I wanted to start by saying “Welcome to my Dungeon,” but I can’t. Because you really aren’t. See, I don’t like doing this. This is where I toil, where I labor, where I create my perfect games. This is MY space. But there’s only so much you can learn from theorcraft. I can tell you how to pick creatures and how to think about terrain and all that crap, but that only gets you part of the way there. Because understanding how the ingredients work is NOT alchemy. And there’s no real way to explain the alchemy without just letting you see how it’s done.

So, I’m letting you into my alchemy lab and we’re going to build a couple of combat encounters. Four, to be specific. And we’re going to talk about how and why I do the things I do. But, let’s be clear, you are NOT welcome here. I’m letting you in on sufferance. Do NOT touch my stuff. Do NOT make yourself at home. Do NOT get comfortable. I DON’T want you here. Got it?

Good. Let’s get this over with so I can get rid of.

The Platonic Reality of Jozan the Cleric

In many ways, I see myself as a younger, sexier, role-playing game oriented version of Frasier. Not the character, the show. Frasier (the character) was a pompous jacka$% pretending he was better than everyone else, whereas I am a legitimate genius and actually am better than others. But the show was very smart. It was loaded with clever references. Just like me. But unlike the show, I am also very sweary. And that’s a plus.

So, let’s talk about Plato and Platonic Reality. I’m not even going to insist that you trust me that this is relevant to combat building. At this point, I assume you either trust me or you don’t deserve my help.

Plato. Platonic Reality. Plato, the ancient Greek philosopher, believed that there was a reality alongside our own where the pure, idealized concepts of things existed. A thing was just a specific manifestation of the idea of a thing. So, the fancy chair I’m sitting here in my dungeon office lab? It is just a specific thing that has the quality of “chairness,” whatever that might be. It is connected to all other chairs by it’s chairness. We can talk about a specific chair (my awesome chair you wish you had) or we can talk about all chairs, the ideal chair, the Platonic chair.

So, let me introduce to you: Ragnar, Jozan, Lidda, Mialee, and Soveliss. You may have heard those names before. Jozan the Cleric, Lidda the Rogue, Mialee the Wizard, and Soveliss the Ranger are the iconic characters from D&D 3.5. Basically, for each class, WotC created a specific character to use in art, examples, and other media. Ragnar the Fighter, you might recognize from Dragon Warrior IV from the Nintendo Entertainment System. He was a cool soldier who saved a bunch of kids from a demon. He’s on the list because I kind of hate Tordek, the iconic fighter. Don’t worry about the reasons. He knows what he did.

Ragnar, Jozan, Lidda, Mialee, and Soveliss are my Platonic Party. They aren’t real characters, they don’t have stats, and they don’t even need names. I just gave them names because the “iconic character” idea was kind of neat and I wish 4E and 5E had done it. Plus, they make great examples when you’re writing thousands of stupid words of DMing advice.

Ragnar, Jozan, Lidda, Mialee, and Soveliss represent my conceptual understanding of the “standard adventuring party.” They are the Platonic reality of D&D groups and each is a conceptual understanding of what a fighter, cleric, rogue, wizard, and ranger can do. They can be different levels, they function a little differently in different systems and editions, but they are just the idea of Adventurerness and Fighterness and Rogueness and so on.

Having a Platonic Party (whether you name them or not) is vital to building good combat encounters, as you will see. But it’s important not to be too specific about them. Don’t assign them specific abilities. Just have a sense of the sorts of things various fighters can do. Or various clerics. Or various wizards.

For example, Ragnar has high survivability and is a powerful melee combatant who manages a zone of control very well. He wants to be near foes to bring his melee weapons to bear, he wants to take most of the hits, and he wants to control who the enemy has access to.

Jozan has reasonably high survivability as well, but he’s a weaker physical combatant. He plays a support role, empowering members of the party. He can also weaken the enemy and he has some capability to deal with the enemy with ranged attacks. He’s extremely versatile and he functions as a firefighter, doing what needs to be done.

Lidda is mobile, but she only has moderate survivability. She can dish out heavy damage as long as she’s free to decide how to engage the opponents, but if she gets pinned down or can’t act freely, she’s hamstrung. She is short range, functioning at either melee range or a few squares for maximum effectiveness. She doesn’t want to stay close to the enemies unless she has an ally to assist and protect her.

Soveliss is also mobile and has higher survivability than Lidda. He is also versatile, but his versatility is in range. He can function equally well (for the most part) at close and long range, which means he has more freedom than Lidda in picking targets and he can function well even when pinned down or hemmed in. But he’s not as strong as her when she’s at her best.

Finally, there’s Mialee. Mialee is a glass cannon. She can dish out a lot of hurt at range and she’s very versatile in the hurt she can dish out. She can disable, weaken, or just damage opponents, sometimes over a big area. But she is extremely vulnerable. She cannot stand toe-to-toe with the enemy for any length of time. She also loses some versatility when she has to work around her party members, especially when there are small numbers of opponents. Once Jozan, Lidda, and Ragnar have closed in with the enemy, her options are reduced to avoid endangering her friends.

