I Just Finished DMing My 526 Day Dungeons and Dragons Campaign – Here’s What I Learned

Writer’s Warning: This entire thing got so dreadfully far away from me in terms of scope. I’ve officially hit 22,000 words, which is dangerously, dangerously close to “publishable book” length. You are now officially not allowed to complain about the length of this because you’ve been warned.

I must confess that this one is largely just an excuse to flex. I am very proud, and I feel like I knocked my campaign finale out of the park*. I am on an emotional high that has lasted multiple days (very rare). I also feel like I have plenty of Hot Takes™ about the hobby and campaign design at large, so now feels as good a time as any to flirt with making a writing pivot all-together as the Trump/COVID defined era of politics continues to depress me to the point of avoiding engagement all together for the sake of my mental health.

*I live in a constant worry that my own self perception is horribly inaccurate. I tend to be pessimistc in this regard, and my players also thought it ruled. That said, consider this a fair warning i might just be a delusional fool.

Before we begin, let’s set expectations – something every good game master* should do.

*”Game Master” and “Dungeon Master” mean the same thing but “Dungeon Master” is technically the correct term for someone who was playing Dungeons and Dragons. However, i’ll mostly be referring to this i as “Game Master” going forward since it’s a more generalized term, and i’m intending for this advice to be mostly applicable to any TRPG.

First, the advice/lessons i’ll be dispensing are broad and philosophical. This is not a list of spells to look out for or monsters from the manual that will make your encounters super dope. That stuff is useful but also plentiful in the sea of TRPG instructional content. I’m also extremely inexperienced in comparison to the sorts of folks who typically dispense TRPG guidance, so I lack the encyclopedic knowledge one really should have for that sort of thing. The perhaps ego driven hope here is I can provide unique insight as a hyper competitive*, hyper fixating psycho can provide with a set of relatively fresh eyes.

*I mean competitive in the sense i see other DM’s online and decide I must become better than them in spite of being far less experienced with appreciably far worse talent.

Second, this is not a “how to” or a “beginner’s guide to DM/GM.” Similarly, I think the world of TTRPG advice content already covers this sufficiently and better than I could. If I wrote up something, it would just be this reddit post* but not as good so if you’re interested in that sort of thing, click there.

*Plus some basics on safety tools and how to make sure you’re respecting your players boundaries, particularly when it comes to mature content. In this regard, I feel unqualified on two accounts: I DM for a really tight-knit group and do this stuff less than I should, and I like my CIS-white maleness makes my perspective lacking in this regard.

Finally, the hope here is that most of this advice can be at least somewhat applicable to most RPG experience. The idea is that if you’re playing Pathfinder or Burning Wheel instead, a lot of this stuff will still be helpful.

What I Learned About Combat

The same gentleman who wrote the reddit post pointed out on his youtube channel that the vast majority of the Dungeons and Dragons rules and manuals pertains specifically to combat. That is to say, making an epic, narratively satisfying, character driven story that doesn’t just feel like a string of random events is a bit like pounding a square peg into a round hole. I think that is an accurate assessment, but one that also directly conflicts from the type of story driven experience most new players today are expecting after being introduced to the hobby by shows like Dimension 20 and Critical Role. Incidentally, I am also much more oriented towards the story side of the hobby than I am the strategic, combat-driven side of the hobby – so i’ve developed some solutions.

Make the Combat Tell A Story

In movies, we generally understand that the action scenes exist to support the story. An action scene should impact the plot, inform character development, or be the cathartic payoff to those sorts of things. If it doesn’t accomplish at least one of those things, it probably* shouldn’t be in the movie. Yet, out of the dozens of published adventures i’ve read and samplings of online campaigns i’ve watched**, it seems this principle is rarely applied to combat.

*I say “probably” because for something like, say, The Raid, the action scenes are the point and the plot is largely superlative. This is the equivilant to a more traditional, dungeon diving type of DnD campaign. I enjoy both of these sorts of things from time to time, and I certainly pass no judgment or disdain if that’s your cup of tea.
**When I say watched, i mean “skimmed for ideas to steal.” Incidentally, this also means I am passing no judgment as the aformentioned sort of “dungeon dive” type of adventure may very well have been what some if not most of these campaigns were aiming for.

My lessons learned on this can be divided into two main categories: context and mechanics.

In regards to context, this is largely about the “why.” The first thing to consider in designing any combat encounter in a story oriented campaign is why (in a story sense) the combat is happening. Once you know the why, use this to inform your decisions in regards to designing and operating your encounter. I am in no way suggesting every random raider should have a tragic two page backstory. Combat can be made to feel immediately varied and immersive by simply taking the basic premise – say, highwaymen are robbing the party – and apply some light realism and logic to the situation instead of it merely being an excuse. To see what I mean, let’s get really into the weeds of this theoretical very basic concept of an early level random traveling encounter.

Our encounter now has a social element to it. The game master can spice it up by adding a little character to our leader of this highwaymen. We also have some options as far as engaging with the encounter. If it’s an ambush, someone can pass a high difficulty perception check to get one up on them. If they succeed, they can choose to surprise them instead OR they can risk losing this opportunity and instead attempt to bypass the encoutner with a stealth check. Once the social preamble to the fight begins, there’s no reason our highwaymen have to demand every wordly posession of our party. Requesting a moderate sum of gold to pass unharmed instead presents another interesting choice. Once combat ensues, our highwaymen’s behavior should be informed by the fact they are real, normal people who just want to make money in an immoral way. If the fight goes poorly for the party, they can surrender and lose their precious magic items, but save their lives. If the highwaymen bit off more than they can chew (likely), instead of more obligatory attack rolls in a scenario that is clearly already decided, they can retreat. In that scenario, the party as yet another opportunity to engage with story and character. Do they feel bad for them because they are in over their heads and in poor economic straights, or are they vengeful and slaughter them?

That’s just all off of the top of my head. There are plenty more interesting choices that can be applied to a very basic, filler combat type scenario so that it is still rooted in story. The point here is even if the stat blocks are as vanilla as a set of saving throws and a basic hit modifier and damage value, tons of story can be unearthed if you take the time in preperation and be mindful in operation that even nameless NPC’s do, within the fiction of your world, have opinions, ambitions and goals.

The other tool to give our encounters a sense of believability is mechanics. Although this feels a bit counter intuitive – the hard numbers of mechanics and undefined emotional concepts of story are often ptited against one another – a lot of these ideas are already baked into the better designed monsters of Dungeons and Dragons. The classic 5th edition Beholder is a great example of this. Its signature anti-magic cone and death rays are extremely evocative of the artistic design of the creature. It has no run speed, instead utilizing a glacially slow fly speed of 20 feet per round, evocing an intelligent predator that relies on these eyes to freeze its prey and the classic horror monster trope of advancing slowly towards you. Really, the only thing I’d ever feel inclined to change when utilizing one for myself is to lower it’s Armor Class to say, 15 and increasing it’s Hit Points to give it that “Jason Vorhees just won’t die” feeling that the slow movement speed evokes.

Which is to say, I strongly endorse homebrewing your own monsters/stat blcoks, or at least heavily monkeying with the existing ones to better enforce the story ideas to be communicated. For example, my campaign’s secondary villain is a flirty, sophisticated bisexual vampire, and his combat style reflects that. His fighting style is about speed and precision, and he backs up sword play with some magical abilties which reflect his backstory as a former wizard hunter.

To communicate these ideas mechanically, he has a relatively low damage weapon attack, but he attacks three times in one turn and he has an expanded critical hit range. (Rolls of 18 or higher result in a crit). He has a signature ability to create 1HP clones of himself, who can also make these weak, high crit rate weapon attacks. He has a high Armor Class but relatively low Hit Points, suggesting someone who can evade attacks with speed and precision, but is not particularly inclined to take a hit well. He has a parry reaction when an attack misses to get yet another attack off, and he has a decent bag of spells which are suited to fighting other magic users. To round out his kit, he receives a small amount of healing when he deals damage to enforce the idea of his vampirism, and he has a trait which increases the effectiveness of his counterspell to enforce the wizard hunting backstory.

I, a lazy individual who never writes anymore, would never suggest this much* shoe leather be applied to every stat block, but applying one of two of these types of ideas to a lot of your stat blocks does wonders to the storytelling of your combat. Make a big slow guy with low accuracy but hard hitting attacks. Have a pack of ravenous wolves retreat automatically when the pack leader is slain. Have a cocky villain who doesn’t use any of his spells until his HP is halved and he realizes the heroes are not to be taken lightly. If my last two suggestions sound like they’re blurring the lines between context, story, and mechanics, that is very much intentional. Combat can be used to create a video game-ish mechanical challenge for your players, but it can also be used for worldbuilding and storytelling. Which leads me very naturally into my second combat lesson.

*This character was originally designed to be the second to last boss in the entire campaign. That, uh, didn’t happen.

Not Every Encounter Needs to be Challenging or Life Threatening

If this seems like a pretty obvious conclusion based on all of the above, that’s because, well, it is. That said, I do think there is enough of a distinction between the sort of encounter specific momentary storytelling and where combat scenarios fit within the grand scheme of a session, story arc, or entire campaign that it’s worth discussing.

While I feel like in my first attempts at DMing i did an okay job of making encounters with humanoids, animals, and other various entities feel like they were beings with interiority and unique traits, basically any fight my players didn’t cheese* in some way ended up being a nail biter. They were, with minimal exceptions, grind’em out slug fests that basically always resulted in one or more players going unconscious and every last spell slot being expended. While at face value, if i tell you there was one party member death and zero player character deaths throughout the entirety of this first campaign, that might sound like it was perfectly tuned, by the end it became an unfun slog for everyone involved.

*This is slang for doing something the design of a challenge doesn’t account for, resulting in the players bypassing the intended challenge. In complex games like most popular TRPG’s, this is inveitable because no one can account for ALL of the various spells palyers can potentially have at their disposal.

The failure of combat the first time i ever ran a campaign can be boiled down to two things: pacing and relativity.

If there is one thing I could fix about The olDMe™, it would be that i was really bad at pacing. This applies to both the different activities and story beats* throughout my first campaign and to the difficulty of my encounters. In regards to the latter, the macro design of most of my later first campaign sessions were failures for the same reason working with maximum effort for a full 8 hours at most jobs is unrealistic. The basic idea of a mental battery should be accounted for in session design. Most players want to be able to go into brain neutral, check their phone, or crack out of character jokes with their friends at times during the campaign.

*I’ll get into the other ideas in session and campaign pacing very shortly.

However, the even larger problem is relativity. If the difference in challenge between a fight with soldiers at the front gates and the final boss of the castle is very small, and the emotional intensity becomes non existant even if the players remain focused and intellectually engaged. It becomes the “And when everyone’s super, no one will be” problem.

I watched in real time as my first campaign’s original strong enthusiasm and enjoyment decayed for everyone involved. Making every fight hard meant that my players began seeking ways to bypass or cheese encounters just to get a relief from the brain strain. It pushed them into making mechanically driven decisions instead of character driven decisions, which, of course, drove me to make fights even harder so they were “interesting.” My players would regularly get frustrated when they spent half a session healing of unconscious because, in my mechanically driven worldview, the fight was very easy if no one even dropped unconscious*. It eventually just became a black hole where having mechanically complex, challenging fights consumed my entire prep time from week to week, and the last session was a 10+ hour grindfest in a castle that went from one fight arena to another.

*5th edition Dungeons and Dragons is extremely lenient about hitting 0 HP once you’re at about level 5 or so. It’s kind of a smoke and mirrors way of making people feel like they’re in danger with death more or less being off the table except for in outlier cases such as unconscious healers or area of effect damage. The problem is, that while a player may drop unconscious in a fight three times and be at no realistic risk of death, if they spend 4 out of 8 rounds in a combat encounter doing nothing, they’re going to be having an extremely unfun time.

Luckily, by the end of the blog post title’s campaign, I got much better about this. Forcing myself to make fights easier meant I was forced to make combat mechanically engaging in more creative ways than difficulty. That brings me to my next lesson.

Make Objectives in Combat Dynamic and Varied

When you’re approaching encounter design in a mathematically and mechanically driven matter like the olDMe™*, the crux of success and failure quickly devolves into hit points and damage – life and death. That ties directly into the prior lesson I learned. If the main dramatic crux of an encounter is driven by hit points, than fights can really only be made more or less interesting by how threatening they are from that mechanics driven standpoint.

