the same XP you would have gained for 12 medium encounters instead.

What’s my point? My point is that actually changing the XP based on the difficulty of the encounter doesn’t change anything. Notice all down the line that the XP Gained for quarter easy, quarter hard, half medium is exactly the same as if you just had all medium encounters.

In other words, as long as you build your encounters such that half are medium, a quarter are hard, and a quarter are easy, you can just give out the same XP for every f$&%ing encounter and still get the same result.

And since we’re already breaking down things by tier, rather than by level, we don’t need that much granularity in the XP system. All 1st level encounters are worth 50 XP for each PC that participated. All 2nd level encounters are worth 100 XP for each PC. All 3rd, 4th, and 5th level encounters are worth 250 XP for each PC.

Now, what about adjusting for large or small parties. Do we really have to bother? Well, obviously, we could, but how far do we really need to worry about it? Here’s the deal. The XP system is kind of inexact anyway. Let’s look at 4 5th level PCs and assume I want to build for them a 5th level fight. They need 2,000 XP worth of difficulty for a medium fight, right? How can I do that?

Well, the easiest way is with a Challenge 5 monster, right? Except a Challenge 5 monster is actually only worth 1,800 XP. I can’t quite hit what I need. What if do a pair of monsters? Well, remember that when I have a pair, I multiple the XP x1.5 to account for the extra difficulty. 2,000 divided by 1.5 is 1,333.33. And I need two monsters, so 1,333.33 divided by 2 is 666.67. So I need two monsters worth 666.67 XP. But I can’t do that. A Challenge 2 monster is only 450 XP. A Challenge 3 monster is 700 XP. I can get close. But not exactly there.

And the thing is, this happens throughout the system. It is impossible to build exact challenges for the right number of PCs. You’re going to be somewhere on the spectrum, but you’re always going a little over or a little under.

Now, fortunately, the DMG kind of lets us know where the cutoff for large and small parties are. A party of 4 to 6 players is a normal group. We only have to start really adjusting the difficulty drastically when we’re outside of that range.

What can do with all of this information? We can make our lives a lot easier. More importantly, the lives of the GM. Now, pay careful attention because there’s some mental gymnastics here.

First, we know that it doesn’t matter if we adjust XP awards based on the difficulty of encounters. If every XP grants a medium XP award, that’s the same progression as granting one quarter easy, one half medium, and one quarter hard. So we can give out the same XP for every encounter, regardless of difficulty.

Second, we know that as long as the party is between 4 and 6 players, we don’t have to do any drastic math to adjust the difficulty. We just multiply the threshold by the number of players. If I have five 5th level PCs, they consider a medium encounter to be 2,500 XP. If there’s four, they want 2,000 XP. If there’s six, they want 3,000 XP.

Third, we know that there’s no combination of monsters I can legally create to give me those XP totals. If I use single monsters, I can do either 1,800 XP, 2,300 XP, or 2,900 XP, meaning I will always be undershooting the average. If I use multiple monsters, I have to deviate even more, I’ll still be unable to hit the right number, and the players will be losing XP because of the difficulty multiplier. Therefore, the best I can ever do is put an encounter on a spectrum between Easy and Hard. Close to either end makes it Easy or Hard. Close to the middle makes it Medium.

Fourth, assuming we have between 4 and 6 players, the easiest Easy encounter is an Easy encounter for four players and the hardest Hard encounter is a Hard encounter for 6 players. That’s the broadest range I can build my encounters into. But, if I put that hardest Hard encounter in the game and have only four players, that encounter will kill them. And if put that easiest Easy encounter in the game and have six players, it’ll be trivial. I can’t do that.

However, and this is where the mental gymnastics happen, the hardest Easy encounter is an Easy encounter for six PCs. And the easiest Hard encounter is a Hard encounter for 4 PCs. If I keep myself in that range, I’m pretty much guaranteed never to overwhelm or underwhelm any party between four and six players.

Fifth, half the encounters should be medium, one quarter should be easy, and one quarter should be hard.

Therefore, I can take the XP Thresholds by Level table, compute the hardest Easy, the easiest Hard, and then take the average of the two, and build EVERY ENCOUNTER to that range. In theory, I won’t have to adjust for party size as long as it’s between four and six. And I can dole out the same XP for every encounter at that level regardless of difficulty or party size. In other words, I can build this table.