Obviously, you need to modify this information a little bit based on the specific edition and experience levels and such. But those are a good, functional, general understanding. You might notice that what I’m calling a Platonic Reality is actually nothing more than a strategy as I defined it two articles ago. And I could have called it that. But it makes me feel smart and self-important to call them Platonic Realities.

By the way, you might be wondering why THAT party. Why Fighter, Cleric, Wizard, Rogue, Ranger. Well, the reason is because it covers a broad range of strategies AND because it is traditional. And that’s what I need: broad range of strategies. And most parties, somehow or other, manage to mix in all those different strategies and roles. The Fighter and Cleric and Ranger might actually be a Paladin and a Druid, the Wizard might be a Sorcerer or a Warlock, and so on, but you’ve got the basic strategies covered.

The other question you might ask is, when I’m designing encounters, why don’t I just use the PCs instead of a mystical Platonic Party. And I realize I haven’t actually talked about how I use this party yet, but I’ll get there. But the reason I don’t use the actual PCs is because that would be brutal and brutally unfair. Unless there is some in-game story reason for an enemy to know everything about the PCs AND you have decided you hate your players and want to punish them for existing, you NEVER build an encounter to your specific group of PCs. It’s tantamount to punishing them for the choices they’ve made. And it means their unusual choices and strategies and the specifics of the races, classes, and equipment they’ve chosen are never rewarded.

That said, there are a few things you SHOULD do. First, adjust the size of the Platonic Party for the size of your actual adventuring party (Soveliss sometimes ends up sitting home and sometimes the party brings along Seelah the Paladin (because she could kick Alhandra’s a$&, sorry WotC)).

So, invent yourself a Platonic Party, a conceptual understanding of the standard adventuring party that you can pull out to build encounters with.

How to Actually Build a F$&%ing Combat Already

Let’s get down to it. No more stalling. No more concepts. No more theories. Let’s get this over with. I’m going to spell it out. Just remember, though, that I’ve said several times this is a fluid process. You tend to bounce back and forth between the steps, tweaking, adjusting, and changing things. You may not even do things in order. That’s cool.

First, the Seed

Remember way back in encounter building when I explained that encounter building was like planting a seed or some sort of bulls$&% analogy like that? Well, combat building is also like that same stupid analogy. You can’t start the process with nothing. You’ve got to have something. Now, you might have an actual encounter. That is, you might have a dramatic question that needs to be resolved and sources of conflict already. Or you might have a specific monster in mind you want to build a fight around. Or you might have a particular battlefield you want to put a battle into. Or you might have a particular ambiance you want to bring into the game (which comes up often in adventure building if you’re doing it the correct (Angry) way, which I’ll talk about someday).

Now, by the end, you need all of those things. You need an actual Encounter with a dramatic question and a source of conflict. And the battle inside the Encounter needs Ambiance, a Battlefield, and one or more Creatures. Building a combat is actually just about figuring out the right mix of ingredients to go with the one you’re already starting with.

Second, The Platonic Reality of a Creature

Now, in order to get started, you need a creature. Specifically, you need the Platonic Reality of a creature. It will represent that creature’s strategy and what it is capable of. It could represent one or more creatures, imagine them like a Cloud of Creatureness on the battlefield. If you aren’t sure yet what the creature is, leaf through your favorite Monster Manual and find something that fits the bill to get started with. You can always swap creatures later. Remember to look at STRATEGIES not actual creatures.

Third, Whatever Else You’ve Got, Put it Aside

Now, sometimes you go into a combat building with a lot of ideas. Maybe you think you already know everything there is to know about the fight. It’s easy to go overboard though with these ideas. So, try to put them aside and leave them in reserve until you need them.

Fourth, Create the Concept of the Battle

Now, if you have a Battlefield in mind, imagine that now. In your head, just picture the Battlefield as vague blobs of terrain. You can even draw it on paper if you want. Sometimes that helps. If you don’t have a Battlefield (or you have one but you’re putting it aside for right now), imagine Purgatory. Imagine a big, formless, open void filled with nothing. You know that place in the Matrix where they load up their weapons and stuff? Imagine that place. Or a big, empty holodeck.

Now, you’re going to put in the Platonic Reality of the Creature you have settled on. Unless you’re starting with Ambiance that defines things like starting position, just shove it in the middle of Purgatory. There’s the Cloud of Creature. If you have more than one Creature chosen, pick the creature that is most important for the combat and put the others aside. Also, assume there is only one Cloud of Creature, even if you’re pretty sure you will need to split it up into two or three or four. You’ll add those things in.

If you have a particular Ambiance in mind and that requires some sort of terrain, imagine non-specific Clouds of Terrain. For example, if you want the PCs caught while death rains down from above, imagine Clouds of Elevated Terrain for the above. That’s all you need right now.

Fifth, Platonic Realities Collide

Now, in your head, imagine how things play out when the Platonic Party encounters the Platonic Creature in Cloud of Concepts you just created. Don’t sit down and roll dice and play it round by round. Just sort of imagine how the general script of the encounter is likely to play out. This is the most important step and it’s important not to overdo or underdo it. Just imagine how that encounter would play out. What do all the players do?