*I will not apologize for this pun.

None of this was obvious because there are very few TTRPG systems that don’t orient themselves around this idea. Even something heavily story based like Call of Cthulu pretty much* boils down to this idea mechanically when it comes to combat. Hell, I don’t think i’ve read** a published adventure that had any other sort of tension in combat other than “the bad guys run away.” Being the generous*** guy that I am, I want to offer a handful of examples of what combat can be oriented around besides the players’s hit points:

*Not an obvious observation on first glance, but really the only difference in how Call of Cthulu orients its combat versus Dungeons and Dragons 5th edition is Call of Cthulu weighs the balance heavily in favor of the monsters. It’s still largely “you live or you die,” but CoC almost always answers this with “you die.”
**”Read” usually means skimmed for ideas and stat blocks to steal.
***Examples are honestly the best way to illustrate what I mean. I’m really not trying to flex here (for once)

• A Pursuit or Chase: This is a classic action adventure trope that I don’t see utilized very often in TTRPG’s. The nice thing about this one is it’s extremely versatile as far story reasons the thing can be happening. A patrol can be trying to run back into the castle to sound the alarm and make the combat scenario VeryDeadly™ and VeryHard™. A group of mercenaries is trying to get away with the captured prince or princess. This has the added bonus of giving your Monk or Rogue type players a chance to show off their usually underutilized mobility talents.
• An Escape or Retreat: The same idea, of course, works in reverse – it just takes a bit more deliberate thought applied in a tapletop gaming context. After all, players generally want to be feel heroic (or anti-heroic) and badass, so they’re disinclined to run away just because a big scary lad is approaching. The idea is to make it escaping feels exciting rather than an admission of defeat. A few examples might be:
  • A Prison Break: If you’re breaking a valued NPC or yourselves out of prison, you’re probably not gonna stick around, right? Escape here is the goal.
  • A Collapsing building/dungeon/terrain: Can’t fight inanimate objects, so running to avoid tons of fall damage is the move here. Errupting volcanos also work.
  • A Wildly out of level monster: Goes without saying KnowYourParty™ with this one, but if your party is level 5 and a Ancient Gold Dragon shows up and all the NPC’s start screaming to run, the hope is that MostParties™ are going to get the hint and play along with your story beat.
• A Hostage: Doesn’t need to be a person. In fact, I think this works better if its not a person. A valuable artifact that will shatter in two or three hits will completely change the dynamic of the fight. Those ten low level mooks are suddenly much harder to deal when you can’t just use Fireball.
• An Escort Mission: A variant of the hostage mission. The reason this works in a TTRPG when it doesn’t work in, say, a video game, is that video games have to use AI meaning the escort character inevitably will just walk straight into your attack and get killed, where a human operated character can still display basic competency (or lots of competency).
• Codependent Results: What I mean by this is that the success or failure of one of these other dynamic objectives results in a future encounter becoming easier, harder or different. For example, let’s say the NPC the party is escorting has a decent amount of health, but if they take too many hits, they won’t be able to use the SpecialMcGuffinOfAwesome™ against the demon in the castle, meaning the final battle becomes much tougher.
• An Attrition Inducer: In the context of a dungeon or long journey, the function of a battle can be to soften the party up for a tougher battle, or the context of an entire dungeon can be a challenge of attrition rather than a single tough encounter. Something basic like a golem with a very high armor class but low saving throws against magic can have a very different context by presenting the party with the decision on whether their hit points or spell slots are more valuable.
• A Pacing or Filler fight: Sometimes, having a fight just because the party hasn’t had a fight in almost two sessions is okay (as long as its justified in the story of course). Something like the goons of a local crime boss trying and epically failing to perform a hit on the party not only allows you to hit some emotional beats that are rarely seen in combat, it’s a powerful tool to enforce how much more powerful a medium or high leveled party is compared to the start of their adventure. It actually serves to make your world much more immersive when not every fight is conveniently around a medium to hardish challenge level.
• A Ticking Clock: The challenge of a fight doesn’t have to be survival. For example, if a DarkRitual™ is going to be completed in three rounds, the party can still be challenged while not being in a life or death situation (for them).

Hopefully, you get the idea. You may have noticed, or even objected, that a lot of my suggestions don’t necessarily fall strictly within the realm of the rules or mechanics that are enforced in the game, and that’s very much intentional. Hand in hand with this idea is the next lesson I learned.

Embrace the Gray Area In Between Combat and Roleplay

Most TTRPG systems have explanations of different types of gameplay that occur throughout a session or a campaign. On one extreme, Pathfinder splits its “modes” of gameplay into Exploration Mode (esssentially, free form roleplay), Encounter Mode (combat) and Downtime (what happens “off screen). On the other side, Legend of the 5 Rings splits its combat into three “scene” types depending on the size of the engagement (duels, skirmishes and mass combat), splits roleplay into unstructured (narrative scenes) and strucutred (intrigue scenes) varities, and then maintains the singular downtime scebe for the “off camera” connective tissue of the story. However, no matter how you cut it, the vast majority of tabletop roleplaying games fundamentally outline two seperate modes* of gameplay: Roleplay and Combat.

*In this context, “downtime” or “off-screen” storybeats quality as either GM narration or Roleplay-lite depending on the extent and depth to which downtime is engaged with.

If I haven’t already called my shot loudly enough here, let me outright declare that I think this is a mistake*. At its most extreme, it’s the TTRPG equivilant of when you kick some boss’s ass in a video game and then they just beat you in the cutscene anyway. In the vast majority** of video games that bother to have a story, the “gameplay” and the “story” exist in mostly seperate silos. Sometimes, particularly narrative focused video games will have you hold the forward button while you listen to dialogue. This serves as a transition from the cutscene you just watched into the radical shooting gallery you’re about to play through. Taken at face value, the roleplay and combat modes outlined in most TTRPG systems would enforce similarly strict silos of gameplay.

*I definitely get why tabletop games are structured this way. If the concept of tabletop gaming is totally foreign to someone, adding the complexity of “and there’s also this gray middle zone” is arguably a bit much. That said, i would still very much endorse some elaboration on transitional or in-between gameplay to encourage the kind of approach i’m elaborating on.
**This observation more or less covers any video game with cutscenes, and arguably any game that isn’t using mechanics metatextually like Undertale.

Luckily, I think TTRPG’s* naturally lend themselves to some blurring of the lines between these silos. Talking (in character) while engaged in combat is understood at most tables to be at least a semi-free** action. I’ve already alluded to or outright described several scenarios that exist in this sort of mechanical gray zone***. When I was talking about my theoretical story driven bandit encounter, for example, scenarios like spotting them in advance, attempting to sneak by them, or deciding whether to simply pay a toll to move past them peacefully all exist in sorts of various states of stuff that isn’t explicitly covered in the rules of Dungeons and Dragons. I’ve probably dove into the why to take this approach way too much already, so let me elaborate with examples of this sort of thing instead:

*This is somewhat dependent on what system you’re using. In general, i think the less structured story elements are in a system, the more it lends itself to this.
**I think virtually all GM’s are going to allow, say, a player character to shout a taunt or flip a bad guy the bird as a non-mechanical action in combat for story flavor. Where it gets into the gray area territory is stuff that might affect behaviors or strategy. Some GM’s will allow characters to talk ad-infinium about the philosophical nature of combat or whatever since there is no mechanical affect on the combat itself, but many (me included) have some undefined gray area where they’ll put a capper on things to keep the immersion of combat scenarios going.
***Best example I can think of this is if an NPC is being held hostage at knife point, most GM’s will simply say “and they get their throat slit and die” if the hostage negotiations turn sour, whereas technically, the DnD rules as written suggest you should roll to attack with advantage and then deal 1d4+modifiers damage, a pittance that is highly unlikely to harm anyone without a “commoner” stat block.

• Onboarding: This encompasses why the combat encounter is happening, any opportunities (or lack thereof) the party has to prepare for (or avoid) the combat encounter, or any sort of roleplaying preamble that may precede an encounter.
• Environmental or Object interaction: This covers stuff that you usually didn’t have the time or consideration capacity to explicitly prepare for. This covers everything as basic as “how much cover does that tree provide” to something like “there’s hard liquor behind that bar, right? What happens if I shoot a whiskey bottle with firebolt?”
• Roleplaying in Combat: This covers everything from the aformentioned semi-free talking action to player characters and NPC’s making mechanically sub-optical decisions that are rooted in character limitations or emotional motivations. Definitely keep this one in mind when running encounters. A pack of wolves doesn’t know to focus on the healer first, but a squad of elite mercenaries probably does once they see who the healer is.
• Logical Rulebending: This covers a wide variety of scenarios where the mechanics don’t account for what would likely be the more logically congruent outcome of an action. This would cover something like saying a defenseless hostage is killed by their captor instead of having the captor roll to attack, as well as not having the PC roll a death save if they hit 0 HP against a group of slavers (since the slavers are trying to capture them alive).
• Escape, Surrender and Negotiation: I find that the better outcome for a lot of combat counters to keep the pace of the game fun is for enemies (particularly sentient beings), to try to fenagle their way out of combat in scenarios where they clearly have no chance in winning. In the most extreme scenarios, this can save you a round and a half of boring, no stakes rolls where every miss is just a groan as the inevitable is prolonged. Instead, the scenario can evolve into a plot twist, an exciting chase, or an opportunity for characterization.

Keep in mind that I am not suggesting that every single combat encounter needs to have an hour spent to craft its specific, precise niche into the story with a variety of outcomes and strategic considerations. What I am suggesting is that there are a lot more tools in the encounter designer’s toolbox than what’s listed* in monter manuals, rulebooks and game master’s guides. Furthermore, while every principle I’ve listed so far is about orienting combat to be more appealing towards story driven players, these principles equally help more mechanically driven players engage with your story because they can engage in the story and express their character through mechanics and combat. It is much easier for a mechanically oriented player to connect to the emotional tension of a goliath that is physically indomintable if its stat block more aggressively reflects the strengths and weaknesses of such a creature.

*If you know of such a text that embraces this ideas, please let me know.

However, before moving on to what I learned about TTRPG storytelling, I did want to throw the more mechanically oriented players a bone explicitly in their direction. Without furtherado, here’s the last thing I learned about combat.

The Map is Far More Important Than the Monsters for Making Combat Strategically Engaging

There’s a fundamental problem in Dungeons and Dragons* enemy design where virtually everything they do can be boiled down to inflicting damage, suppressing enemies, and mitigation. In practice, even this understates the problem of how undynamic enemies can be because suppressive tools like a stun or mitigation tools like a legendary resistance are extremely unfun** to be on the receiving end of, so they should ideally be employed very sparingly. Yes, i’ve already gone over how an enemy with 15 Armor Class and 150 hit points and an enemy with 20 Armor Class and 100 hit points can feel different even if the cold hard math suggests they are basically the same. Additionally, I’ve also already alluded to the Beholder, a predesigned monster that breaks out of this mold with its signature anti-magic field. However, the bare bones simplicity of most DnD monsters is not just laziness – but an intentional*** design choice. There is very much a limit on the complexity**** a game master can track when it comes to more complex mechanics. By contrast, applying a lot of intentionality to the design of the location a fight will take place adds more effort in preparing an encounter, but adds very little complexity to the actual operation of the fight because the geography naturally lends itself to the unique strategic dimensions it entails.

*This particular bit of advice is going to inherently focus on the system i’ve run two campaigns in. In my very limited experience,this problem is particularly amplified by the simplicity DnD 5e in comparison to something more tactically oriented like Pathfinder, but still exists in that system.
**If you have five people at a table, a single round of combat can easily take fifteen minutes if you have some hiccups or an indecisive player, meaning if a player is stunned for one round, they can spend thirty minutes (the round they are stunned+the time in the second round waiting for their turn) doing literally nothing.
***In addition to the issue of operational complexity, the usage intention of a monster manual stat block versus one prepared for a specific campaign or encounter is entirely different. Monsters designed for the manual have to be cookie cutter enough that it can be broadly usable AND a GM can estimate its challenge level for their party. By contrast, Lava Dude – guy whose only attack is grabbing the lava in the pit below and chucking it at the party – can be designed to be employed for a specific map.
****The intentionality of this feels questionable when it comes to their spellcaster stat blocks, which list every single spell they have even though there’s like 4 tops they’ll ever use in an encounter and then the GM has to go through all the sweat-work of tracking all their spell slots. That said, the pain in the ass of employing even one spellcaster in an encounter sort of proves the need for simplicity for the GM’s sake. God have mercy on your soul if you want to have a campaign centered around a corrupt mage’s college or something.

