An Attempt at Truly Improvisational DMing

Over the course of my DMing career, I’ve slowly moved from a rigid style to something looser. When I started off, I was one of those DMs who had to have every beat of the story planned, every NPC statted out with motivations, every monster unique and meaningful, and every map complete and detailed down to each five-foot square. The more I dealt with actual play, the more I realized I couldn’t work that way, and I adopted a looser policy, where I’m comfortable enough making things up on the fly but I do have some idea of where we’ve going. Lately I’ve been thinking I could handle full-on improvisational play, so this past week I put it to the test.

I like food and I like television, so unsurprisingly I’m a fan of the Food Network. It has a long-running game show, Chopped, where contestants are asked to make dishes with a limited time frame so judges can determine whose dishes were best. The quirk is that at the beginning of each round the contestants are given four ingredients, which they have not seen until immediately before the round begins, and they must use those ingredients in their dish in some way. A typical round might begin with the host asking “Can you make a delicious appetizer using ground lamb, napa cabbage, pomegranate jam, and boxed macaroni and cheese? You have twenty minutes, time starts now.” It’s basically improvisational cooking, and it’s a concept I’ve been mulling over applying to D&D for a while.

On Saturday my opportunity arose, as we’re in a lull for our normal campaign and I had three players who were up for shenanigans (one specifically requested something, and I quote, “crazypants”). I told them to create characters in Pathfinder, the system we know the best. That was literally all the advice I gave them. In fact, I told them specifically not to tell me their character concept or even level until we got to the session, and I deliberately planned absolutely nothing. When game started, I had them provide me with some Mad Libs-style adventure seeds. Here’s what I asked them for and their responses:

  • Place: A giant ship traveling between continents
  • Thing: Map
  • Creature: Hill giants named Fred

They all sent their answers over chat at once so none of them knew what the others had chosen, for maximum chaos. Since building an adventure off that seemed too easy, I asked for another round:

While they built macros for their characters in MapTool, I started building an adventure that involved all six of those seeds, appropriate for their level and party composition, complete with monster tokens and adventure maps. I ended up with a cult of Pazuzu who wanted to summon their lord to the Material Plane. Two of his initiates, hill giant monks with a combat style based around jumping around the battlefield, gained passage on the same giant ship as the party. When the ship passed by their island, they stole the ship’s maps and gave it to a belker acting as their accomplice. The party followed the belker to the island, where another initiate tried to talk them into joining the cult of slavering devotion to and domination by an evil extraplanar overlord friendship with a helpful, powerful god. In true D&D fashion they instead killed everybody.

The players, of course, had their own agenda. They conferred with each other on their characters, and they selected a theme they opted to keep secret. I was to induce what they theme was based on their character descriptions and actions. I only figured out shortly before the final battle that they were all playing Jedi as a bloodrager, a bard, and a swashbuckler/paladin, though I contend I would have figured it out earlier if I hadn’t spent so much time trying to figure out how the bard worked into things. They had a good time bantering about their secret theme, and I got my revenge by telling them their Jedi expies were now canon in a campaign setting based on Edwardian Europe, so we all had a laugh.

My goal was to test myself, to see whether I could pull a session out of thin air with zero prep time, and I think I succeeded. I’m certain I could have done it even faster than I did if I hadn’t had to deal with finding enemy portraits, writing their attack macros, and creating combat maps out of our tile resources. If I’d done it at a table, with my miniatures handy and a battlemat, I probably could have built the session in half the time. I will deduct points because I legitimately thought the Rock of Gibraltar was an island instead of peninsula, but the players were gracious enough not to call me on it. All told it worked out fairly well. I haven’t asked the players for a detailed grade but I’m giving myself a tentative B+.

Designing and running this session was a ton of fun, enough that I’m trying to again next week with the bonus seventh adventure seed “A sort of Pathfinder version of pre-Empire Star Wars”. I do think this concept for session design requires a really light mood. Everything was already a bit wacky, so nobody batted an eye when the cultists started talking about becoming one of Pazuzu’s friends by welcoming him into your life and mind. It also helped that I had resources available to me, like the NPC information on the Pathfinder SRD or the tokens and macros from previous campaigns where I’d already solved a lot of my output formatting problems. But that’s largely what I’ve said all along, where improvisational DMing works best when you have some way to fall back on numbers when you need them. As I’d thought, I wouldn’t be able to do it all the time, but as a one-shot or a breather session it’s a great break from normalcy.