Sixth, Now Make Life Harder for the Players

After you run the first pass of the encounter in your brain, the next thing to do is get in the party’s way. You know what they are likely to do. How can you prevent them from doing that? I know that sounds like you’re trying to screw the party. Well, you are. I’m not going to lie. That’s exactly what you have to do. And THAT’S why you don’t use your specific party of PCs. Because if you do, you can pretty much wreck ALL of their strategies in one good pass if you’re good at this.

If you want to get technical, the idea is to kick the party over to their Second-Order Strategy. You can think of the First-Order Strategy as the party’s “go to” option, their standard operating procedure. If everything aligns perfectly for them, they simply execute their script and win. And that’s boring as hell. No one wants to play that game because there’s no thought or decision making involved. You want to wreck the general First-Order Strategy so the party HAS TO think of a new strategy and execute it. That’s what makes combat interesting. Decisions.

So, figure out what elements added to the Conceptual Battle will trip up that First-Order Strategy given the elements you’re already starting with. It might be as simple as splitting the monsters in two so the PCs have to divide their attention. Or obstructing terrain so they can’t trace lines of sight. It could be more complicated, like adding a different creature (which is always a fun and interesting option). You could change the starting positions. Add these things to your concept map.

Seventh, Platonic Realities Collide 2: Electric Boogaloo

Now that you’ve tweaked to get in the PC’s way and kick them to a Second-Order Strategy, rerun the encounter in your head and see how it plays out. It should be more interesting and exciting. Keep an eye out for glaring issues though. Most importantly, ask yourself it is an uncontested victory or a unmitigated slaughter. Those are okay, if that’s what you want, but generally, you don’t want those. If you’re looking at either a fight that is far too easy or far too difficult, go back to the beginning and try changing things from the start.

Otherwise, the fight should be good enough as it is. And you can get away with stopping here. But some DMs like to tweak the difficulty a little bit and tinker with a few things.

First of all, if you spot a PC that doesn’t seem to have anything to do (in your imaginary Platonic Party), you MIGHT want to consider facilitating a strategy for them. I don’t generally do this for a couple of reasons. First, it’s okay to occasionally have one PC hamstrung and have to go to lesser abilities or use less effective choices as long as it happens to different PCs every time and every fight doesn’t do it. It actually adds drama and tension to occasionally have someone disempowered.

Second of all, if you want to tweak the difficulty of your encounter, here’s where you do it. If you want your encounter to be easier, look at the way it played out in the second pass and see if there is anything you can add to make it easier. Maybe you can move the creatures a little closer together for the melee fighters. Maybe some defensive terrain for the party or give them lines of sight for ranged attacks or something. Those are fine.

On the other hand, maybe you want an encounter that is a little more challenging than normal. There’s two things you can do, and you can do either or both. First, you can add things that make the PC’s strategies more difficult or time consuming. Don’t actually block the strategies, just add things that make the strategies a little more of a pain to execute. Maybe add some distance. Or give the monsters a slight defensive advantage. Second, you can add elements to the battle that make the monsters’ strategies a little easier to execute.

Two caveats. First, don’t make every battle easier or harder. Stop after the second pass on most of your battles. When the players actually interact with the encounter, they will do things differently from the way you ran it out in your head. They may even have a completely different makeup. That will ensure plenty of surprises and variability as long as you’ve gotten rid of the First-Order Strategies.

Second, and this is so important I’m breaking out the bold and the underlining:

DO NOT UNDER ANY CIRCUMSTANCES EVER BLOCK A SECOND-ORDER STRATEGY

See, you might be tempted to add challenge by actually trashing the Second-Order Strategy, countering it and iterating the imaginary combat again. NEVER DO THIS. You have no idea how difficult that can make a fight. Even tactically strong players might take a round too long to come up with a good strategy and they will get themselves killed. Look, I always say you can run your game any wrong way you want and I always laugh when PCs get killed. So imagine how significantly bad something has to be for me to say it is brutally unfairly cruel. And I’m saying that now. The vast majority of players can’t handle it. And even if they can, it isn’t fun. It just feels unfair.

Eight, Run the Numbers

Now, run your numbers through whatever stupid system your game of choice gives you for encounter building. Add up your XP budget, decide how many creatures are in the Clouds of Creatures, and finalize that s$&%.

Now, occasionally, you’ll run into a problem and discover you can’t make the combat you designed work with the budget allowed. Well, dips$&%, I told you to know your system, didn’t I? You should have a general sense of the power level of the creatures. But everyone makes mistakes. My advice is not to run a combat too far outside what you know to be good numbers. Sure, you might know your players can handle easier or harder numbers, and that’s fine. But don’t go outside of that range.

So, maybe you need to scale back or scale up. You’re going to have to tinker. Maybe even discard the encounter and start over. This is the part where it becomes alchemy and playing around. But my advice is, after you make any major changes, run a Conceptual Clash to make sure it’s still running the way you imagined.

Nine, Finalize the Map

Now, you should have a blobby, crazy, conceptual combat map. It’s time to turn that into a real map. Replace all the Terrain Concepts with actual terrain that suits your dungeon, wilderness, city, or whatever.