Once again, demonstration is the best proof of concept, so let’s take a look at two level 7ish encounters that I employed in my first and second campaigns respectively. The first is extremely complex mechanically but not very good. The second is very simple mechanically, but the intentionality of its map design makes it far deeper strategically.

Encounter 1

Encounter 1 is a boss fight with an ancient construct which is the finale of a very large dungeon. There is effectively none of the onboarding I talked about in regards to that sort of gray area lead up I mentioned, but that’s fine*, because there’s a McGuffin in the dungeon and this section of the campagin was designed to be more combat oriented than story oriented. It takes place in a fairly featureless empty-ish room other than the healing pools, which i’ll get to in a moment. Mechanically, the boss is setup like this:

*Fine is the operative word here. For exactly this reason, my adventure design has slowly drifted away from explicitly pulp-adventure “here’s the combat” sections like dungeons for basically all the reasons this blog post exists.

• Medium Armor Class and very high Hit Points
• Movement speed of sixty feet.
• Decent all around saving throws give it no distinct tactical weakness, but it has no legendary resistances as one might expect.
• It’s immune to the Charmed, Poisoned, Exuastion and Frightened conditions as a logical result of it being a construct, and it’s also immune to the Petrified and Stunned conditions to compensate for a lack of legendary resistances.
• It’s immune to Poison and Radiant damage, resistant to Fire damge, and weak to Necrotic and Cold damage.
• A wide variety of attacks which i’ll go over now:
  • Multiattack: The boss attacks four times per turn, twice with spike , once with fire blast, once with set trap.
  • Spike: a 10 foot range melee attack that deals 2d6+4 piercing damage and inflicts the “bleeding” condition – a mechanic unique to this fight, which deals damage over time and has other consequences in this fight.
  • Fire Blast: a 60 foot ranged attack that covers an area that’s 20 foot in radius. It has a very high dex save and deals 1d10+4 fire damage on fail, or no damage on successful save.
  • Set Trap: Places a “trap” that covers four squares of the map. Stepping in it inflicts a dex save, and failing it results in taking 2d6+4 piercing damage and once again, inflict the the “bleeding” condition once again.
  • Bleeding: Any time a player character is bleeding, they make a medium difficulty constitution saving throw at the end of their turn, and take 2d6 damage if they fail. Additionally, any time they move while they have this condition, they leave behind a trail of blood which gets tracked on the map. Finally, in addition to making the constitution saving throw, this condition can also be removed via the healing pools on the map I mentioned and by the boss. That becomes relevant because this boss also has:
  • Two legendary actions, which it can to do the following:
    • Any of its standard round attacks lsited above – spoike, fire blast and set trap respectively.
    • Drain bleed: It costs actions to use, but deals 2d6+4 necrotic damage to all visible targets with the “bleeding” condition and the boss heals all the damage it deals. Once drain bleed is used, all targets damaged no longer have the “bleeding” condition.
    • Light blood trail: This also costs both legendary actions. All the blood trails i mentioned get lit on fire, and anyone standing in the blood makes a dexterity saving throw. This attack deals 3d10+4 fire damage on a failed saving throw, or half of that on a success.
  • I mentioned there are “healing pools” on the map, and you’ve probably observed this seems like a ton of damage for a level 7 boss fight. The idea is that in the four corners of the room which fill up as the fight goes on. At the end of each round, each pool fills up one third of the way if it’s not already full. 1/3rd of the healing pool heals the player for 1d8+3 healing if they end their turn in the pool, and it cures them of the bleeding condition. once the pool is used, it’s emptied and begins filling throughout the fight.
  • Finally, there’s also what I can a “desperation” action. Once the boss is at 25% or less health, it immediately it immediately takes a legendary action and fires chunks of its body at the players as a ranged weapon attack that deals 4d6+4 piercing damage on hit and inflicts the bleeding. From this point in the fight onward, it can no longer use its spike attack, and can only attack twice on its turn. Its movement speed is also halved.

Did your eyes glaze over while reading all that? I don’t blame you. The concept of this fight is that because the boss has a high movement speed, it will run around the map to get into melee range and try to get the bleeding condition on as many opponents as possible, creating a dynamic fight with lots of moving around. The reason that this fight actually sucks, other than the complexity, is that it doesn’t actually encourage movement.

Instead, the very high movement speed results in the boss running away from melee characters, forcing them to spend an entire turn not attacking to get back into range when it pursues the spellcasters and snipers. The second problem is that the set trap attack is an positively hair-brained idea of a mechanic that does nothing but add “complexity” to the fight, since players will obviously just walk around it. Both of these lead to the ultimate problem of the fight, which is that the optimal strategy quickly becomes very clearly to have the entire party group up in one corner of the room near a healing pool since so many of the mechanics punish players for moving around, and the only downside is everyone is in range of the very weak Area of Effect fire attack.

It’s not a fight completely without merit. The various strengths and weaknesses to damage types is evocative and forces our spellcasters in particular to get out of their comfort zone*. The final “desperation action” is extremely evocative of, well, a desperate last stand. However, all the various mechanical chicanery just made the fight go very slow due to all the mechanics to track that almost unilaterally boiled down to “it does damage, but differently,” and of course, it failed to even accomplish the strategic decision it was designed to induce.

*Although this is where my system specific choice of Dungeons and Dragons backfires on me to a degree. 5th Edition stat blocks effectively never have weaknesses because they prioritize modularity and ease of difficulty evaluation when designing their monster stats. Because even brand new players will catch on to this quickly, the learned behavior in this system is if you find a non-resisted damage type, spam it instead of risking hitting another resisted or immune damage type.

Encounter 2

As mentioned, the second encounter is far simpler mechanically, and instead relies on the map to create tactical depth. So logically, let’s take a look at that map.

Map made with Inkarnate.com

• The white letters are the player characters. The onboarding is pretty simple. This encounter is an ambush where the party has the opportunity to pass a high difficulty perception check to instead spot it in advance. If so, they’d start off of the map, with the wolves hiding in the south chunk of woods. They would then have the opportunity to reverse ambush them or attempt to bypass the encounter stealthily.
  • The story context is similarly simple. This is a group of mercenaries who have been ordered to capture the party dead or alive by a rival political faction. Z, their leader, stands out in the open once the trap is sprung. Instead of simply attacking with a surprise round, he attempts to persuade the party the party to surrender. This allows for them to learn about the aformentioned motivation, have some interesting roleplay, or even potentially bribe their way past the encoutner at substantive financial cost – perhaps a valued magic item if i’m feeling particularly spicy.
• The two red x’s are trained dire wolves. They do decent damage especially as a pair, and they have high hit points but low armor class. Their melee attack also inflicts a strength check which, if failed, inflicts the Prone condition.
• The two y’s are archers that have decent armor but medium-lowish hit points. They deal a little less damage than the wolves at range and significantly less damage when forced to engage at melee range, and they don’t have any boosting synergy like the two wolves do. You’ll notice they’re standing in the woods, which provide three quarters cover.
• Finally, there is the red z, who is a glass cannon mage who deals heavy damage, has low AC and medium-low HP, but also has magical resistance.

It took me far less text to describe three different stat blocks than it did to describe that one boss stat block, which reflects the far simpler operating complexity. In spite of this, the map offers far more strategic complexity with relatively low effort.* However, it provides far more strategic considerations via two very simple geographical features – trees and a river. Here’s the basic things to note about the layout:

*This map is pretty because its made using a web plug in, but you can easily draw trees and a river on graph paper in under 5 minutes and still present the same strategic consdierations.

• The first thing that immediately sticks out is traversal. A running river seperates the party from the heavy hitter of the encounter. There are two different paths to traverse the river conventionally.
  • Swimming across (yellow circle) is a medium difficulty athletics check, where failure means you fail to make any progress across, and also, you’re now in an exposed position to the two archers and mage.
  • The stones up top (red circle) can be leaped across. This is a medium-high difficulty acrobatics check, where the consequences for failure are greater since you’ll fall, take damage, and get swept downstream where, presumably, your lower athletics score will make it harder to escape. However, this path is high risk/high reward, since you have the advantage of getting to use cover on both sides of the river. Basically, the design here is to give both our burly barbarian types and our lithe monk types different traversal options.
    • I would also allow taking a stealth approach across the second path, the advantage being you get a leg up on the artillery on the other side, and the disadvantage being that you move slower across.
  • There is also an implied option of traversal, which is using a spell like Fly or Misty Step to get across. The advantage is its guaranteed to work and get a damage dealer to the other side. The disadvantage is that you use a valuable spell slot purely for traversal purposes. Secondary disadvantages like having a squishy spelclaster on the other side or needing to worry about concentration checks may also apply depending on the spell employed.
• The second component is, of course, the cover-providing trees. The first instinct in this sort of situation is probably to take cover behind the trees so the artillery damagers across the river are far less of a threat while the wolves are dealt with, but that’s where the “trap” element of this encoutner truly comes in. You see, the wizard (z) will cast fireball, which of course will deal 6d8 fire damage to the presumably grouped up party in the trees, but will also catch the trees on fire. Anyone who ends their turn in the burning areas for the forest will take 3d8 fire damage, and the cover provided will be reduced to half cover after burning for one round, and provide no cover after that.
  • That’s obviously unfair – intentionally so. However, it does provide the party with an additional option to deal with the ranged fighters on the other side, which is to employ similar fire based tactics to get them out in the open.

It’s not just that there are tons of tactical considerations. There are tons of valid solutions. The party could focus on taking down the wolves as quickly as possible as a group, or they could have the goliath barbarian chuck the halfling rogue like a football across the river. It’s a mundane enough obstacle that plenty of wild, off the book solutions exist, and the basics of the encounter are pretty scalable anywhere between level 5ish* to level 15ish**.

*Below level 5 the margin for error just becomes too thin. Two bad rolls on a swimming skill check, for example, runs a very strong risk of a party wipe.
**Past level 15, a running river is just too mundane a challenge to place in front of your chad barbarian types, who will almost certainly succeed any reasonable athletics check given the context and then obliterate the glass cannon. Before that, though, y ou can definitely spice this up with giving the wizard counterspell and/or giving them a second wizard.

If this sounds too hard, keep in mind that because of the strategic complexity, i don’t need nearly as much raw damage dealing or sponging. In fact, this fight functions great as sort of a pacing or filler type fight because even if its undertuned from a numbers perspective, it doesn’t need to be hard because it’s intellectually stimulating.

All of this might seem like a lot of words written about combat for someone who has professed to run a game that’s more story oriented. That’s fair, but I also think since I chose to partake in the somewhat round-peg-square-hole act of making a story centric experience in Dungeons and Dragons, it presented unique challenges and a large bounty of lessons learned. However, enough is enough, so without further ado:

What I Learned About Roleplay

We’ve now reached the part of the excessively long article where we dive into the HotTakes™ portion. Yes, i’ve hopefully presented ideas that are pretty novel, but I think their lack of uniquity boils down to the official published products not containing them. Most* objections I would anticipate to these ideas are that they involve way too much prep work** and that they’re ill-suited to the objector’s particular style*** of game.

*I’m sure there are also objections to specific critiques or generalities i’ve lobbed as well, but on the HotTake scale, i think these are pretty mild.
**This is in no way meant to be a slight towards people whose preperation process prioritizes efficiency and having free time to do other stuff. I realize I am a single, no child having, no second job working dude who is an over-preparing psycho even by these standards.
***I would argue that most of what I laid out is actually especially effective in more traditional, hack and slash style tables because you can do a lot of story work without distracting from the main appeal of your table, but I also acknolwedge that some tables are just fully uninterested in story and that’s fine.

The reason for that is, up until now, there are very tangible, mechanical things I can point out with a lot of what i’ve said, and often times, even if someone objects on a sort of qualitative preference, they can’t really argue i’m objectively wrong because objectively* it’s very difficult to argue I am not accomplishing my goal even if you disagree with the goal I am pursuing which most** people would at least acknowledge is now a discusison of subjective preference. That’s all about to change, not just because there’s a lack of tangible basics that are beyond reproach, but because there’s nothing people on the internet*** love to do than argue their takes about fictional storytelling are objectively**** correct.