Posted in Campaign Writeups, DMing, Pathfinder | 1 Comment


Some years ago, when I frequented the Wizards D&D message boards, there was a user whose signature said something to the effect of “the spiked chain exists to find the DMs brave enough to ban Core material”. That is, the spiked chain is intentionally broken, and it should be stripped from the system even though it’s in the very first rulebook, and this is a secret test of character to find which DMs are truly good. I’ve never been sure whether was a self-satisfied delusion from a lone DM who thought they spoke for everybody, or a cry for help from a player who’d seen too many games go bad because of rules abuse. The very concept that a single item could damage the game that badly seemed alien to me, and it still does.

I generally argue that there’s nothing in the rules guaranteed to be a game-breaker. Some things are incredibly powerful in specific situations or combinations, like the blaster sorcerer with access to every cure spell or the monk who can jump so far all witnesses become their slavering sycophants. But in a cooperative game, you can usually solve these by telling the player “hey, that’s not fun, can you power it down some?”. A lot of these builds are complicated structures, where if the player removes a single feat or spell they remain powerful but no longer disgustingly so. And even if they don’t, there’s usually a way around these builds if you’re willing to exploit them. Nothing can turn a game upside-down on its own. Only specific, borderline malicious actions you take with those items can.

Still, I’ll admit it’s easier to cause a problem with some things than others. It’s easier to do insane damage with a hulking hurler than with a samurai, with haste than with a two-weapon fighting tree, and with a vampire than with a gnome. Some things require a more delicate hand to keep things fun for everybody.

Chief among these things, of course, is the spell wish. Wish is less a spell in the rulebook and more a creature of legend, a dusty passage in the back of the Player’s Handbook included for completeness but not to be taken seriously, in the same way the edge of a map might say “here there be tigers”. DMs treat it as a nuclear option, so enemy wizards and demons can fiat their way out of a bad situation, and as a campaign capstone reward, so players can satisfy their desire for ludicrous requests but the DM absolves themselves of adjudicating the consequences. In the eyes of D&D players, it’s not there to be realistically used.

Wish does have game-breaking potential. Its powers are intentionally loosely defined, though they include “pluck anybody from anywhere in existence and deposit them anywhere else in existence”, “heal maladies at the level of a deity walking the earth”, and “be an 8th-level spell, because why not”. It’s this ambiguity that causes the problem. DMs are worried players will use the wish too well, either getting themselves some distressingly powerful gear, granting themselves a disruptive power, or sabotaging the game’s intended direction. Players are worried they won’t use the wish well enough, and the DM will interpret their request in whatever way hurts the player the most. It’s a cold war where neither side wants to use wish because they’re worried the other side will twist it out of control.

Well, we’re in a campaign with genies, so I figured wish was going to come up at some point. In our last session the players befriended a malik, and here are their wishes:

  1. As many diamonds as wish can create, teleported to the party’s home base.
  2. Eternal youth for the party’s middle-aged leader. Specifically, “I wish to enjoy the physical benefits of youth forever.”
  3. The exact location of their missing party member, since the point of the campaign was to find her.

An astute reader may notice that these are the exact concerns I gave above about how players can damage the game with wish: inordinate wealth, a disruptive power, and a near-instant solution to the campaign’s driving conflict. They also all gave me ways to hurt the players with them: they didn’t specify the source of the diamonds, they left “eternal youth” open-ended, and they asked for a location open to misinterpretation. We were set up to demonstrate everything wrong with wish.

And here’s where we get back to my original argument: there’s nothing wrong with wish, just with how you use it. My players and I do not have an antagonistic relationship, no matter how much we I pretend we do. All of us knew the pitfalls of wish, and not only did we avoid some, we deliberately invoked others. When the players asked for wealth, they explicitly noted that they did not specify from where the diamonds came, expecting it to be a plot point later (their origin is, of course, a spoiler). When they asked for youth, they deliberately chose an open-ended wording and left the rest to me. I puzzled over it for a few days trying to find an option that gave the player what they wanted but didn’t punish them inordinately for it, because I didn’t want to ruin the character any more than they did.