While you’re doing this, feel free to scatter some props, furnishings, and odds-and-ends around. Those tend to have a neutral impact on the battle. You can also add a terrain feature if one strikes your fancy a little out of the way of where the battle will actually take place. Players and monsters might use it or it might get ignored, but it won’t change much.

Ten, Figure Out How the Fight Starts

Now, with the map drawn and the creatures finalized, figure out the starting positions. Also, figure out if the starting positions (part of the Ambiance) can change based on the actions of the PCs. For example, if the monsters might have raised an alarm in an earlier encounter, decide how that could affect the starting positions. Don’t run any more conceptual iterations. That’s one of those random events the monsters and PCs couldn’t be ready for and it is going to impact the battle however it does. Let the game play out the way it does from here.

Build Fair, Play to Win

Congratulations, you can now build an exciting, interesting, and tense battle. You’ve also built a battle that is fair. It might be hard. It might be easy. But you’ve built it fair. You’ve given the PCs every opportunity to create victory with their own strategic decisions.

But all of that falls apart if the monsters don’t fight their best. After all, you’ve already figured out what they want and you’ve figured out what their ideal strategies are. Implement that. In short, play to win. When you run that battle, you, in the persona of the monsters, should be doing everything you can to win that fight. You should have a very hard time actually winning. D&D is built not to let you win. So it’s safe to try. But if you don’t try, everything you’ve done for this combat is going to be wasted effort.

Got it? Good.

Now, let me show you four examples of the process and then you can get the hell out of here and we can stop talking about combat forever. These are going to be pretty basic, simple examples, but each one will yield a decent combat that would fit into the beginning or middle of any adventure.

Encounter Number 1: Hobgoblins at the Gate

The setup for this encounter is pretty simple. There is an underground hobgoblin fortress the party is meant to be invading or murdering or something. In a cave, there is a defensive outpost. A gate, basically. When the gate is attacked, the hobgoblins there have the job of raising the alarm and then delaying the invaders as long as they can to give the rest of the hobgoblins inside time to plan their defenses.

First, Seed

The seed is pretty straightforward. I’m starting with an actual encounter, right? I have the dramatic question, the reason for the encounter, and what the PCs hope to get out of it. The PCs must either stop the alarm from being sounded or break through the hobgoblins’ defenses as quickly as possible. Now, I’ll add that I want preventing the alarm to be very difficult. The party will have to recognize the threat immediately and act fast. Otherwise, the real goal is to break through as quickly as possible. You can also assume I’m planning for the rest of the encounters to change based on how long the hobgoblins have to organize.

Second, Creatures

Obviously, the creature I’m going to start with is the hobgoblin. So I have to understand the hobgoblin. The hobgoblin is defensively strong. More importantly, when hobgoblins band together with other creatures (including other hobgoblins), they become more effective melee combatants. They also have pretty good ranged attacks, but they give up some defensive capability for it (using their longbows does not allow them to use shields). Ideally, they want to form a defensive cluster and let the PCs come to them, then cut them down in melee.

Third, Put Aside All Else

I’ve decided to use wolves and goblins with the hobgoblins in the dungeon in general, but I’m going to hold them aside for right now. I’ve also thought about using barricades, standing tower shields, scaffoldings, and other wooden war constructions to emphasize the hobgoblins’ military nature. I will also keep those in reserve.

Fourth, Battle Concept

So, the bare minimum the combat is going to have is an open cave with an exit that has a gate blocking it, the hobgoblin’s gate, an alarm gong near the hobgoblins, and, of course, a cloud of hobgoblins between the gate and the PCs.

Fifth, Fight!

In my brain, I now see how this is going to play out. The PC party enters the room. The hobgoblins dig in in their position in front of the gate, waiting for the PCs to come to them and ready to hold them off in melee. Or maybe they make some ranged attacks while they wait for the PCs. One hobgoblin runs for the alarm, Soveliss or Mialee tries to take out the hobgoblin and either it works or it doesn’t. Assume it doesn’t. Now Ragnar and Jozan charge the hobgoblin line, engaging in melee. Lidda slips around behind them using the open space to her advantage and ganks while Ragnar and Jozan absorb the blows. Meanwhile, Soveliess and Mialee can rain spells and arrows down on the hobgoblins from range and the hobgoblins are tied up with Rganar and Jozan.

It isn’t a terrible fight, but it’s pretty pat and standard. And honestly, the party doesn’t HAVE TO move to the hobgoblins. They can hang back and let their superior ranged fighters deal with the hobgoblins, hamstringing their best abilities.

Sixth, Block

Let’s figure out how to block the PC’s primary strategies. First and foremost, two of the PCs can fight uncontested. Mialee and Soveliss can hang back and use ranged attacks and never be in danger. Now, I can wreck the lines of sight easily enough, but that still won’t do much more than force the ranged fighters to find a better position and then they can start in with ranged death again. What we really need is a ranged answer to the ranged PCs. That is, another creature that (a) can attack the ranged combatants and (b) draws fire away from the hobgoblins. Goblins are accurate ranged attackers and already on my list. So we’ll add a cloud of goblins.