Thresholds 1

As long as I build my encounters using colored ranges, where Low is an Easy encounter for 6 PCs and High is a Hard Encounter for 4 PCs, my adventure should work okay. 4 PCs will find it more challenging. 6 PCs will find it easier. But neither group should find it broken. And if I further make sure about one quarter of the encounters are Low and one quarter are High, with the rest falling close to the middle, I can award each PC exactly the same XP regardless of the difficulty of the encounter or the number of PCs.

In other words, if I have a group of six 2nd level PCs and they have a 2nd level fight, each PC gets 100 XP. If there are only four 2nd level PCs and they have the same exact fight, each PC gets 100 XP.

Of course, in my adventure, it’s even easier. Because I’m breaking things down by tier. So, here’s where I build my encounters to.

Thresholds 2

So, every Fixed Encounter gives a set amount of XP to each PC. No matter what. All of the encounters are built to the ranges shown in the tables. Done and done.

XP for Milestones

Now, let’s talk about milestones. Milestones are the other component of the Fixed XP. A certain number of story events are significant enough to be worth an XP bonus. How much of a bonus?

This COULD get really tricky. After all, we budgeted all of the Fixed XP in our adventure based on a certain number of encounters. Right? And we figured out exactly how many encounters we need to get to exactly the right level. And now we want to throw in this other XP that ISN’T optional and it’s going to f$&% everything up, right?

Except it doesn’t have to f$&% anything up. What if we made Milestone XP equal to one encounter. A 1st level milestone is worth 50 XP per PC. Why? Well, I’m just spitballing here. An encounter is the most significant thing the PCs can do, right? But we want the milestones to feel big. If a milestone is worth as much XP as an encounter, those feel like big victories, right?

But that means we’re cramming a whole bunch of extra encounters worth of XP into our fixed budget. Now we have to recompute everything! Time for more spreadsheets!

Or not.

See, remember that we computed the number of days and encounters based on the XP needed and on the safe number of encounters in an adventuring day. Right? But when we figured out the safe number of encounters in a day, we were pretty aggressive. Remember how I said I was going to round up? Yeah. That means every one of my days is potentially pushing the players a little harder than they might push themselves. That means I can remove encounters from a day and the day will probably still be pretty challenging.

On top of that, we didn’t take into account random encounters and optional encounters. Those things chew up resources. And that means they slow the party’s progress. So, again, the number of encounters in a day might be slightly aggressive.

What that means is we have some leeway in every adventuring day to remove encounters. Except, we don’t want to remove the XP. Do you see where I’m going? If a Milestone is worth the same XP as a combat encounter, it’s effectively a free encounter. It’s bonus XP. It’s story XP. It’s quest XP. Because remember Milestone XP is a separate bonus. If we have a Milestone called “drive the orcs from the dungeon,” and to do that, the party has to beat “Kraid the Orc Boss,” well, they get XP for that fight with Kraid. But then they get bonus XP for the orcs fleeing once Kraid is killed.

Milestone XP is just Encounter XP without an encounter. Or added onto an encounter as a bonus because that encounter is really super special.

Now, right now, I have arbitrarily decided there are 10 milestones in my dungeon. That doesn’t include the boss. Why 10? It’s about one per level. And since the party will spend two to three sessions at each level, on average, that means every second or third session, they’ll get Milestone XP. That’s a nice rate of story progression. Especially if the milestones don’t always align with gaining new levels.

Fixed XP Progression

Ultimately, here’s the Fixed XP Progression for my dungeon by tier.

Fixed XP

Basically, it just takes the information from the Master Progression Spreadsheet we’d already made, swaps out some encounters for milestones, multiplies by the XP per PC per Encounter or Milestone, and makes sure we get the PCs to the level they need to be. It doesn’t include the Final Day of adventure. And it actually gets the PCs kind of close to 12th level. Encouraging them to maybe explore a little more.

Optional XP Progression

Now, let’s talk about Optional XP.

What are our sources of Optional XP? Well, we have Random Encounters. And we have Discoveries. But that’s not all. We also have this big, open dungeon, right? And the PCs can wander around and find stuff that isn’t on the critical path. Just because it isn’t on the critical path, though, doesn’t mean we should include some Fixed Encounters, right? If the party decides to go left instead of right and wander off the critical path (on purpose or by accident), does that mean they shouldn’t be able to find a cool, carefully designed trap or obstacle or monster lair? Of course not.

So, we want to include the possibility for Optional Encounters. They are exactly the same as Fixed Encounters, but they aren’t on the path from start to finish. They are off to the side. And, thus, they should give the same XP as any other encounter and follow the same rules.