*Given the pure length of my thoughts on combat, there’s almost certainly some objective flaw just purely as a failure of fact checking of system knowledge. The idea here is that reasonable people aren’t going to look at advice like “have your NPC’s behave like they would logically based on the fight’s story context” and say that’s objectively dog-shit advice.
**Given how appalled I am at the average redditor’s DnD takes, please note there is an implied “people I’d be able to enjoy a single converation with” disclaimer on “most”
***Real life conversations, particularly about art, have gotten very internetified to the point I might as well not even include the qualifier “on the internet.”
****Without getting into the worth its own article subject matter of whether objectivity even exists in art, let’s just say, yeah, this often includes yours truly.

I’ve never been one to shy away from the HotTakes™, and this already feels like excessive foreplay, so without further ado, let’s dive right into my HottestTake™

The DMPC is One of the Most Valuable Tools We Have

The Golden Rule of GMing a TTRPG is “Don’t railroad your players.” This piece of advice is two steps above worthless as it lacks any specificity as to what “railroading” is, explicitly ignores any nuance*, and often directly contradicts lower rung but popular advice like “make your world believable.**” The Silver Rule of GMing, by contrast, is more useful on face value – “Never use a DMPC.” The reason this rule is clear at least in its construction is that its a binary statement about a specific behavior rather than a general game philosophy. Really, the only flaw is some variations in defintion of “DMPC” which, yeah, that will definitely become relevant. However, As if the title of the article section didn’t already give the game away, I have opposition to this common wisdom as well on more basic, tangible grounds.

*Railroad vs Sandbox- also known as Linear vs Open-ended – is not a conflict between two diametrically opposed concepts as usually presented. Instead, it is a spectrum where a noticable portion of the community will argue all campaigns should be purely Sandbox/Open-ended. I think a lot of this is self-deception. If you are doing any sort of substantial pre-session prep or running a published adventure, than there is inherently an element of linearity to what you are doing. That’s why “don’t railraod your players” isn’t very useful advise unless the advisor is suggesting all session prep is inherently bad.
**The reason world believability inherently contradicts “don’t railroad” is that every world has at least a few basic rules and many more implied rules. For example, it should be near impossible in most settings for a level 5 party to assassinate the ruler of your setting’s kingodm, and if they accomplish this feat, the consequences almost certainly need to be severe and life altering because of the wide ranging political and societal implications. This is inherently a limit on player freedom and thus, at least to some minor degree, railroading.

All of that preamble is essentially an acknolwedgement that before I can even explain why the DMPC is a valuable tool, I have to address the mountain of common, valid criticism and GM horror stories before I can do so. In short, here is the bulk of said mountain and my response to it:

• All (or most) players hate DMPC’s
  • My players liked it.
    • Originally there was a whole other section of preamble attacking strawmen who refused to believe a DMPC worked in my game in any sort of way. This was ultimately scrapped because you’re either willing to believe that basic statement or you’re not.
    • There are a few valid criticisms of the above, however, that still deserve a bit of attention.
      • “Your game was good in spite of your DMPC.” It’s entirely fair if you have mountains to evidence to suggest otherwise. Unfortunately for me, there is not a single retort to this but rather the long explanation to come on why the DMPC made my game better, after which you can either agree or disagree with my assessment.
      • “You’re just an uber special/talented GM. Just because you made it work doesn’t make it good.” Not usually the sort of complimentary statement you hear from dissenters, but I do acknowledge the possibility of someone being wildly impressed with the improving combat section up to that point. To this, I’ll say I think my advice is very much usable by most GM’s, and those lacking the particular type of restraint needed to utilize this tool can still utilize these tricks (less effectively) via companion or NPC type characters.
  • I think a lot of modern, more story oriented campaigns have some NPC or guest-character that serves a lot of the functions I’m going to get into. The distinction between my HotTake™ and these is that they tend to be temporary party members that pass what I would call a DMPC-flavored baton. I think this is perfectly fine, and I’ll probably utilize this style of NPC to serve this purpose in the future. I just think the true DMPC is a little bit better at certain specific things i’ll dive into.
• The DMPC is extreme power gaming.
  • There’s a basic assumption built into the most common DMPC criticism that craving statistical/mechanical power (i.e: magic items, character abilities, etc.) or diagetic/in-universe power (money, titles of import, followers) is inherent to the experience if not the primary reason the experience exists. To a lot of this criticism, I would narrow my eyes with suspicion and ask why we don’t feel like TTRPG’s can have higher ambitions than this, but at the same time, I would acknowledge anything that involves a player progression that makes you better at things inherently does mean this is at the very least some component of your game.
  • My solution to the above really just boils down to “have restraint,” since any sort of power gaming/power fantasy component is inherently at odds with the fact that you, the GM, have literally unlimited power at all times. However, I do have a couple of distinctions between a DMPC and a traditional player character within that general principle to ensure there’s never a feeling of this sort of thing. (Because even if you’re correct about the fact you’re not power gaming via the DMPC, that’s completely irrelevant to your players’ enjoyment if they feel like you are.)
    • A DMPC and a traditional player character should always have their distinctions in function. A player character is an avatar through which an individual player experiences your world and interacts with it. A DMPC, like a normal NPC, is a semi-flexible entity of personality traits and backstory which serve a specific story function. If you’re attempting to experience your own world through a character avatar, then you’ve gone full Kanye. Never go full Kanye.
    • The DMPC does not partake in the post battle looting. With gold, this can either be handwaved through narration (“Everyone in the party gets 70 gold!” leaves it entirely ambiguous how many ways its being split) or by the DMPC explicitly being uninterested in taking blood money for reason x. As far as magic items, the DMPC doesn’t get any during the course of gameplay unless one of more party members explicitly state “here [DMPC], you mean a lot to me, so please have this item.” If the DMPC can naturally stay strong enough not to be a burden on the party through their own generosity, that’s great. Otherwise, the DMPC will on rare occassion get a single magic item via off screen shopping that’s just strong enough to keep them a relevant participant in combat.
  • An obvious benefit to a DMPC (so obvious, in fact, i’m just covering it here) is that they can cover up holes in party composition. This is obviously much more relevant to smaller tables than large ones, but players being able to just pick what class, sub class and spells they take with minimal to no concern about party composition is an immediate benefit. No one’s playing a class with healing spells? The DMPC can be a cleric.
    • The idea here is if you’re being smart and building your DMPC with this in mind, they won’t be “power gaming” or “stealing player glory,” but instead be accomplishing seperate combat functions the party is tangibly unable to. In other words, no one can be mad the DMPC is stealing the glorious task of healing the party for 2d8 when no one else took up the job opening.
    • This is doubly a bonus because, inherent in the above explanation, the DMPC is specifically taking up jobs and tasks that the players didn’t want to do.
• What you’re doing (with distinctions listed above) is literally not a DMPC.
  • In my personal opinion, DMPC comes closest to the type of NPC I am describing. I would personally distinguish them from a Companion character because i am still (usually) operating them in combat and they are not tied (usually) to a specific player character’s plot or backstory. I would distinguish them from general NPC or “ally” term because of their general omnipresence. The idea of the DMPC is they are almost always present for major group story beats.
  • Ultimately, this comment is semantics about made up terms. By all means, apply a different label to my guidance if that makes more sense for you.
• The DMPC takes the spotlight away from the players.
  • I’ve more or less covered this already with the ideas of DMPC vs traditional PC story function and the DMPC’s practical role in combat. The ideas in the “power fantasy” section basically apply here.
  • The one distinction between this critique and the “power fantasy” critique I would want to address specifically is in story function – specifically the idea of player characters receiving adequate story attention and the risk a DMPC can inherently pose to this.
    • In short, I think this critique is valid, and the DMPC that by defintion inherits the title “most improtant NPC in the campaign” is obviously the biggest potential offender. My response to this is simply that this specific concern applies to any NPC of any stature, power or importance. Once Ser Dragonslayer the Radical has interacted with your level 5 party, their status in this world has immediately diminshed (slightly) purely by the existance of a CertifiedBadass™.
    • Certain “the players are everything” type extremists would argue that Ser Dragonslayer the Radical should never, ever exist in a campaign for that exact reason. I disagree, but my point here is that this “spotlight” concern is much larger than the DMPC, and if you’re following my other basic principles of function, it shouldn’t be a problem unless it’s one more rooted in the basic fundamentals of your campaign.

I’ve already alluded to in my very long defense of the concept some of the basic advantages of the DMPC. Really, the philosophy of this all can be boiled down to one idea, which itself is quite controversial:

The vast majority of players are bad at initiating story without guidance.

This simple statement cuts at the core of a particular gaming sentiment i’ve criticized in small doses or implication up until this point, but it’s now necessary to directly address.

There is a very popular idealized concept of a TRPG campaign where players are set in a location (usually a village) and then explore and interact to their hearts content without direction from the GM. The idea is that the location is full of NPC’s that will create conflict or provide story hooks which become quests if party latches on to them. Planned story beats or carefully prepared encounters are the enemy of good campaign design, this philosopy proposes, because DnD (or other similar systems) are not storytelling games period. They are roleplaying games where story is supposed to evolve organically from the players perfect little foreheads like the beautiful rushing water of Niagra Falls. And if the events of your campaign instead resemble wierd, wacky shennanigans with no emotional engagement or any resemblance to a satisfying story arc, that doesn’t matter because the god given purpose of TTRPG’s is to give players freedom of expression and anything that could potentially limit that is bad by default.

To be clear, I don’t think there’s this massive population of non-storied or non-planned campaigns going on, but I do think it’s a very popular idea that striving towards this as an idea is the correct way to play. I disagree strongly with this for two reasons.

• This is GM cowardice. Placing any potential for story or engagement purely on the players just means if a session, story arc or campaign isn’t satisfying (or less satisfying than its potential), then its the players fault and not the GM’s.
• I disagree with the notion that a TTRPG can’t be a story driven experience. The appeal to roleplaying is not just that you are inhabiting a different fictional person from yourself, but that it’s the kind of person that gets into exciting, interesting things that you, the player, don’t get into in real life. If this was not the appeal, then everyone would just play Sims instead. That’s not to say there isn’t some appeal to simply going goblin mode on a group of highway men or turning the rude duke into a chicken, but to suggest this is the highest possible aspiration of the hobby when something like the ending of Critical Role Season 1 (SPOILERS, obviously) exists is just a depressing concept to me.
  • I’m not suggesting following my advice or having planned story elements will ever result in something this spectacular. Accomplishing something like this takes extraordinary amounts of both luck and talent. I’m saying if we’re going to have this perfect ideal to strive for in our hobby, I much prefer this be that ideal.

And yes, to accomplish this ideal, it does require some player freedom to be sacrificed. The trick is to sacrifice so little that the players don’t even notice its missing.

The single most insightful thing I’ve ever heard about satisfying story within the context of player freedom came from Brennan Lee Mulligan in this roundtable*. The key dichotomy to story driven DnD is that players want to live out the satisfying story experience of Sam and Frodo carrying The One Ring up Mount Doom, but they also want to roleplay as a character who desperately wants that to happen as painless and efficiently as possible. That means what their character wants to do is have the eagles fly the One Ring to the top of Mount Doom. In this dichotomy, player choice is like a rushing body of water, and the GM carves the path of least resistance to form the oddities and curves of a river bed which metaphorically becomes our satisfying narrative. Thus, player freedom is sacrificed minimally to acheive story. The players don’t feel the hand of the GM or the rails of planned story elements because the world and the sessions the players explore it in are designed in such a way that makes satisfying story the logical outcome of both character motivation and the fictional reality.

*I was kind enough to timestamp the youtube link to this moment 🙂

That’s what the DMPC is meant to acheive. Carving a river (metaphorically) requires many tools, and the DMPC is a tool capable of many things while also (when operated skillfully) remaining invisible. I have two main tricks when it comes to utilizing my DMPC in this way, which i’ll call the Nudge and the Clean up.

The Nudge is my bread and butter – a baseball pitcher’s fastball, if you will. The idea is that while great* story design means all player decisions** result in a narratively interesting scenario, there will be times players come up with stuff that you didn’t consider. Shit happens***. Nobody’s perfect, and player freedom means sometimes you have to improvise. This is the business we’ve chosen.