Now, Lidda can outflank the hobgoblins easily and backstab them, cutting through their defenses quickly. We need to make the PCs work for their flanks, pulling the hobgoblins out a little. Maybe give them a chokepoint to hold.

Finally, Ragnar and Jozan are doing exactly what the hobgoblins already want, facing their defensive line. So there’s no need to make that harder. But we need to worry about no one coming to face down the hobgoblins. In theory, the party could hang back and try to weaken or eliminate the hobgoblins at range. They probably won’t, but the hobgoblins would be ready for that. So digging again into my list of put aside elements, I’m going to put a couple of those big rolling tower shields (basically obstructions) near the chokepoint. If the hobgoblins come under fire and no one is fighting them in melee, they will take up cover behind there and use their longbows, which will encourage the melee fighters to come forward.

Seventh, Tweak

So, the fight will probably play out much better now. The melee PCs will have to move up while taking fire from the goblins. The ranged PCs will engage the goblins. The goblins could support the hobgoblins or fire upon the ranged PCs. Lidda might try to find a way to circle around the hobgoblins or she might try to eliminate the goblin archers. And the goblins might harry Lidda with arrows while she moves into position. The party might try to draw the hobgoblins out of the choke point.

But I do want to tweak the difficulty just a little bit. The goblins are kind of weak, so I want to make them a little harder to get to. So, I’m going to put them on a rampart (a wooden scaffolding with walls) slightly above the battlefield. That puts some distance between them, slowing down melee combatants and gives them cover.

Finally, I will put the alarm gong within a move of a goblin’s starting position. A single well-placed arrow or spell could eliminate the goblin if the party gets the jump on them. It’ll be hard to prevent the alarm from being raised, and it will take a little luck, but it won’t be impossible. Especially if the party scouts stealthily and gets the lay of the land before they get spotted. Think of it as a reward for working cautiously and gathering intelligence.

Eighth, Numbers

This is not a lesson about how to build encounters using the math in D&D 5E (though I did a video about that). Let’s assume 4th level PCs. So, for a moderate encounter, we can fit about 3 hobgoblins (100 XP each) and 4 goblins (50 XP each).

Ninth, Map and Tenth, Starting Positions

I convert my concept map into an actual map. Easy enough. But how big is it?

The PCs will be approaching from the opposite end of the encounter area as the gate. I want the melee to be joined quickly, but not too quickly because the goblin archers would like to get some shots off. So I will start the enemies and the PCs about two moves apart. The alarm will be within one move of the goblins. The party has one quick chance to interrupt that, as mentioned. If I wanted to make it easier to interrupt the alarm, I would move it farther away. I will split the goblins into two different groups as well. Finally, I will give the ranged PCs a choice. I will put some covering terrain in the form of natural columns or cave formations within two moves of the entrance. If a ranged PC is willing to give up a round, they can find some cover from the ranged goblins.

Encounter Number 2: Ghoul Patrol

For this encounter, I’m just going with ghouls. I know the party will be tromping through sewers and I want a nice surprise encounter I can drop on them as a fun surprise. Apart from sewer and ghoul, I don’t have much to go on. It’s just that I love ghouls.

First, Seed

Ghouls. Done and done. Also, sewer, but that’s more for theming.

Second, Creature

Also ghouls. See how easy this is?

The thing with ghouls is that each ghoul has a chance to paralyze and incapacitate a PC for several rounds with each attack. And ghouls are individualistic cannibals. At least, that’s how I see them. And I’m always right. So any given ghoul is more concerned with disabling someone and dragging them away to be eaten, as long as it’s safe to do so. Strategically, the ghouls will probably spread out and fight like individuals, each trying to paralyze someone and drag them off.

Third, Put Aside All Else

I have ghouls and sewer. I’ll put sewer aside and worry about ghouls.

Fourth, Battle Concept

Lacking any firm idea of a battlefield, we’ll start with an open area with ghouls on one side, PCs on the other, and see what happens.

Fifth, Fight!

Ragnar, Jozan, Lidda, Soveliss, and Mialee are walking through a sewer when they come upon some ghouls. How does it play out? First thing’s first, Jozan is going to spend his first action trying to turn the ghouls and he might drive one or two of them off, removing them from the fight. That’s fine. That’s what clerics do. He chose the class, he deserves that one.

Then, if the party is smart, they will fall back into a defensive posture with Ragnar and Jozan forming a wall for Soveliss and Mialee to hide behind. Lidda should hang back at first until Ragnar and Jozan get things under control and turn whatever ghouls are going to be turned and then break out and try to murder who she can safely. If he can, Soveliss will hang back with Mialee so the two of them can make ranged attacks with impunity. Lidda might also go that route.

Sixth, Block

Honestly, that fight is okay but it isn’t great. With Jozan instantly reducing the ghouls’ numbers and Ragnar and Jozan establishing a line, the party is in control. All it really needs is a slight tweak or two. First of all, Jozan and Ragnar should not be able to establish a line. Instead, the ghouls should come from multiple directions. In fact, unless the party somehow detects them, they should start surrounded. That way, the party is forced into a circle rather than a wall, with either Soveliss or Lidda forced into a defensive posture.