What about Random Encounters? These include wandering monsters and monsters moving into cleared areas while the party is away from the dungeon. The question is, what are they worth AND how often do they come up. And I’m going to admit I’m being sort of arbitrary here. Because, again, we can make them worth whatever we want and decide whatever frequency we want.

I wanted Random Encounters to be a major pain in the a$&. I want them to be worth a pittance. In fact, the only reason they are worth XP at all is because, if the XP value is low enough, that XP award is a slap in the face.

Imagine you’re an adventurer. And you just finished two big fights. And you earned 200 XP for those big fights. And then, on your way out, you got jumped by some wandering death spiders. You were just trying to get to the exit, but bam, death spiders. And they almost killed you. But you won. And the GM gives you 10 XP. That’s actually kind of insulting, isn’t it?

That’s exactly the point. Remember, Random Encounters are supposed to suck. We are using them to keep the PCs from leaving the dungeon for extended periods of time or trying to sleep in the dungeon or dick around doing nothing or setting off alarms. Basically, Random Encounters mean the PCs f$&%ed up. To some extent. I mean, they are going to happen. But they suck. They should always suck. And the party should never get it into their head that attracting Random Encounters is a good idea.

So, I want a Random Encounter to be worth one tenth of what a normal encounter is worth. That’s a sucky number. As for how often they should come up? Well, at one tenth, it hardly matters. They have so little impact on the XP Progression. To be on the safe side, I’m going to assume a party that does everything right will have ZERO or ONE random encounter per day. Some parties might be unlucky or stupid and have more. But ZERO or ONE per day is my target.

Finally, Discoveries. Discoveries are neat. And I want them to be valuable, but not as valuable as Fixed Encounters and Milestones. But I also don’t want them to be cheap. Half an encounter seems like a noticeable amount, right? A 1st level discover is 25 XP. That’s half what a PC gets for an encounter or milestone. That’s not insignificant.

Now, admittedly, I started off with those assumptions as arbitrary numbers. And then I decided to shove them in a spreadsheet and start fiddling to see how many of each thing I could tolerate. I assumed one Random Encounter per Day and then tweaked the other numbers. I knew I needed at least 26,000 XP worth of optional XP and I knew most of it should come toward the end. Based on that, ultimately, this is the table I came up with.

Optional Progression

Notice two things. First of all, Optional Encounters. There’s 30 Optional Encounters in the adventure placed on the map. And most of them become available at high levels. That’s because those are the highest XP sources. And I needed to make sure most of that comes toward the end so the party doesn’t start overleveling too early.

But that also makes sense based on the progression through the game. Remember, more of the dungeon opens up the more you explore it. So, as the party finds new ways through the dungeon and opens up new passages, all sorts of new possibilities open up. There might be a 10th level hallway right off the entrance to the dungeon that has some obstacle the party can’t overcome until they hit 10th level. And behind there might be some secret treasure guarded by a stone golem, a 10th level monster. So it stands to reason that the number of optional encounters ratchet up toward to the end of the adventure.

We can also use those to build optional objectives. Yes, you could go confront Shub Niggurath right now. Or, because you found the clue to the Shub Niggurath slaying sword, you can go hunt that down to make the encounter easier. So some of that stuff can be payoffs for clues and earlier exploration.

As for Discoveries, having a specific number of Discoveries is really useful. It requires me to break down the secret knowledge of my dungeon into 21 chunks and hide those chunks in the dungeon. It tells me exactly how I have to break down my secret details. But otherwise, it’s a completely arbitrary number. I just filled out numbers that would get the Optional XP total where I wanted it.

The Final Table

We’ve covered a s$&%load of ground, but we’ve nailed down our XP system. It’ll encourage the right behaviors and it’ll also guide our construction of absolutely every encounter. It tells us precisely how much stuff our dungeon needs to contain, including story milestones and significant discoveries.

But by way of summary, I just want to throw this last little table out there. This is how much XP stuff is worth in our dungeon. The GM might never see this, because we’ll tell him when to award what XP, but this is how we’re going to figure it out.

XP Awards

Now, one little side note. Notice that I broke up Defeating and Circumventing an encounter. I kind of hinted at that in the text and it’s something we’re going to revisit later. For now, I just want to tell you the basic premise.

If you manage to circumvent an encounter without actually defeating or removing it, you get half the XP. But you only get that the first time you circumvent the encounter. So, if you sneak past a goblin guard post, you get half the XP you would get for, say, slaying all the goblins. If you sneak past that guard post three more times, you don’t get any more XP. You only get the other half of the XP when you finally, permanently remove the obstacle. The same goes for any other encounter. If you avoid a trap, but that trap remains active, you get half the XP. You only get the other half when you finally destroy or disarm the trap.