*I know, “just simply account for every player decision, lol.” I promise, it’s not as impossible as it sounds. It mostly boils down to accounting for player choice in very broad terms. It’s not about the how, it’s about the what. “Bad guy is caught” versus “bad guy gets away” are extremely broad ideas with very little gray area. Details like the bad guy dying versus being captured alive can be sorted out between sessions.
**All Reasonable player decisions anyway. By this, I mean it’s reasonable to expect players to take the informed realities of your world seriously, I.E: not attacking randomly rude NPC’s on sight in broad day light because no reasonable human being does.
***Of course, sometimes player decisions are “unreasonable” becuase of genuine misunderstanding of the scenario being narrated or presented, and that’s one of the numerous ways we risk running into situations that can be problematic to our story

The idea at its core is that when the party or a particular party member is considering a course of action or about to make a decision that will result in a negative experience* for said players by no fault of their own, we can utilize the DMPC to subtly nudge the player away from that direction. The key to keeping it subtle is that the DMPC must always do so in a way that is rooted in the character’s motivations and personality. This only works when the character remains believable, and that is how the nudge remains invisible. If you cannot accomplish the nudge while maintaining the DMPC’s character consistency, then you shouldn’t do it. Period.

*As in deeply unfun or narratively unsatisfying. Not necessarily in the sense that you are attempting to prevent negative consequences.

Most of the time, I do this one of two ways: presenting information and distraction.

Presenting information is the DMPC sharing information about the world or about other characters that they didn’t know or didn’t consider as part of their decision making process. This information is anything that may dissuade the party from taking this course of action or anything that may make the party want to take a different action instead. Most of the time, the existance of this knowledge has some degree of improvisation. The idea is that given the DMPC’s backstory and personality, it would make sense that they would both have this knowledge and be willing to share this knowledge. It is surprisingly easy to convey this information indirectly, which assures the gentle, guiding hand of the GM remains invisible.

Let’s use an example. For my imaginary scenario, let’s say the party is debating whether to go to Capitalville to assassinate Duke Evilton, and you don’t want them to because you have BigPlans for Duke Evilton that you’ve already foreshadowed. The information I’ve decided to convey for demonstration purposes is to remind the Orc Paladin that his estranged uncle from his backstory lives in Capitalville, and therefor as the GM I know there will be hesitancy from the Orc Paladin to go to capitalville. I could have the DMPC just remind everyone Estranged Uncle lives there and that might create an inconvenient or awkward situation, but this really would only work of the Orc Paladin and the DMPC are very close, and even then, it’s not a very subtle* nudge. However, if I have the DMPC metion they know the streets of Capitalville very well because they had to delivery a large weapon shipment to this guy called Estranged Uncle well time, it will not only present the same information to the Orc Paladin but indirectly, but because the character is trying to help the players acheive this inconvenient** plan, it’s the perfect disguise for your guiding hand.

*That doesn’t mean it’s a bad idea. Go for it if it enriches the DMPC’s character and their relationship to the party. The idea here is that all things being equal, go for the subtler approach.
**Inconvenient for the whole BigPlans angle.

As mentioned, there’s not always a way to nudge the players while more importantly maintaining the integirty of your DMPC’s characterization, and that’s where the distraction comes in. While this also must remain rooted in character, because it lacks specificity, there’s pretty much always a way to do so. It honestly takes very little to reroute the conversation somewhere else. This works on three principles:

• Players at a TTRPG table are pretty much exactly like cats. They like to be where they shouldn’t, and they have very short attention spans. The key calculation to all of this talk of redirection and trickery is to think what would you do if you want to make your housecat do something else instead. Applying this basic logic to both the nudge and the distraction, this alone has an incredibly high success rate.
• Players are very forgetful due to the above, mention something off topic and they’ll probably decide to do something else if you can keep their attention for one minute.
• Otherwise, simply add a detail to another plan that has been presented as an option. The thing about plans at a DnD table is the one with more details pretty much always seems like the better option because it has contingencies while plans with less details don’t.

All of the above does lend itself to a question: what constitutes a decision you should try to redirect the players from? A InconvenientDecision*™ is character choice that is made arbitrarily that will derail planned story elements in some way. In this context, I define arbitrary as any decision made without** intentionality rooted in the the player’s character, and I define derail as a situation where a character action would force you to choose between changing the reality of your world*** or allowing a narrative outcome or depriving the players of potential narrative enjoyment.

*As hopefully clear by now, I do not use this label as a judgement. Players do not know your plans and make their decisions made on incomplete information. InconvenientDecisions™ are inevitable.
**I don’t mean to suggest player decisions are “bad” or “lesser roleplaying” if there is not a specific, tangible character reason it’s being made. You don’t decide what to eat for lunch based on childhood trauma, and similarly, roleplaying means a lot of times a decision is just based on the first idea you had. The idea here is to distinguish decisions made arbitrarily from ones made because it is a deliberate choice rooted in character personality or backstory.
***There’s definitely some gray area here. In the Duke Evilton example, the foreshadowing not paying off would have resulted in losing potential narrative enjoyment while maintaining the payoff would likely involve changing what the foreshadowing was pointing to, thus changing the reality of your world even if the players don’t explicitly know that.

The key to making this work is that less is more. I keep alluding to this idea, but it very rarely takes much resistance to trick players’s cat brains. By contrast, the cat brain has a deeply rooted desire to misbehave, so if players detect that you don’t want them to do something, you’ve just ensured that’s exactly what they’re going to do. The key to maintaining enough confidence in the less is more approach is to avoid letting perfect be the enemy of good. Players will make InconvenientDecisions™. Part of the magic of this slight misdirect is that it turns arbitrary decisions into deliberate ones. If you remind the Orc Paladin of their Estranged Uncle, and their character tells you they’ve been running from their past from too long, do not under any condition stop them, because now they’re doing the very thing that is the foundational goal to being a GM. They’re investing themselves into their character and making decisions rooted in motivations other than their own. Because it’s no longer arbitrary, it no longer qualifies as an InconvenientDecision™.

And though the above is true, the decision may very well still be inconvenient in the practical sense. Luckily, we still have the clean up technique at our DMPC’s disposal.

Clean up is when you make a mistake and you need something in the story to happen to bring things back on track. A lot of these are extremely simple. If you didn’t draw a door very well on your battlemap, you can utilize your DMPC to make a comedic beat about characters testing out the window sil instead of simply as the GM telling them there’s a door. This kind of simple action when still rooted in character has a bunch of benefits. It keeps the pace of the session’s momentum going because no one breaks character. It’s a small opportunity for roleplay and characterization. Most usefully, it doesn’t accidentally point players in a certain direction when that’s not your intent. Basically, it still gives players the freedom to explain in character they’re sneaking in because they got in trouble at this inn last week if you, the GM, misunderstood their intent.

A rare but incredibly vital DMPC clean up is when it comes to combat and combat related planning. I’ve already mentioned the concept of the DMPC’s abilities filling team needs so that players can play what they want, but this doesn’t stop there. An inconvenient dichotomy* for story driven campaigns is that player character death is the ultimate story derailer**, but player character death has to remain a credible threat to maintain the integrity*** of player characters’s decision making. Basically, while the GM never wants a player character to die, there are lots of situations**** they especially don’t want a player character to die*****. Luckily, it’s usually not very hard to prevent combat scenarios from going HorriblyWrong™.

*Admittedly, I must confess this is something the “plans limit player freedom/players should drive all possible story elements” crowd have going in their favor. They have a far easier time handling player death than I do.
**If you get two thirds of the way through an epic, emotional campaign and one of the beloved PC’s dies unexpectedly. The best case scenario is that event will obliterate the party’s enjoyment for the next few sessions as opposed to the entire campaign.
***I have the utmost confidence in my very lovely players to approach their roleplaying with the best of intentions. I just think it’s basically impossible for anyone to prevent player knowledge of literal invincibility not to derail their character integrity to a certain extent because being the kind of adventuring badass who ignores warnings of danger is pretty much baked into the cake of most TTRPG systems and campaigns.
****I don’t want any of my players to die because that introduces the incredible inconvenience of needing to figure out a way to introduce their next character and make them feel included in the adventure. That said, I’m perfectly happy to let a PC go out in a blaze of glory in an epic clash against an undead army or a confrontation against the evil king if the dice fall that way. By contrast, a player character dying in some random filler highwaymen encounter is just the absolute worse because it introduces all of the aformentioned inconvenince while also obliterating any narrative satisfaction whereas by contrast, the blaze of glory route has that going for it.
*****I have a huge problem with DnD and DnD influenced systems making death incredibly trivial for lots of parties because of the aformentioned problems a lack narrative threat of death presents to character decision integrity. It’s for this reason I think the majority of DM’s put up pretty substantial homebrew roadblocks to character resurrection if not disallowing it almost entirely. That said, the rules as written approach does have the aformentioned advantage of not having to worry about this whole mess, which is why the rules are written this way.

The simple version of this is the same in character reminder as the poorly drawn door. Remind the character they just took 4d8 fire damage from the lava golem’s skin last turn. I think it’s very reasonable for the DM to just simply remind players out of character of something like that, since their characters would remember, and they only forgot because the in-universe time frame of 6 seconds took ten minutes in the real world*, but an in character reminder has the aformentioned identical advantages of the door situation.

*And at most tables, the player did not suffer second degree burns in real life.

The subtler version of this is when things are already teetering on the edge of going HorriblyWrong™. This is also surprisingly easy. Simply have the DMPC fight extremely optimally because you, the GM, know the strategically perfect approach to your own fight. This is pretty much always enough to swing the situation* in the players favor again.

*This is another reason why strategically deep and story driven combat is so important. If the players are simply just overwhelmed by the statistical power of some absolute unit’s beefy HP pool and damage output, then this doesn’t work.

The final type of clean up is hopefully* the rarest, but it’s undoubtedly the most important. I presented the Duke Evilton scenario where a theoretical party of players were making a decision which, unbeknownst to them, would derail future story beats. I also presented a scenario in which as the GM, we are compelled to allow this derailment to occur because, as I opined, empowering your players to maintain character integrity must similarly be protected at all costs. Even if that cost is story derailment.

*If it’s not, there are fundamental problems to your session or campaign design that need to be addressed.

In this scenario, the only thing that matters is getting to the next session. Hopefully, this doesn’t involve the DMPC at all. If you’re three hours into a four hour session, just say that’s all the content you had prepped this week – that’s all folks. Totally okay. Encourged even.

If you’re on hour one of four – not so much. In that case, hold on tight. Every trick conceivable. “Oh hey, before we head out, I need to go to the store” and cue 20 minute shopping sequence. “Hey, if we stop by the village just outside Capitalville, i know someone who can help” and cue overland travel mechanics, a filler encounter, and then a fully improvised social interaction and hey what do you know, the Village Helper has some vital intel, but you need to find his lost pet pig first.

It’s fucking scary, but if you can just get to the next session, you’ll find a solution. What you want to avoid is attempting to fix the train while it’s running. There’s usually a way to maintain the story plan if you just have some time to think. You can usually rework some things, and if you can’t, there’s at least time to mourn the story beat’s loss and get excited about something new.

A natural question after all of this is why does this have the DMPC have to do all this? It’s a fair question, and the answer is that they don’t. By all means, I encourage you to employ these tactics through a variety of NPC’s and companion characters. However, I do believe the DMPC is uniquely qualified to employ these tactics.

The reason for this is the DMPC’s invisibility. This is far from obvious. The DMPC is, after all, omnipresent. But it’s that omnipresence that allows them to fade into the background. A regular NPC is usually being interacted with for a very specific reason, meaning that even if they have a grand total of ten minutes for the duration of the campaign, they are the center of attention for those ten minutes. That means those ten minutes are scrutinized. Meanwhile, the DMPC will only be the center of attention for those first few sessions when the party is getting used to the idea of considering this non-player entity a part of their team, but the new situation will quickly* become normal. Humans managed to normalize the worst pandemic in over one hundred years, an imaginary friend is child’s play in comparison. The players know, of course, that the NPC is not one of them on a basic intellectual level, but their mundane omnipresence means they become a basic reality of the campaign’s world – not all together different from the reality that rolling a natural one will result in failure. Once they’ve blended into the very fabric of the players’ reality, well, now you’re getting the idea.