Seventh, Tweak

The fight is pretty cool with the simple change to being surrounded (with the caveat that the party might detect the stealthy ghouls BEFORE they walk into the kill zone and the ghouls surprise them by ambushing from multiple directions). But let’s assume you want to make this a little extra tense and frightening, as I do, by making things a little harder on the party.

Let’s make life a little tougher for the party in general by have the ghouls start close to the PCs. The avenues of approach are guarded and the ghouls start within one move of the PCs. That makes it harder for the PCs to reposition themselves as the fight starts. We can counter this by giving them a chance to somehow detect the ambush, possibly hearing the ghouls move around. This works a little against the ghouls too because they will have a more difficult time scurrying from PC to PC and spreading their attacks out in the tight quarters, but it will also work for them. See, they also have excellent avenues of retreat down which to drag paralyzed PCs. See, the PCs are likely to recover from paralysis after only one or two rounds, based on how the numbers line up. So separating the PCs from the party quickly is very effective. But the tight quarters also make it easier for someone to intercept a ghoul dragging off a party member.

If you really want to make the battle more difficult for an expert party, or if you really hate your players, you can have the ghouls emerge piecemeal. One or two at a time. That makes it more difficult for Jozan to decide when to turn. But more importantly, since any given party might or might not be able to turn undead, it means the party might adjust their position to deal with an attack from one direction and then get surprised by an attack next round from another direction. But that could be devastating to some parties.

Eighth, Numbers

If we assume a 5th level party, 4 ghouls (200 XP each) makes an encounter halfway between easy and moderate. Considering I’ve implied that this is a side encounter, just something to scare the motherloving f$&% out of the party while they pursue other ends, that seems fine.

Nine, Map and Ten, Starting Positions

 I want the open area in the middle of the battlefield to be tight, but not impossible to move around in. An area that is about a move across is just fine. But I need an approach that lets the ghouls appear pretty close to the party. So, what if we have some sewer pipes that exit into the open space that are partially boarded up. They have about two feet of space on the bottom that the ghouls can crawl out of. That means a ghoul can also drag a PC away into the pipe, forcing another PC to either crawl after them or hack through the boards. That creates a nice, scary moment.

As for how the fight starts out, obviously, the ghouls are hidden out of sight and will attack once the party gets into the middle of the area. But, the party MIGHT hear the ghouls sneaking into position (even though they don’t breathe) which will negate the surprise but probably won’t change the starting positions.

Encounter Number 3: The Rescue

So, this is one of those encounters where I know how it starts but not what is in it. I knew the party was going to be traveling across trackless wilderness, in this case, a very hilly coniferous forest. And I wanted them to meet an NPC. I wanted to establish the NPC as rugged and brave but also wanted the NPC to be overwhelmed and in need of help. I wanted to give them a chance to rescue or lose the NPC by their own merits.

Basically, I wanted the scene to open like this: the party emerges from the tree line into a long, narrow clearing. Suddenly, from the underbrush at the far end of the clearing, the NPC explodes into the clearing, panicked and stumbling, injured. Then, something horrible bursts into view behind him, set on giving chase and bringing him down. The party is unnoticed, but they have to react quickly to intervene.

First, Seed

In this case, the seed isn’t really the battlefield, it’s the opening. It’s the ambiance. I have a scene, a mood, I want to create. Basically, I just have an open space with defined edges that allow the battle to open the way I want to.

Second, Creature

This is a tough one. Literally, I could have anything chasing the NPC and this is where that list I briefly mentioned of game elements to include in the adventure comes in handy. I could use a single big thing like a manticore or some other forest beast, but I will always prefer a group of monsters to an individual thing. A pack of things, like wolves, might be interesting, but I find humanoids more interesting creatures in general. Since I’ve already done hobgoblins and goblins, let’s do something else. Either orcs or gnolls. Arbitrarily, I’m going to decide on orcs.

Strategically, orcs are bruisers, pure and simple. They don’t have very impressive defenses, but they have some staying power. Most importantly, they can cover a lot of ground quickly (with a bonus action that grants them extra movement) and they deal a lot of damage in melee. They have javelins they can throw, but they’d much rather close the distance and get up close and personal and murderous if they can.

Third, Put Aside All Else

Done. Easy. I don’t have much else. Just an open battlefield with some orcs chasing someone. Now, I’m not going to put aside the someone because they figure into the battle. The NPC is essentially the finish line. This is a rescue, after all.

Fourth, Battle Concept

In the middle of the clearing, I’m going to put the NPC. About a move away, I’ll put the orcs. About a move away in a different direction, I will put the PCs.

Fifth, Fight and Sixth, Block and Seventh, Tweak

Now, I’m going to do something a little unusual here. I’m going to go back and forth between fighting and blocking and tweaking a bit. As you will see, this fight doesn’t work out quite right at first and it needs some back and forth. So this will give you a good sense of how to combine the steps. Because, often, you end up doing just that.

A lot depends on whether the PCs or the orcs win initiative. If the orcs win initiative, they can immediately close with the NPC and lay into it, dealing five attacks. And even if they don’t directly close in, they will hurl javelins. That means the NPC will probably end up dying or dead immediately. If the PCs lose initiative, they don’t have a chance to rescue the NPC except for winning the fight as quickly as possible and then stabilizing and healing the NPC. Now, that is okay, but it limits the PCs’ agency. Notice it doesn’t matter what the PCs do.