I realize that sort of pushes PCs to deal with living creatures with violence, and I’m okay with that. I’m not going to apologize for preferring combat. That’s why I’m writing this for D&D and not f$&%ing Fate. But it also encourages PCs to go back to problems there were unable to remove completely and finish solving them. If there is a powerful monster they can’t beat but they can avoid, I want them to go back and beat that monster once they are powerful enough. If there’s a trap they can dodge but not disable until they have a certain magical item, I want them to be able to fix the problem permanently.

That, too, is a part of exploration based gameplay. And that also plays into the idea that sometimes the best reward in a dungeon filled with wandering monsters is a safe path or a short cut. So, I’m leaving that option open.


Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

38 thoughts on “Megadungeon Monday: How to Award XP!

  1. I really don’t see the logic in wanting players to destroy traps instead of just circumventing them. If the PCs destroy all traps, you’re discouraging one of the cooler moments of old-school play of luring monsters into a trap that the PCs have discovered or altered.

    • I’d think that would be destroying the trap and killing the monster at the same time, getting the XP for both. That seems right to me – getting extra XP for killing a monster with a pre-existing trap. It’s both more fun and more rewarding.

    • What I see there is that the players don’t have to destroy the trap for full xp. Rendering it harmless to themselves while making it an active threat against wandering monsters would qualify as a permanent defeat of the trap worth full xp. Leaving the trap operational enough to still harm the party, and choosing to not disarm or destroy it to use themselves is a choice the party will have to make, weighing the tradeoff of bonus xp and a safe path vs the utility of the trap as a defensive tool. Since we are assuming that the players will know what the sources of xp are, that sounds like a valid choice.

    • I think angry is using the term “trap” here in a broad categorical sense, as in a dangerous area that will continuously pose a threat that isn’t a monster or a social encounter. Maybe it is a deadly virus that some hags are brewing up that will later be released on the town. By killing the hags and making their dungeon sink back into the ooze, the “trap” has been disarmed. Maybe rickety bridge that will collapse when the PCs are on it, so they have to find a way to get around the bridge. They don’t have to destroy the bridge itself, but they do need a way to reliably circumvent it so they can get back where they came from. that would count as “disarming” the bridge trap.
      I think basically it might lead up to being the “other” column when you don’t know how to allocate the exp for a dangerous situation.

  2. Another excellent piece to the puzzle of adventure building. My mind was filling with ideas for adventures as I read this. Thanks for this illuminating series Angry!

  3. It might be worth mentioning that the DM should give a breakdown of each end-of-session XP award. It communicates to the players where they succeeded and failed, and helps to incentivize them. For example, a DM could say,” “You only circumvented the goblins, so you get half XP for that. If you find a permanent solution for them, you’ll get the rest. The time you spent deciphering the runes is a Discovery and gave you 50 XP each.”

    • I mean, if you want to give XP for them, good. I mean great – frankly, I was sad they took that out. And it’s always the DM’s prerogative to give whatever whenever. But officially it’s not in there. Unless there was some sort of Mike Mearls Encyclical that I missed.

      • The DMG gives rules for how much XP an encounter should be. It gives the saving throws and such for each CR of difficulty. A trap usually targets you via that. It’s a good basis for how much XP that trap should be worth for defeating it. A trap is really just a different kind of monster/encounter. One that doesn’t necessarily chase you and only reacts to your presence, but it’s an encounter all the same.

    • I also think he’s using the term “trap” loosely. A room that the doors close and start to fill with sand and the PC’s have to figure out how to get out is a trap and an encounter. A pit with spikes and a fake covering is part of the larger encounter of the goblin encampment leading up to it and not worth xp by itself but factors into the overall encounter difficulty.

      At least that’s how I think he’ using traps.

  4. I was going to argue with you about the Circumvention piece. And then I parsed through my objections and saw how what I perceived as problems might not be problems if you handle circumvention in a particular way.

    I think you need to define what is a circumvention. If they sneak past, and have to keep sneaking past as they go back and forth, clearly they have not dealt with the problem. This may resolve itself eventually with a failed stealth check, or if the goblins realize someone has been through and become more viligant.

    If they befriend the goblins, then in my view they have permanently dealt with the problem, and should get full XP. Tell me if I’m wrong.