*And also, early sessions rarely need much of anything in the way of directing player energy productively. This is because you have the fewest conceptions for the long distance destination of your campaign at this point, and a low level party lacks the ability to greatly affect the world in a strictly mechanical sense.

And yeah, this doesn’t always work. Sometimes you only have a choice between bad and worse. It’s tabletop gaming. After all, this is the business we’ve chosen.

Plan the Short Term as Much as Possible. Plan the Long Term as Little as Possible.

The thing about creating the sort of convoluted twists and turns that create a satisfying story arc is there’s not really a way to do so even when the themes and points of emphasis are derived primarily from player improvisation. The key boils down to the prior mentioned river analogy – player freedom must be preserved as much as possible while accomplishing this. With all the advice and philosophizing* about utilizing the NPC to steer the direction of a sesssion, I think the prescience of this statement becomes pretty obvious. Rebuilding a train track might be theoretically easy when you have the power of God, but it can quickly devolve to disaster when you have no blueprint and attempt to do so while the train is running. What’s less obvious is what exact this statement means and how to accomplish it. It runs into the same vagueness problem that the “don’t railroad” GM golden rule does. Let’s fix that.

*Really, the DMPC discussion was a vehicle to discuss gentle player direction to create satisfying story. The DMPC is jsut my favorite tool to accomplish this.

“Short term” in this context means the next story objective that will take the players between one and four sessions to accomplish. Think of what the focus of your preperation should be as the point where a television episode* would end. There’s a lot of variety in what this constitutes. This can be player driven: such as a executing a heist or busting themselves out of prison as a result of said heist going poorly. This can be GM driven: such as an attack on the player’s hometown by a horde of undead or an attempt to assassinate one of the player characters. In an ideal world, the end of one of these story objectives will align with the end of the session, but if not, you’re usually fine improving. This is typically where stuff like shopping shennanigans goes, and if your party isn’t into that, you’re usually completely fine to get the early next objective onboarding done without a plan.

“As possible” in regards to short term planning basically means “while maintaining player agency and your own sanity.” I keep saying this, but perfect, once again, cannot become the enemy of good. There will always be flaws within a long narrative construction that’s driven heavily by improv and pure chance, so you might as well try your best not to let the vague threat of possibility consume all of your free time*. As I mentioned before, the key to having a decent plan for most scenarios is to boil things down to very vague and broad outcomes. Stick to “They live/die/succeed/fail” type of things. Unexpected details within these broad construction will often force you to improvise, but improvising is far easier when you have a solid foundation and a basic direction already planned out.

*DISCLAIMER: This and all other future statements along the lines of “don’t overprep” or “don’t overwork yourself” are extremely “do as I say not as as I do statements.”

“Long term” then is any ideas are further out than that 1-4 session chunk that falls under “short term.” The less stuff you have prepared outside of that window, the easier it is to prep and all together avoid needing to worry about InconvenientDecisions™. The less you have prepared in the long term, the more control the players have over the direction of the story. Easy enough.

The much trickier part is the “as possible” when it comes to the long term planning. This is the single hardest thing for me to advise upon because this is very specific to an individual campaign on both the details of an overarching story and the moment to moment story problems and obstacles presented to players. However, I do think it’s fair to suggest that basically every campaign has at least a little long term planning, and I also think I can split this into two categories that are broadly applicable for most GM’s.

The first type of long term planning pertains to character ambitions and objectives which are too big to fit into that 1-4 session “short term planning” bubble. For example, “assassinate the evil king” is an objective which could be compressed into that window, but it is probably both more narratively satisfying and honorable to the integrity of your world if that is a tall mountain to climb* that consists of a few different smaller objectives which build toward impacting the actual assassination plot itself. In this case, “as possible” becomes limited because by definition, those small objectives need to impact that very large objective for them to matter. Therefor, you need to know things like “who** is the evil king?” and “who are the main power players in Capitalville?” so the world’s reaction to those smaller objectives feels, well, reactive. You definitely don’t need a whole lot of details, but it’s very helpful to know ahead of time some basic reactions to things like “what if the king finds out about the assassination plot?” or “how do these small objectives impact the assassination plot’s success?”

*Metaphorically speaking, unless it is very literal because the evil king lives in a mountain fortress.
**”Who” as in who they are as a person. Not just the fact their name and title is “King Evilton.”

The other type of long term planning pertains to story beats, ideas for the future, the sort of “this would be cool” that starts itching at the GM brain when you spend multiple hours every week on this very specific thing. In spite of the many black colored pixels expended specifically spent to argue against a sort of planless sandbox style game, this is where I must admit the implied criticism against a more planned, story driven game has merit. The harshest of my theoretical critics at this point are surely screaming “obviously you just want to write a book, so do that instead of making your players suffer!!!”

Well, for starters, I think there are a few examples of this type of planning all but the most extreme sandboxers can get on board with. A very common thing that happens with TTRPG’s is character backstories are presented which leave a implied expectation of something which should be addressed in the campaign – or at the very least, something that’s very rich with potential storylines and adventure hooks. Potential pitfalls* aside, I think “incorporate players backstory into the campaign” is a pretty popular sentiment. Now strictly speaking, player character backstory elements do not need to be planned, but I don’t think it’s hard to understand why for this particular thing component is much better as a planned long term story element. Character specific backstory which is also heavily driven towards by the individual characters runs a serious risk of of one or two characters feeling like they’ve absorbed all of the narrative attention – particularly when this playstyle naturally already means there is less narrative in the metaphorical pie to be split between the players. With long term planning however, these elements can be incorporated naturally into the overall broad story, which on top of avoiding the risk of said sentiment, has the powerful advantage of naturally creating bonds between player characters because they become invested in each other’s personal lives.

*This can become dicey on the extremes of player backstory where there is either very minimal detail or pages upon pages of detail. The former means that either you the GM must fill this in yourself or simply provide less narrative attention to that character. The latter can be heavily restricting due to the lack of gray area, which means there must either be a heavy degree of coordination and collaberation with our very detailed backstory-haver or, once again, simply providing a lack of narrative attention to that character. Really, what all the potential pitfalls boil down to is spotting them in session 0 when character concepts are unveiled and discussing with your players on their degree of interest in these sort of elements and how much narrative they may want or not want during the campaign.

The other element of long term planning which I think is fairly uncontroversial as a net positive is world building elements which imply story elements or have the potential to cause them. Something like the continent of your world being ruled by an demonic entity implies very different middle to late game story beats versus the conflict that might arise from a queen ruling an otherwise patriarchal society. The standard classical fare of traveling town to town and dealing with their problems for coin or kindness has its limits. Not only does classical storytelling convention suggest the stakes should raise over time, but the mechanics of most RPG’s similarly call for more and larger threats. Eventually, most RPG fictional universes do dictate given the idea of character progression that the party should directly interact with or even become the politically and culturally important figures of your world. Even if your world building approach is one of fill in the empty canvas as your players travel across it, I think it’s, once again, pretty obvious why some planning far ahead as these elements become visible is extremely helpful. Detail and deliberate thought in worldbuilding provide grounding logic to assist in the improvised session to session GM decisions as the players capacity to influence the world grow both in their worldly stature and their pure mechanical capabilities.

There is still the elephant in the room which is “just write a book.” Yes, there are those elements which I think a lot of if not most campagins have like “this character will betray you 10 levels from now” or “the helpful inn keeper is actually the bastard son of of the king” which cannot be adequately defended against this sandbox oriented criticism. It is purely the selfishness of myself and other GM’s that create these beats, and my defense is simply that that’s okay.

That is not to say that these are purely valueless elements. A long, drawn out twist in the story can delight your players in the same way “I see dead people” delighted millions of film goers in 1999. Players can enjoy and even connect with your favorite NPC in the same way they do with each other. They’ll even ignore that whiff of author avatar as long as they don’t threaten their spotlight. These things ultimately constitute a lot of that beating heart that makes this your campaign and not just any campaign. They do provide a similar enjoyment to that individualistic lead creative vision that draws people to the “auteur director” like, well, the aformentioned M Night Shyamalan.

The problem is that these elements pose far more risk to your campaign than potential enjoyment. With this particular aspect of long term planning, “as possible” boils down to your own restraint. The reality with the foreshadowing, the plot twists, the NPC’s you spend all night making art for, and the 41 page lore bible* you share with everyone on google docs is that they’re there for you**. The level to which your players engage with this material is the cost they pay for your GMing services.

*For the record, I did not make my players read my 41 page because even though I am the sort of psycho who has a 41 page lore document, I am not so egotistical as to think any other human being would be interested in such a thing. However, I am the sort of insecure psycho that feels the need to Flex™ on how deep and well thought out my world is, so I did share said document on the premise that now my players can look up individual components they may be curious about or have straight up forgotten.
**This is not strictly true. Detailed world building and NPC backstories definitely help provide a logical foundation to feel comfortable when making improvised GM decisions. But it’s mostly true.

The key is that while you may gently nudge and prod your players in the direction of these things, you absolutely must be willing to sacrifice this stuff at the altar of player freedom. And even when you do, there is no way around the fact doing usually feels really shitty. Even when your players are completely unaware of their own destruction of such things, it feels like they are plotting against you personally because their little wish fulfillment vehicles are so much more important than the four hours you spent on this epic boss fight ten levels from now because you were feeling VeryInspired™ that day and now you have to completely scrap* it because they used the power of friendship instead.

*Someone who is not the kind of psycho to make a 41 page lore bible would probably point out you can just reuse this elsewhere and repaint the story relevant parts. Unfortunately, I am that kind of psycho which means I’ll always know that it’s not LoreAccurate™ to reuse or secretly retcon this sort of thing, so I really do end up scrapping large chunks of this sort of thing about 75% of the time.

Deal with it. This is the business we’ve chosen.

Pacing is the Second Most Important Aspect of Your Campaign

Clickbait tactics were not necessarily my plan going into this thing. If they were, I wouldn’t have spent (checks word count*) 15,000 words** and counting***. But naturally, I did manage to save the best for last, which is nice I suppose****.

*Not going back later to double check after the edit. DealWithIt.jpg
**And about 5,000 of which were spent attacking an extreme non-planning sandbox strawman.
***Just kidding, I did in fact double check after the edit.
****Assuming one made it this far, in which case, hi Maja.

Pacing is hard for me to give solid, concrete advice on. There are many reasons for this. The first is that pacing has always been one of my best strengths* as a GM. Because running a game with a satisfying pace has always come naturally to me, there is less deliberate thought to this aspect of my game than there is to, say, the geography of encounter design. The second is that there are a ton of variables that affect a campaign’s pacing that are specific to an individual table. The quantity of players is the biggest factor, as that impacts every single aspect across the board from the length of combat encounters to how individualized story attention is divided amongst characters. This also includes things like your tables preferred combat to roleplay ratio and how much your table likes to talk to each other unprompted in character. Finally, pacing is something that, even more than everything else I’ve talked about, is largely intangible. Bad or floundering pacing is not so much observed as it is felt. It’s that moment when “Oh, sorry. What was that?” becomes a phrase your ears are intimately familiar with.

*And to be fair, let’s also mention some things I’m bad at. Thing #1: my narration, particularly my description of environments and characters, is doo-doo bad. Thing #2: I plan too much. In particular, I cause myself lots of unnecessary pain thinking about potential far out story beats even when I have the discipline to not write said things down.