Now, even if one or two PCs manage to put themselves between the orcs and the NPC before the orcs go, the remaining orcs can run circles around them with their bonus move and still cut down the NPC, depending on how many orcs there are, of course. Increasing the distances involved doesn’t make anything better. The orc’s bonus move is a problem.

This is one of those instances where I am actually going to reject the fight before I go any further, and redo things. Let’s switch it out for gnolls. And I’m going to do something else. I’m actually going to hamstring them a little by rearming them. I’m going to take away their longbows and switch their spears out for hammers or axes. The gnolls do not have a ranged attack option.

Now, running the fight, it is better but still not great. The PCs and the gnolls can both close in on the NPC in one move and then do something, but if the gnolls win initiative over most of the party, they are going to be able to kill the NPC. Then, the heroes are going to drop back into standard tactics: Ragnar and Jozan establish a front line, trying to control the gnolls, Lidda tries assassinates whoever she can partnering up with the melee heavies, Soveliss and Mialee will blast the gnolls at range. Jozan MIGHT try to heal the NPC during the battle, but with the gnolls active on the battlefield, that’s dangerous and the party is probably better off (and more likely) to try and mop up most of the fight first. It still doesn’t feel like a “protect the NPC” fight.

The fight still isn’t running quite the way I want it to, so we’re going to give the concept another tweak before we move forward. Let’s change the distance. What if the NPC is two moves away from everyone?

Now, the gnolls will close the distance to the NPC, but they won’t be able to attack right away. Ragnar and Jozan will definitely also close with the NPC to defend him. Mialee will probably stay back and throw spells. But what about Lidda and Soveliss. Things get tricky there. See, Lidda would probably like to hang back for a round and wait until she can team up with a melee heavy and Soveliss would prefer to stay out of harm’s way and use ranged attacks, but if the gnolls have Ragnar and Jozan outnumbered, they have to decide whether they can risk that. Soveliss and Lidda may end up having to close with the NPC just to provide extra defense, even though its not optimal. Meanwhile, Mialee hangs back and one of the gnolls may break off and try to take her out and she will be defenseless. That has a much better feel to it.

The only issue is that Ragnar and Jozan have it too easy and Mialee may be in a lot of danger. Maybe you like that part about Mialee being in danger, but I’m going to balance things a bit.

Instead of starting the party two moves away, what if I start them one move away but put them at the top of a ridge. That is, they emerge from the tree line atop a ridge overlooking the chase which happens in a narrow cleft or valley between hills. If Ragnar and Jozan (and Lidda) go barrelling down the hill, they have to make a skill check. Failure means they tumble down the hill, take a bit of minor damage, and have to use their action to hustle to close the distance. Success means they close the distance and still have their action. Soveliss might stay on the rise and use ranged attacks because it is an attractive bit of terrain, but then Ragnar, Jozan, and Lidda might get overwhelmed. Mialee is relatively safe on the hill. But, to make sure she isn’t completely ignored, we’ll give one gnoll back his longbow. Yes, he could use it to assassinate the NPC, but if we assign the NPC enough HP to survive one arrow, that should be fine.

Now, one of the brutal but neat things about this fight is that gnolls get a bonus move and lesser attack if they drop someone to zero HP. So, if the PCs let the NPC get dropped, someone is going to get attacked. And if one of the PCs get dropped, the NPC might get a nasty surprise.

Eighth, Numbers

Let’s assume the PCs are 3rd level and there’s five of them. So, a Medium fight for them is 750 XP worth of challenge and a hard fight is 1,125 XP worth of challenge. If we assume there are three to six gnolls, each gnoll eats up 200 XP of the difficulty budget. So, four or five gnolls is a good number depending on how rough we want to make this on the PCs. If I was going to make all the gnolls melee only creatures, I would definitely say four because five is likely to overwhelm the party and destroy the NPC. With one of them being ranged, assuming the gnoll hangs back, doesn’t carry a melee weapon (other than their weak bite), I would be more willing to have five. I’m going to assume the archer is a female huntress and the others are male gnolls who serve the huntress a bit like hunting dogs. That seems cool.

It’s important to note that the difference between four and five is actually bigger than the numbers in this particular fight because the party’s ability to tie up all the melee gnolls and protect the NPC is what is really being tested here. Parties that excel at clever plans, teamwork, and battle control can handle five, but other parties might find five gnolls too challenging. If I were writing this encounter for a party I didn’t know, I would stick with four.

Ninth, Map and Tenth, Starting Positions

The map and the starting positions are already essentially worked out. That happens sometimes. This particular battle doesn’t need any other moving pieces. The creatures and the situation create enough to make it interesting. In order to spare ranged combatants that stay on the ridge, I will let them take cover in the tree line.

I should also note that putting the PCs above the opening the scene also aids the ambiance. It is empowering and makes them more likely to rush into the battle to aid the NPC, ironically giving up a favorable position and some power over the battle. It’s a neat little mood setter.