    If they find another way past that can be readily duplicated without risk (a blind spot, trick or bribe, or finding a risk-free way to temporarily disable the trap each time they pass), have they permanently dealt with the problem? Why not?

    If they sneak past, and then find a shortcut so they can avoid the goblins entirely, have they circumvented the goblins or defeated them permanently? Do you then award higher discovery XP for the shortcut, because you only awarded half XP for the goblins the first time through? If the shortcut is part of your fixed progression, do you bother with the extra record-keeping, or just grant full XP from the start?

    I’ve just talked myself into arguing about it again. (a) If the result of circumventing the encounter is that they will have to face the challenge again, isn’t facing the challenge again without any hope of getting extra XP its own disincentive to a temporary fix? (b) If the result is that they will not have to face it again, should they not get full XPs, because ultimately it was not circumvented? (c) If the result is that they may or may not have circumvented the challenge, depending on whether they later find a shortcut, then either (a) or (b) will apply eventually.

    In any of these cases, I am not sure the extra bookkeeping is worth it, particularly in a published dungeon.

    • One major purpose of “circumventing xp”, a purpose which Angry covers at the very end, is providing incentives to return to areas you’d previously avoided and deal with the threat for good. If you just give full xp for circumventing and then never give xp later, even if they defeat it permanently, that incentive vanishes.

      There’s not much more empowering than coming back to something you weren’t able to deal with before and finally kicking its butt. Especially if you can do so with ease. Gives a real sense of advancement. Angry is just providing an incentive for this, to weight the odds in favor of players discovering just how empowering this feels.

      • That is true of some of the scenarios he described, but not all of them. I also note that in the two examples that are as you described (need to be higher level to defeat than you will be the first time you encounter it, or need a specific item you don’t have yet), the designer can predict when they will occur, which would allow the half-XP award to be baked into the encounter.

        I am fine if that is the way it is applied. What concerns me is if it is a general principle of the dungeon that the DM is expected to apply to every encounter. In the latter case, it would just be more work for the DM with, in most cases, little benefit. If it were me, I would not apply it in the goblin guardpost scenario.

  5. AngryDM, why do you say the Deadly column has no use? It’s the Easy column that has no use. If it’s below Easy it’s trivial (and therefore still easy), if it’s above it’s Easy, ergo it has no use. The Deadly column at least tells you what the ceiling is on Hard encounters.

    IIRC, Basic Rules v0.1 used encounter thresholds as ceilings, but by the time the DMG came out they were floors.

    • Hmm – I just double checked the DMG and you’re right that these are diificulty thresholds. They should be at least the numbers given to avoid falling below the desired threshold.

      • If you’ve read lovecraft, it would be apparent that he actually was a huge racist. That aside, if one looks into anything enough, there will be something offensive or wrong with it. just my 2c. (rounded down in canada lol)

        • Yes, and finding a chink in the armor sounds racy even though it isn’t. And making things spick and span sounds racy even though it isn’t. And homogenized milk isn’t gay. Grow up, people.

  6. I was looking at your table of Encounter XP threshold and found that you are tiering the encounters unlike me. I looked upon the XP Threshold per Level and saw that there is some jumps in difficulty that match the Tiers of Play from DMG. For example, you grouped the levels 3, 4 and 5 but even on your table first Encounter XP Threshold you will see that Deadly encounters for level 4 are Easy Encounters for Level 5, so characters of level 5 will cakewalk this entire tier.

    So… I prefer to tier levels this way:
    – Levels 1 to 2: Babysitting levels, players level up in one day and encounter building have tight constraints.
    – Levels 3 to 4: The truly starting levels of 5th edition.
    – Levels 5 to 10: Low level campaign that can be divided in 2 parts of 3 levels.
    – Levels 11 to 16: Middle level campaign that can be divided in 2 parts of 3 levels.
    – Levels 17 to 20: High level campaign that can be divided in one part of 3 levels and Level 20 can be a tier itself.

    • That does correlate with the DMG (pg 37):

      Levels 1-4: Local Heroes
      Levels 5-10: Heroes of the Realm
      Levels 11-16: Masters of the Realm
      Levels 17-20: Masters of the World

  7. I think you’ve got the encounter multiples table on DMG82 backwards. It’s to estimate encounter difficulty, not encounter reward.

    For example, a band of 8 goblins are worth 400xp to the players.
    But due to the number of attacks, it should cost 1000xp from your encounter budget. It’s not awarded.
    So for a party of 4 level 2 characters, they would earn 400xp to divide amongst themselves.
    As a DM, expect someone to die, because you’ve made a ‘deadly’ encounter.