That said, after a very thorough deep dive into my campaign’s session summary notes and session preperation notes, I have found patterns in the structure of the most successful sections of my campaign, and I think the least successful sessions mostly broke away from these tendancies. Furthermore, I am somewhat confident I can translate the more table specific tendancies of successful pacing to be broadly applicable for just about anyone. Without furtherado:

• Broadly speaking, a campaign can be broken down into units of time. Going from largest to smallest, these would be the campaign itself > acts or story arcs > large problems > small problems/game sessions. This conveniently sorts itself out from least to most important, and I’ll tackle these ideas in that order since that also means it sorts itself out from least to most i have to say about the matter.
• The Campaign:
  • The Campaign is technically speaking the most flexible unit of time. However, I think the RPG system being used almost always inherently lends itself to pretty consistent end points.
    • Systems like Burning Wheel or Legend of the Five Rings which focus on gamifying story and roleplay itself tend to have very specific end points because character progression is tied to the aformentioned gamification of roleplay. Therefor, there’s very little sense in continuing much longer once character progression has reached its end point.
    • Systems like DnD and Pathfinder which have almost structureless story and roleplay are more open ended about this, but tend to eventually reach the maximum possible stakes of DefeatingGod(s)™ because of the player character’s escalating power levels.
    • Because of the aformentioned close connection to system the pacing of the campaign, the rest of my advice will largely pertain to being Dungeons and Dragons specific.
  • There are a ton of problems that naturally present themselves if attempting to take DnD 1-20 leveling with the classic XP rules as written approach.
    • This is because like the classic RPG video game levelling systems that were largely inspired by OG DnD progression becomes slower over time. This is very unobvious because of how abstract and meaningless actual “XP” point listings are and how monsters with higher Challenge Ratings also scale up in how much XP is provided for defeating them. You need to really dive into level appropriate stat blocks to see just how much slower monster XP scales in comparison to player character levelign requirements, but in short, it works more or less like modern video games work and take my word for it.
    • This is extremely incongruous with good storytelling because you want the stakes to escalate on a fairly even level and really, you want the kind of “fight God” stakes to be the smallest part of the story because that very quickly becomes stale and well, the way DnD’s scaling works suggests the most time in your campaign should be sent fighting extradimensional beings if you’re following the classic “level 1 through 20” adventure implied to be the wholistic experience of a campaign by said range of levels.
  • Other than the basic fact of how unruly XP is for players (or the GM) to keep track of, this is why pretty much everyone has moved to a “checkpoint system” which is basically a fancy way of saying the party levels up when the GM feels like it. Hell, I think literally every published adventure I’ve read provides level up checkpoints.
  • However, that hardly solves all our problems. That’s because the type of challenges that can be provided to very low and very high level parties is extremely limited and therefor very unfun.
    • Level 1 and 2 isn’t really a problem because it’s very obvious even in the XP rules as written that it’s meant to be a tutorial based on how rapidly players get to level 3 naturally. Therefor, I think it pretty much naturally occurs to GM’s to start campaigns at level 3 or 5 for experienced players and use levels 1 and 2 for table setting and teaching for inexperienced players.
    • The much bigger problem is with level 15 onwards. My slightly hot take* is that once level 7 spells are on the table, most forms of challenge that can be presented to players is completely obliterated. Levels 15 and 16 aren’t so bad because at least when the party uses a combination of Scrying (level 5) and Teleport (level 7) to completely bypass a challenge, it represents a greatly substantial expenditure of resources. For this reason, I would advise these being the “Avengers: Endgame” section of your campaign. Once the even heavier hitters come on board, the only challenge** left to present is just slapping giant numbers onto the enemy HP and damage numbers, and god have mercy on your soul if you attempt to create a story scenario that can’t simply be cheesed by bending reality to the party’s very will.
      • *The vast majority of people’s campaigns don’t go to 20 and I think this is a fairly popular take. Didn’t bother to research to be sure though 😉
      • Admittedly, my specific approach to encounter and adventure design is probably uniquely ill-suited to deal with very powerful spells. That said, you really do just need to have literal decades of knowledge on DnD mechanics to have any chance of building challenges for high level players that don’t just get steamrolled.
    • The bulk if not the entirety of the story being levels 3-14 might feel too miniscule, but I don’t really think that’s the case. I think traditional story structure fits rather nicely into this pocket for reasons i’m about to get into.
• Story Arcs
  • Unsurprisingly, this is the next most flexible unit of campaign time.
  • 3-5 story arcs or “acts” feels like the correct amount. More than 5 starts to feel like a series of unconnected events. Less than 3 feels like insufficient time to have that natural escalation of stakes throughout a story.
    • If your individual group tends to run through your big ideas very quickly and 5 feels too short, my advice would be to split your campaign into two “seperate” campaigns. Still star the same player characters, but insert a large time skip between the two campaigns. Fill the in between with a session or two of group downtime which concistutes a “Campaign 2 session 0.” This will make the two chunks of story arcs feel connected to one another, with the downtime being the connective tissue for the whole.
    • If three feels too long for the reverse reasons, make the first arc very short with a sudden and unexpected ending. A bunch of ways to do this, but generally, i’d lean on some sort of catastrophic unpreventable failure whether that be a beloved NPC dying or a calamatous event occuring the world. Because your party will be at low level at this point, you won’t need to go to extraordinary lengths to make this development obviously super unfair bullshit.
  • Think of the end of a story arc as where the season would end if its a television show.
  • Between these 3-5 story arcs, I think they usually lend themselves to a large problem or small problem here or there that fall outside of the arcs. The exception to this is when story arcs end in CalamatousEvents™ or introduce BigStoryProblems™ that present some level of urgency. I certainly wouldn’t recommend ending most of your story arcs this way though, because I think these sort of connective tissue or lower tension sessions is what distinguishes RPG storytelling as unique.
  • These in between spaces are also usually when its appropriate for your party to faff about with things like egregious shopping sessions or birthday parties if your group is into that sort of thing.
  • I’d also recommend a “prologue” or sort of an extended sequence that’s more open ended and unstructured. The idea here is that the campaign’s main story can be introduced to the party organically so it feels like something the character’s actions naturally led them into instead of something they’re participating in because metatextually that’s the content you have prepared for them.
    • It’s really not very hard to heard low level players into the direction they need to go if needed. On top of the basic mechanical reality that it’s very easy to kick their ass in combat if needed, players will be busy getting a feel for how to play their new character, how their character interacts with the rest of the party, and how your world feels through the eyes and ears of their new character. For all of those reasons, players especially tend to not have a lot of self initiative and are very happy to latch onto the bait you place in front of them.
  • If just wanting to get straight to the “good stuff” is more appropriate, then I would recomend having a very specific concept for why the party exists in close approximity to one another if not already as a functioning unit in media res. An example of the former might be “you’re students at a military academy” while an example of the latter might be “you’re part of a band of mercenaries.”
  • There’s a natural progression of problems the party is mechanically equipped to solve and therefor the relevant story stakes. I would describe it as roughly following a path of individual problems > community problems > political problems > international or worldy problems > and existential or interdimensional problems.
    • You definitely don’t need follow the entire path from beginning to end. Covering 3 out of the 5 in their logical order of progression is more than sufficient, and I think even 2 of 5 will often do. A very focused campaign with a small scale might end its finale contesting political control or sway over a large city, while a very grandiose campaign might start with our party as Lords and Ladies of various competing noble houses.
    • Pretty much every story arc will consist of one of these “levels” of stakes.
    • The stakes do not need to necessarily escalate to a new level every story arc. The grandiose political campaign, for example, my proposed grandiose medeivalish politics campaign might go like this
      • Story Arc 1: Noble houses vie for installing their preferred monarch on the throne.
      • Story Arc 2: Noble houses fight for the new monarch’s favor.
      • Story Arc 3: Spies from the neighboring kingdom stir up problems in the kingdom’s capital to soften them for an invasion
      • Story Arc 4: Said invasion occurs.
      • See, 3 out of 4 with political stakes with an international problem thrown in as a finale.
    • Incidentally, all of this why I mentioned fewer than 3 story arcs being “not enough” time to get through a satisfying progression of stakes over the course of a campaign.
  • Based on my really bad session notes, I would estimate a story arc for our group constituted 6-12 sessions – each session being approximately 4 hours. I think this is generally a pretty good number, but probably on the shorter end due to our small player quanity.
    • For reference, this resulted in about 40ish sessions worth of “main story” content and 15ish sessions worth of non story arc sessions. The 15 was definitely larger than ideal which was a result of needing to justify in story why i had a few changes in player lineup early on and a few smaller group/individual sessions when i knew in advance scheduling wasn’t going to work for the group.
  • I think how the story arcs fit together and how the stakes progress more or less occurs very naturally throughout the campaign. I don’t think there necessarily needs to be a ton of intentionality to designing the overarching story, but a little bit of basic overall planning helps prevent the campaign from going into a nothing happens holding pattern.
• Large Problems
  • The easiest way to distinguish a large problem from a small problem is to think of story problems in terms of video game quests.Large Problems are those mandatory story quests you get which say very basic things like “find proof that Saren is crooked” or “save the citadel from the reaper invasion.” Small Problems are the quests that tell you to go to a specific person or location to complete an individual task. Some of the Small Problem type quests are required to complete the big story quest, while others exist as entirely seperate, optional entities.
    • How your small problems interact with your big problem varies a lot, and I don’t have a ton of advice other than “do what makes logical sense.”
  • A story arc consists of 1-3 large problems. Generally speaking, a story arc that tends to stay stationary geographically will involve one very complex problem to deal with, while a story arc that involves a traveling from one location or another will typically include the first large problem being the task of the traveling itself and then dealing with the problem once you arrive. A seperate large problem may evolve organically or be planned in the middle for pacing if there’s a blockade the party needs to somehow get past or if the party loses a fight while traveling so has to break free from their captors.
    • I honestly don’t think I ever plan on there being more than two big problems to deal with in a story arc, but as mentioned, big problems will often occur organically as the result of setbacks or unexpected character decisions.
    • Naturally, there will be smaller problems or deviations from the big problems of the story arc that occur. This is especially true early on when story stakes are lower, and therefor time is usually not incredibly urgent.
  • Honestly really tricky for precise advice on this other than the above, so i’ll pull the first story arc from my campaign and try to be as general and brief as possible.
    • The set up: During the “prologue” concept I mentioned earlier, my party was kidnapped by the big bad and forced into a Suicide Squad-esque situation. Basically, do my dirty work or your head will explode.
    • Said dirty work consists four MacGuffins for the big bad. Obtaining the first MacGuffin, as one might expect, constitutes our first story arc.
    • MacGuffin 1 story arc consists of 3 Big Problems: Finding the macGuffin, stealing the MacGuffin, stealing the real one when they find out the first one is a fake.
      • Big problem 1: The big bad got our heroes to the city he knew it was located in. The first two sessions were spent exploring the city and talking to people until they ound out where it was.
      • Big problem 2: Once they found out where it was, they spent a session gathering additional intel so they could figure out how to steal it from the museum, and then a session actually executing the heist.
      • Big Problem 3: The Macguffin is a fake. Luckily, the big bad informs them its such a quality fake he can use it to find the real one. This gets our heroes pointed in the direction of a large, traditional pulp fantasy dungeon which holds the reason shard. They spend one session relaxing in the city, preparing to adventure to the dungeon, and actually traveling to the dungeon. Then they spend 3 sessions in the dungeon, where they eventually acquire the real MacGuffin.
      • I understand all of that sounds very railroady, and it was. I was dealing with mostly a group of players brand new to the hobby, so the first arc was designed to hold their hand while they learned how to play. They were not trapped under the big bad’s thumb forever i assure you.
• Small Problems and Sessions
  • Now we get to the meat and potatoes of designing your individual sessions.
  • To clarify, “small problems” and “sessions” are sort of different things but directly related to each other. A “session” is the regularly scheduled occurance where you and your friends play RPG’s. I think 3-4 hours is a surprisingly common standard for an average run time based on my experience with multiple gaming groups and perusing of other people playing online.
    • Note that when planning this 3-4 hours includes people inevitably showing up late, bathroom breaks, “oh wait, we’re starting? I don’t have my character sheet!” etc. etc. so realistically, you can usually shave off 30 minutes to an hour of this time spent actually playing.
    • If you’re running super long sessions, i think you can generally follow this advice and split prep time into “chunks” of content. Because by god, if you’re the type of folk running 8 hour sessions on Saturday, I hope on god you are getting a nice long lunch or dinner break in there.
  • By contrast, a “small problem” is some form of objective within the story. The most common type of small problem is a combat encounter. In the form of roleplay, this takes on basic tasks like “find the thing,” “talk down that angry guy,” or “negotiate a settlement.”
  • In sort of “balancing” a session or preparing activities for the week, GM advice or general DnD discussion tends to talk about this in terms of how much combat there is versus how much roleplay. I think this is totally the wrong way to think about this idea. Instead, think about it in terms of “mechanical” play or “logical” play. Mechanical play at its core boils down to situations that are primarily resolved with dice or rules in the player’s handbook. Logical play boils down to situations that are primarily resolved by how NPC’s and the world reacts to player decisions.
    • These are by no means strict defintions, and if you’re wondering how one could measure how much of their play is mechanical versus logical, well, now you understand why I said the pacing section was very hard to give solid advice about.
    • Within this context, there is still some element of balancing out combat versus roleplay because combat tends to be more mechanically oriented while roleplay tends to be more logically oriented.
  • When it comes to planning out how much the balance of mechanical versus logical play, a lot of it comes down to your table’s preference. That said, I still think there are sweet spots to aim for.
    • For roleplay, I would say the sweet spot is about two thirds logical play and one third mechanical play.
    • For combat, i would say the opposite. Two thirds mechanical play and one third logical play should be easy enough to acheive if you’re following my principles of story oreinted combat and in particular, all the stuff I mentioned about embracing those “gray area” spaces between roleplay and combat.
    • Since my table is more story oriented, I tend to be about two thirds roleplay and one third combat within that sort of balancing context, and additionally, I tend to make my combat a little more logic oriented with more story driven gray area and a sprinkling of what I’d call “set piece” design where specific, cinematic things will happen and work mechanically in a way that’s designed exclusively by me (i.e: not consulting the rulebook) for that specific situation.
    • With all of the above, I would advise not going past the halfway point for logical play in combat or mechanical play in roleplay. In combat, this will lead to outcomes feeling “fake” or “pre determined.” In roleplay, this will make the people in your story feel like non-human animatronics or video game characters.
  • If you want a session or problem to be very roleplay centric, then you need to create some form of structure. This is because a single combat small problem inherently almost always takes much longer than a single roleplay small problem.
    • Don’t believe me? Come up with a random topic and try to ad-lib speak about it for ten minutes. If that doesn’t very hard to you, stop reading this right now and try to speak out loud for ten minutes about animals. That’s it. Time yourself The only restriction is you can’t look up anything to specifically go off of ahead of time. Done yet? I promise you that was extremely hard if you managed to accomplish it at all.
      • Let’s be real, you didnt’ accomplish it.
    • The aformentioned talk with animals exercise also sort of explains why we need structure for roleplay oriented problems. It’s not just the players that can’t talk for ten minutes straight, it’s very much a GM problem as well.
    • Without getting too terribly much into an extremely long spiel I have about designing social encounters, there are two forms of structure you can apply to a social scenario.
      • The first is with a tangible, physical sense of location. Like, literally draw a map and know where the points of interest are. The map can look like shit, and the players don’t need to see the map. You just need to know where these points of interest are, why they’re points of interest, and how they relate to each other in function and physical space.
      • The second is with a list of relevant NPC’s. Each NPC should have a name, race, basic description, 2-3 sentences about their personality and 1-3 sentences about their ambitions or objectives.
      • The idea with this structure is it gives a foundation on which to improvise. If you’re planning a single roleplay oriented session, just one of these two things is fine, as preparing one will allow you to mostly improvise the other. If you’re preparing an entire Large Problem that’s centered mostly on roleplay, you really should have both as well as some basic ideas on how the NPC’s interact with each other and the various points of interest.
  • A well run combat encounter should usually last 10 to 30 minutes. Inevitably, you’ll almost always have one of your prepared combat encounters run way longer than you wanted it to because the group spent ten minutes debating what to do or the wizard didn’t realize they were out of level 3 spell slots so they panic when they can’t use fireball.
    • If you’re running a very combat oriented session, 4-6 is usually want you want to have prepared.
  • Particularly once you get to level 9 (i.e: when 5th level spells come into play), it’s really important that your combat oriented scenarios account for when player’s could potentially rest. This is because DnD still maintains the philosophy of its original design that players should expect to be engage in 6-8 fights between long rests. One gander at a dnd stream or real play podcast and you’ll quickly find that absolutely no one does this. This is a huge balance problem because a barbarian is balanced around the idea that they can pretty much always rage while a wizard is balanced around the idea that they’ll have to split 14 total spell slots between an average of 7 encounters per day, so it’s no wonder why spellcasters feel overpowered in comparison to their martial oriented breatheren in most campaigns. (Incidentally, another reason DnD quickly becomes unfun at high level).
    • Sometimes, a scenario will naturally solve this for you on your own. Something that creates natural urgency like an epic final battle outside the castle walls doesn’t really leave a lot of room for long resting.
    • By contrast, other scenarios, like diving into a huge dungeon, usually don’t present much if any urgency. That’s where the rest balance becomes a problem because there’s really nothing inherently preventing the party from just long resting after each encounter.
    • Usually, the suggested problem for this is patrols that may ambush a sleeping party. I find this to be completely ineffective for a whole host of reasons, the main ones being that the optimal solution is then to just hunt and kill the patrols before long resting.
    • Unfortunately, the only real solution to this is to introduce some form restraint that limits rest that is obviously (to your players) very gamey and artificial. My go to for this was the idea that traveling somewhere like a dungeon requires food and water. Since it will spoil very quickly, it needs to be protected in magical ration kits you can buy at the store, which are quite expensive and therefor essentially serve as a number of “rests” the players can take while traveling in dangerous uncilvized areas. It works well enough and also serves as something that gives gold actual value for the players without needing to stuff stores with valuable magic items like all my merchants are video game vendors.
      • Off the top of my head, I can certainly think of other ideas. A shorter dungeon-ish area might have a curse that prevents sleeping. A more linear dungeon might have a creepy, invincible Mr. X type figure that will positively crush the players if they rest too much and let him catch up.