Most importantly, they didn’t actually ask for their ally’s location. First they tried to annul the contract that caused their ally to leave in the first place, and I told them that wasn’t possible, as that actually would end the campaign. They then asked if they could find out where she was, so I looked through rulebooks until I found a spell they could duplicate that did exactly that, and I adjusted my plans to compensate for cutting out a few sessions of trying to find her. We worked together to find a solution that met both the characters’ criteria and the players’, and chief among the players’ was “this is a good game and I want to play it next week too.” There was no battle of wits, no attempt to damage each other’s play experience, and as a result this overtly powerful spell not only didn’t break the game, it improved it.

Wish isn’t the problem. Neither are the spiked chain or the hulking hurler or firearms that target touch AC. The problem is an antagonistic relationship between players and DMs. Somewhere along the line, we as a hobby decided this was how the game should be run, with the DM trying to prove his or her mastery over the game while the players try to outsmart them at every turn, and that’s rubbish. You don’t solve a problem like that by pruning every part of the rules with the potential for abuse. You solve a problem like that by not abusing them.

Posted in Campaigns, D&D 3.5, D&D 3rd Edition, DMing, Laws, Pathfinder | 3 Comments

The Art of Art

Our campaigns tend to have three names: an real one, a colloquial one, and a snarky one. I’ve talked about how The Eight Arms and the Memento Mori quickly became The Mosnter-Hunting Campaign and soon after became The Monster-Friending Campaign. Less frequently used are the transition from The Eight Arms and the Shadow Invasion to The First Eight Arms campaign to Victorian Greyhawk (especially interesting because the campaign is set in neither the Victorian era nor Greyhawk), or from Fortune and Glory, Kid to The Mystara Campaign to The Terrible, Terrible Mystara Campaign. Our current campaign is no exception. Ostensibly its name is The Eight Arms and the Contract of Barl, but before it even started it became The Plane-Hopping Campaign. It doesn’t have an official snark name yet, but I’d like to propose The Art Campaign.

Visual art is a double-edged sword in deliberately non-visual media like tabletop gaming. When done well, it can put everybody on the same page about something or provide seeds for exploration and investigation or give the players that sense of wonder they can’t get from a simple description of “a big city on an island”. I’ve had whole adventures happen because I found a piece of art for an area, saw something in it, and adopted it into the session, or because I found a picture of a character and decide he or she was too awesome to not use. When done poorly, it disrupts the players’ mental picture or changes the style of the game to something you don’t want or provides a source of endless frustration as you search and search for the perfect piece of art that you know exists but can’t find. Nothing throws me out of whack like seeing five characters in five different art styles and being told they all exist next to each other in the same space.

Contract of Barl, weirdly is doing both. I knew it was going to be an art-heavy campaign from the start because of the plane-hopping. Places like Bytopia are inherently weird and we need a visual frame of reference to understand why it’s different from what we expect and how it works. I also decided it was high time to populate the campaign wiki with pictures, after I spent something like days getting the wiki to accept file uploads in the first place. Everything kind of came together here and now to make this the campaign about visuals.

The good news is that we’re in the process of getting character portraits for key Eight Arms personnel and it’s going absolutely swimmingly. Before too long we should have all the characters in this campaign, and then it’s on to the founders from the original campaign. They seem to keep popping up in stories even if their players don’t, so it strikes me as a good investment, and a bit of a present to the players for putting up with me. It you’re as thrilled with the portraits as I am, that picture is a link.

The bad news is just about everything else. Art of the planes is a great idea, but a lot of the planes and the sites on them are really specific. Bytopia is a plane of communities and civilization, except the sky is another parallel continent of untamed wilderness, and you will be hard-pressed to find a picture of that without very definitely finding a picture specifically intended as Bytopia. The more generic the plane, the easier it is to find art of it, but the more boring the plane itself is. A true plane-hopping campaign needs to go to the weird places, and I’m finding myself making do with whatever I can cobble together with all the Photoshop skills of a sleepy koala.