    • Yeah, I was also confused at how Angry completely dismisses the “phantom xp” thing without explaining why. It’s there to account for the fact that a group of monsters make more attacks than a single monster, and so are harder than their combined xp value would suggest. This whole series is great so far, but that one thing I do not understand.

      • But doesn’t phantom XP only come into effect when you’re actually designing an encounter? There’s no distinction between encounters here except between difficulty. We’re not at the point where phantom XP for an encounter can be calculated.

      • I believe that Angry is saying that it makes sense award the “phantom XP” (RAW only used to calculate encounter difficulty) as “actual XP”. So, to design an encounter at a difficulty of 1000 XP, throw in 8 goblins, and award a total of 1000 XP. RAW, 400 XP should be awarded, but that’s actually a bit silly. The amount of XP for a good game should be based on how difficult the encounter is, not simply “a goblin is worth 50 XP.”

    • Rereading the article Angry says that the encounters he’ll design will not exceed the target XP threshold. So if you have multiple monsters their individual challenge will necessarily be lower in order to account for the Phantom XP. All that matters at this point is that the XP and encounters needed to level appropriately are figured out.

  8. I don’t know why Wizards didn’t just use CR in the budget tables instead of XP. CR’s only purpose is to provide the difficulty of the creature (and ultimately the encounter itself). If the amount of XP from a creature is the driving measure of its challenge, then CR becomes mostly redundant (other than being an easy reference for the DM). Seems to me they could have just used CR-related math and gotten rid of the mess that is phantom XP.

    If anyone can provide a convincing argument that XP is better for budgeting the challenge of encounters, I’m all ears.

    • hey, dmForLife.

      CR is listed specifically as one measure: If you have an average sized party of a level equal to CR you don’t risk taking characters out, if you use higher CR monsters you do.

      If you challenge a lv4 party with CR5 monsters, there’s more risk – higher damage output, etc. But if you simply take multiple CR4 or even CR3 monsters, the same doesn’t occur. So, CR is a warning label – using higher CR monsters is more likely to wipe a party even if from XP point of view they should be able to do it.

      Also, the XP progression is not identical to CR progression. A CR2 monster is worth 450XP, a CR3 monster is worth 700XP. This pretty much means that the weighting of their difficulty in encounters is different as the numbers are not simply multiples of another. Adding CRs will result in a different number than in adding XP, so both have value: CR describing an assumption of the deadliness of an individual monster vs. party level, and the XP allowing to do easy/medium/hard calculations with different, not identical weights.

    • The third edition of D&D worked exactly that way, using CR as a measure of a creatures ability, and it was how DMs would assemble an encounter. From what I can tell, the reason it was changed to primarily an XP-driven budget was a design decision that was behind much of 5e’s design: complication:

      Can you tell me at a glance the difficulty of the following encounter?

      2 CR6 creatures
      1 CR4 creature
      8 CR2 creatures

      What would the difficulty be for a Level 6 party? A level 8? A level 4? The math involved to determine this was more complicated than finding the sum of XP, and that made all the difference.

  9. I find it brilliant to award less xp for random encounters, since one of the main problems I had with them was that some players actively looked out for trouble, just to grind levels.

    Do you have any tip on how to explain that in in-game terms? Of course I could explain to the player the game design reasons behind it, but an in-game explanation on why they got 100 xp for killing the five goblins in the throne room and only 10 xp for the 5 goblins who ambushed while they were sleeping would definitely be useful.

    • There is already NO in game explanation for XP. None. It makes no consistent sense and anyone who tries to explain why it does is ignoring something or glossing over something. XP is purely a game construct.

  10. Instead of “backfilling” the dungeon, perhaps making the main entrance require reagents of a certain value (that scales as the dungeon is cleared and/or increases each time the door is opened), would be less work for the GM and equally punishing? You can also prevent sleeping in the dungeon with scary noises, apparitions, bugs, etc…

  11. So let me see if I have this correct: ignore ‘phantom xp’ not just for the purpose of calculating encounter difficulty, but also for rewarding xp? So from what I understand if you use the system the way it is laid out in the DMG, the more monsters you include in any given encounter, the higher your multiplier becomes and therefore the less xp you would dole out each individual encounter. So if i’m not running a deficit each adventuring day, does Angry’s way of doing things shorten the campaign by any significant amount?

Leave a Reply to BeoricCancel reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.