“Oh my god,” my theoretical reader might say. “How does anyone keep track of all that?” Fret not. The idea is not to sit down and go mad trying to keep track of all this. The idea is to have these general ideas in the back of your mind. Trying to build a campaign with some pacing formula or exact structure will without a doubt end in disaster. It’s the same reason that most critically acclaimed creatives will tell you that screenwriting classes are bullshit. Shit like “state the theme on page 5” or “midpoint on page 55” is mostly worthless.

The reason why movies tend to adhere to fairly formulaic structure is because a good screenwriter simply writes their story as it comes to them and then edits it so that it has some resemblance to storytelling structure. A director (sometimes the same person) then takes that script and adjusts it further as actors add their own element of interpretation to individual characters and the pacing of the story evolves as it transforms from written prose to visual. Finally, a film editor takes the existing footage and pares it down further to form it fully into what is usually some recognizable structure. Even then, the vast majority of film editors will tell you that while they have a variety of learned techniques and basic guiding principles at their disposal, at its core, it comes down to “feel.”

Pacing an RPG adventure is no different other than the fact that a lot of the “editing” is occuring completely improvised by players and the gm alike during hte course of playing a session. In this context, of course it’s going to be mostly a “feel,” thing. The purpose of writing this massive outline on how my adventures have been structured is as a diagnostic tool. When the “feeling” is wrong, when players and checking their phones or having side conversations, intellectualized concepts on RPG structure serve for you to diagnose your problems and adjust accordingly.

And the thing is ultimately, all (checks word counter again) 20,000 words matter very very little in comparison to my last point.

Panning for Gold is the Most Important Aspect of your Campaign

Remember towards the end of the DMPC section where I said don’t stop the players when they’re on a roll? Do you remember when I said the end of Critical Role Season 1 was our unattainable highest posisble aspiration? Those incredibly finite moments where the stars align and you create a genuine moment of heightened human experience is what I call gold. They’re why we do this hobby. In spite of my beef with a lot of DMing convention, I think this noble spirit is at the heart of most of this advice. Gold comes from the players’ interactions*, so the logic** is to build the campaign in such a way where nothing happens unless the players actively initiate interactions.

*Incidentally, I polled some friends – including my players – for their favorite dnd moments. I think this is usually the case, but I was surprised pretty much everyone had at least one moment that i’d call mostly “scripted” or fairly deliberately planned content. In short, while I stand by the general idea that we should be clearing the way for when our players feel the confidence or narrative momentum to really engage, i think the premise that those talk about it forever moments must stem purely from improv or character initiation is a little flawed.
**And I’ve talked in length about why I think this idea is flawed – mostly that it boils down to roleplaying is hard enough so don’t make it harder by also forcing your players to drive the narrative.

Let’s go back to the river metaphor. If we recall*, the characters are rushing water taking the path of least resistance**, and the planning and session preperation are the river bed which provide that path of least resistance. The river in its convoluted and pleasing shape then is a story which somewhat resembles emotional ups and downs and gradually escalating stakes and all those other basic story ideas that make narrative satisfying. We at the table, GM and players, are prospectors sitting at the rivershore and panning for gold. You won’t find gold every session. If you have fantastically wonderful elite players*** like I do, you’ll be blessed enough to find gold every 3 or 4 sessions. Or rather, you’ll find gold every 3 or 4 sessions after you get past the growing pains of getting the basic inate understanding for how characters play off each other and after the GM gets sort of that flow to which nudges the players and their characters react best to and how the session can remain smooth and steady while maintaining clarity****.

*And my sincerest apologies to those who remember before with perfect clarity.
**Remember, players want their characters to want the eagles to fly the one ring to mount doom. It is not the players job to think about things like character development or story arc. It is your job as the GM to create situations where that naturally occurs, which again, is at the core of my general critique against “dont plan anything” GM styles.
***Whom incidentally, also happen to be very attractive.
****Clarity of intent had its entire own section before it was cut for time. Yes, haha, i know cutting things for time is a very funny concept at this point. In short, situations presented in narration, maps or encounter design should always be crystal clear to the players, even if this means damaging the pace of your game to do so. That is because it is extremely immersion breaking when a player makes a character decision they would not have made if you the GM had done a better job explaining the thing. Basically, whenever possible, player and character knowledge should be aligned. It’s totally fine if a bad decision is made because a character lacks sufficient knowledge. Yes, there’s a lot of gray area here when it comes to sort of the strategic element of combat. There’s not really a great solution there.

And every once in a blue moon, you’ll hear that “tink” in your pan. The moment where you’re not just digging flakes out of sill, but finding a genuine solid nugget of glorious story excellence. These are the moments you will think about on a weekly basis and talk about on a monthly basis. This is the shit you will remember for the rest of your life.

There is undoubtedly a certain deep frustration that happens when that doesn’t occur naturally over a long period of time. The harsh truth is that you cannot manufacture rapport or DnD gold for the same reason you cannot manufacture chemistry on a sports broadcast. It is a stew made of the intangible ingredients of the human soul. If you try to “chase the dragon” in regards to gold, you will only find that refuses to show itself to the metaphorical miner who arrives to the rivershore with a pickaxe.

The other hope with this intentional and deliberate design philosophy is, since 95% of your campaign at best will not be just basking in the rays of the sunlight as it reflects of your gold, that we are raising the floor of enjoyment your players experience in the campaign on a moment to moment basis.

I’ve talked ad nauseum at this point about story and character, and there’s a bit of an implication, at least from those who don’t understand, why I’m not just writing book or watching a movie. Well, there’s the somewhat obvious reason that TTRPG’s are an actual social activity you can do with your friends while the others are not social activities even if you are using things like a movie theatre as pretense* to gather socially.

*You are not, unless you are a real jerk, socializing while watching the actual movie. Therefor, the movie itself is not a social activity.

The other is that there is a very specific itch that is only scratched by this hobby. Science has shown that when you are playing in a DnD campaign, your memories are stored in the same “jar” as the memories of things that happen to you – the normal boring person – are stored in. This is in contrast to the “jar” that memories of events outside of the self happen – the same jar where memories of movies and books and TV are stored. I don’t know if I’d go quite so far as to suggest that DnD at its best is therefor more powerful or other qualitative states that imply “better,” but I don’t think it’s off base to suggest it is unique and different. Even before the science of this idea existed, we had a vague, inherent understanding with the common observation that players use personal pronouns like “you,” “I,” and “we,” when describing sessions to each other.

I hope I haven’t been too dreadfully long (rhetorical, of course I have), and I hope at least one person* made it to the end. I have one very brief last message for you all. Don’t be intimidated by this. Go try to DM a campaign. This is really all philosophy – and what is tangible is purely meant as a diagnostic tool when things are not going the way you want them to go. This was as much (probably more so, considering the length) an exercise of me trying to get my intangible ideas down to some level of tangible thought for future design. Anything the rest of you learn from me is just a bonus.

*Hi Maja

And hell, even if you are as violently opposed as ever against every single word I’ve said, that’s ultimately okay as long as you and your friends are having fun.

Advertisement