The monsters are no better. Past campaigns have had enemies like orcs (incredibly common as long as you like the World of Warcraft style, and I do), weird nightmare creatures (also easy, especially if you look for art and build the monster around it), animals (DeviantArt loves weird animals so much I’m not even kidding) or simple, actual human beings. The most relevant creatures in this campaign are efreet, which are not illustrated all that commonly. I can’t really do a search for “like a genie, except with red skin and horns and legs instead of trailing off into wispiness.” The closest I can find are devils, but even that’s a generic term and I end up searching through hundreds of pictures to find one or two I can use. Then consider the most common PC race in the campaign, ifrits, who are normal people but their skin is orange and their hair is on fire. They’re pretty much Pathfinder-exclusive, and finding art of them that isn’t from a Paizo book is rough. Then come fire giants, which you’d think would be fairly common, but no. I managed to come away from my search with maybe two good fire giant pictures, which is significantly lower than the number I need.

Sometimes when I hit walls like this I can take a step back, reconsider what I’m doing, and come up with an alternative that fits my resources. Here that’s not as possible. This is a planar campaign, and we need pictures of planes. It’s a campaign about an ifrit, and we need pictures of ifrits. The Zelda campaign had exacting standards for what worked and didn’t work for enemy portraits because the players already had an expectation for every monster, and I still spent less time for better results than I am on this campaign. I’m mostly finding whatever art I can, then building NPCs and locations around it, which means the art is defining the campaign even more than normal. It’s working out okay so far, but it’s not great and I worry it’s unsustainable.

So for better or worse, this campaign is as much about art as it is anything else. I’m not happy about it, but I’m happier struggling through pictures than I am changing the campaign concept. It does make me long a bit for the days before DMing software, where I described things to players instead. There were misconceptions and misinterpretations, and we regularly forgot ongoing status effects or hit point changes, and I had to print or write monster stats to bring them to a session, and the more I reminisce about them the worse those dark ages actually sounds. But at least I spent less time on Google Image Search.

Posted in DMing, Pathfinder | 2 Comments

House Rule: Three-Axis Cosmology

Especially astute readers may have picked up on this in the last article:

…the players suffered a planar travel mishap and ended up on Carceri, the plane of prisons…Instead I told the players on their way out the door that they had actually landed on Utopia, the plane of city.

This is important because they’re mutually exclusive. The Pathfinder planar cosmology has Utopia but not Carceri, and D&D’s Great Wheel has Carceri but not Utopia. I wanted both in my setting, so I had to find a way to squash the two together without creating problems with the overlapping planes.

More than five years ago (cripes), I wrote about a house rule in my campaigns, three-axis alignment. In additional to good/evil and law/chaos, my worlds have a third axis, action/inaction. A character can be lawful good active, where she rides out and slays evil or performs great works, lawful good passive, where she lives quietly but nobly, or lawful good reactive, where she is mostly passive but becomes active under duress or when a need presents itself. The issue is in how much she feels the need to impress her views on others or the world. It’s a measure not of how a person feels, but how strongly they feel about it. To be entirely nerdy, if good and law are the direction in a character’s alignment vector, active/reactive/passive is the magnitude.

Since that post I have started leveraging this axis in my campaigns, effectively making it mandatory rather than optional. Nobody has wielded a gumption sword to deal extra damage against passive players, but that’s less because I’m not committed to the idea and more because it’s a really expensive weapon enhancement at the levels at which we play. The option is always there, within reach for any player or NPC who wants to use it.

This presented a small problem when we hit the seventh Eight Arms campaign and got to plane-hopping. Both Pathfinder and D&D use two-axis alignment and their cosmology reflects that. D&D uses the Great Wheel, a circle of planes aligned along moral and ethical lines, and Pathfinder uses the Great Sort Of A Box With Planes In, a smaller set of planes aligned largely the same way. Both are two-dimensional. If I wanted a third alignment axis, I needed a third dimension, because that’s what the word “axis” means.

The thing is, something like this already existed. Most versions of d20 have the elemental planes, a set of planes representing fire, air, earth, and water. Alongside them in a separate-but-equal capacity are the Positive and Negative Energy Planes. They don’t really fit in a wheel; you can’t say fire is closer to positive energy than it is to negative energy because those concepts don’t have meaning to fire. They’re defined more by their opposition to each other: fire is far from water, air is far from earth, and positive is far from negative. It’s like an octahedron, or a d8 if you can’t reach Wikipedia from here. Each point on the octahedron is an energy plane, and each plane is “adjacent” to four other planes and opposed to the sixth.