Encounter Number 4: The Tree Root Cave

This last encounter is one that starts off based entirely on a map. To set the background, I have a series of natural caves underneath a giant tree. The dungeon itself is partially flooded and filled with giant tree roots that form bridges and passages through multilevel caverns. And I need to do something at the entrance.

First, Seed

So, the seed is a battlefield. In this case, the tunnel that descends into the cave opens onto a tree-root bridge that crosses the chasm (and suspends a flat-topped rock island) and then winds down the wall to the ankle-deep pool of water below. The ceiling is covered with hanging roots. I don’t know what lives here just yet. But I just want a neat little set piece to set the tone of the dungeon.

Second, Creature

I sit down and leaf through the list of monsters looking for anything that might suit the dungeon. My initial thought was a shambling mound, but that isn’t in the D&D Basic Rules. So, maybe a troll would fit this environment. So, let’s start with a troll.

With their regeneration, trolls are damage sponges. The party has to crank out a lot of damage every round to overwhelm the regeneration or shut it down with fire or acid. In the meanwhile, trolls have a moderate armor class for their power level, but they can also put out a lot of damage and they can engage multiple targets at a time.

As for the reason why the troll is fighting, honestly, we’re going to go with hunger here. I admit this is sort of lame, but that’s the problem with generating these encounters in a vacuum. Realistically, if the troll is seriously injured, it will probably try to escape. Outside if possible or deeper into the caves to recover (quickly).

Third, Put Aside All Else

Done. Troll in the root cave.

Fourth, Battle Concept

The hungry troll starts out on his spongy, mossy little island at the bottom of the cave. The party enters from the top.

Fifth, Fight

Assuming the party sees the troll from above and the troll sees them coming, everyone is on equal footing. Ragnar and Jozan, if they are smart, will run down to hold the troll low down on the root. Lidda will clamber or jump down to try and get behind the troll or she might hang back because the troll will deal a lot of damage and use lesser ranged attacks. Mialee and Soveliss will stay safely behind the chokepoint and lay damage down on the troll, presumably with fire spells. The troll will soak up a lot of damage for three or four rounds and then it will be dead.

Sixth, Block and Seventh, Tweak

The chokepoint is a major problem. The terrain works against Lidda, but it works for everyone else. Jozan and Ragnar can just hold the troll anywhere along the root path and Mialee and Soveliss can attack with impunity. Now, part of me is tempted to dump the troll altogether and do something else, but I suddenly have a couple of nasty ideas.

First of all, why does the troll live in water? Because water protects it from fire and acid attacks of course. So, if it starts taking damage, it will withdraw into the water and submerge itself. Maybe there are some deeper parts of the lake, difficult to see, that function as pit traps. The troll can hide under the water for a round or three protected from fire and acid attacks and letting it’s regeneration run. That forces Ragnar, Jozan, and Lidda to chase it and engage it in the water. The troll is not a natural swimmer, but it can hold its breath for several rounds, lurking under the water.

Second of all, we need a way to make life dangerous for Mialee and Soveliss. What if there is something else living in the cave too? In the ceiling? Hidden amongst the roots? What if there are some stirges up there? See, the stirges present an interesting possibility because they could just lair here, flying in and out to hunt the forests, easily avoiding the troll. Or, at night, when the troll sleeps, maybe they feed off of the troll. After all, the troll recovers quickly from their blood sucking. And perhaps the troll doesn’t even notice being bitten in his sleep. So the stirges like their little lair.

So, we have Ragnar, Jozan, and maybe Lidda blocking the troll on the bridge. Once the troll takes some damage, it withdraws to hide in the water, drawing them out. Meanwhile, stirges descend on Soveliss and Mialee. Stirges don’t do much damage, but they are an action denial tactic. Once they get attached, you need to expend an action to get rid of them. That forces Ragnar, Jozan, and Lidda to each decide whether to keep the pressure on the troll or rescue Soveliss and Mialee while the troll spends rounds regenerating in the water. Now, that sounds good.

Eighth, Numbers

Numbers are going to be a little tricky here. Since I am looking at a CR 5 creature, we can assume my party is around level 5 or 6. And I’m going to be honest, because of my tendancy to add new creatures to the fight, I tend to look at creatures a little lower than my PCs. So, let’s assume I have a 6th-level party. Seeing as how I’m going to give the troll at least a couple of friends, the XP is going to count double, so I’m already at 3,600 XP worth of difficulty for my 3,000 XP – 4,500 XP encounter (medium to hard for five, sixth level PCs). Each stirge is going to add about 40 XP worth of difficulty to that. I don’t want to add more than 5 stirges because the fight will get too big, so that’s a 3,800 XP difficulty fight. That’s a tough fight to start the dungeon, According to the rules, that is about a fifth of the adventuring day, though. So as long as my dungeon has about five encounters in it, this is a nice, strong opening to the dungeon.

Ninth, Map and Tenth, Starting Positions

These have already been decided. I just need to add some deep “pits” of water where the troll can work and where a PC can suddenly fall underwater and have to swim. Fortunately, they won’t risk any damage, though. And perhaps the troll keeps some valuable shinies in one of the pits.