But then, some versions of D&D have the para-elemental and quasi-elemental planes that exist in the confluence between two energy planes. For example, between the Plane of Air and the Plane of Water is the Para-Elemental Plane of Ice. Between the Plane of Fire and the Negative Energy Plane is the Quasi-Elemental Plane of Ash. If each energy plane is a point on a d8, the para- and quasi-elemental planes are the lines between sides. Each connects to two and only two energy planes, and it is what happens when those planes interact. These planes have a long and varied history throughout D&D, in which they sometimes exist, sometimes don’t, sometimes exist but without being separate planes, and sometimes exist only if you squint real hard and turn the book sideways.

The point is that this octahedral concept gave me the idea for a custom three-dimensional Outer Plane cosmology that let me add the effort axis to the planes. It also let me marry D&D’s planes, which we know and love, with Pathfinder’s, which is what every published monster, spell, and deity assumes. I ended up with this:

Each of the planes on points represents pure chaos (Limbo), law (Utopia), good (Nirvana), evil (Abaddon), action (Purgatory), and inaction (The Outlands). Each edge is a place that connects two of these concepts: Pandemonium is the chaotic/active plane, Elysium is good/inaction, and so on. Pathfinder’s planes mostly got to stay where they were except for Elysium. A bit of this is because I didn’t want to disrupt the game’s assumptions too much, but mostly it was because they were so generic; the D&D planes have more character, so they were easier to shift to active or inactive. Just like the standard planes, these inflict penalties on characters of opposed alignment, and the outsider races living in each generally act according to their role on the effort axis.

I did have a finite number of places to put planes and every one of those places had to be filled, so there were changes. After some deliberation I cut Ysgard, because I kind of feel like the whole Norse concept is played out and doesn’t fit much with an Industrial Revolution setting besides; Gehenna, because a plane whose whole concept is “place with steep surfaces” doesn’t merit preservation; Arcadia, because it’s incredibly boring and the “perfection” angle is already done by Utopia; and Acheron, because screw Acheron. I merged the Beastlands with Arborea, because that separation had always seemed weird to me. Instead of two planes whose gimmick is “nature, like, really really hard”, now there’s one with elements taken from both. None of the existing planes made sense for chaos/inaction so I added the Maelstrom, which I think I invented out of whole cloth except for the name. Everything else fit nicely into the new framework with a minimum of jostling or reinvention.

We’re also using the para-elemental and quasi-elemental planes, which means the Inner Planes and Outer Planes share a shape and make sense together. This happened to tie into a seed I planted in the first Eight Arms campaign about the nature of the universe, in a complete accident which I had planned the whole time because I’m incredibly smart and attractive. Since it’s not a wheel or a box, I’m calling this cosmology the Great Edifice (also foreshadowing) and I’ve started adding information about the planes to our campaign wiki. I don’t know how much of this I’ll fill out, but at least it’s there in case a player wants to know.

Posted in House Rules, Pathfinder, Setting Design | 2 Comments

The Art of the Retcon

I’m wrong a lot. Like, a lot. It’s a big part of the reason my first campaigns were so incredibly linear, because I couldn’t reconcile being the all-powerful expert running the game with being the guy who thought his rubber pot lid was oven-safe. I thought the DM had to be right all the time, and that included knowing how the rules worked and where the plot was going and whether using this monster was a good idea and so forth. I still kind of feel this way: Law #1 is “The DM is always right”.

But the first corollary is “But not always entirely right”. In this case, that means the DM is right, but not necessarily right now. Tabletop gaming is a storytelling medium that lets you tell stories over time, and that gives you a chance to revise earlier decisions in a way self-contained media can’t. A DM can say anything he or she wants, but also has the power to change their mind later and issue a retcon if that makes the game better.

Retcon is short for “retroactive continuity”. It’s the act of making a change to a story such that an event that already occurred instead occurred differently, and everything else in the story uses that change instead of the original version. Rather than belabor the point with further examples, I’ll just link you to TV Tropes, though I also recommend the full version of the manga page in the trope picture.

In D&D and other tabletop games, retconning is easy: the DM tells the players something happened a certain way, and that’s how it happened. It’s the equivalent of a text box in a comic book that says “From now on Superman was raised in Queens — Ed.” It very obviously breaks the fourth wall, but it doesn’t matter if it’s more heavy-handed than in other media. It’s partially because of the direct relationship between the creator (DM) and consumers (players) that other media doesn’t have, but also because the consumers are themselves creators. A player can say “I’ve decided my character is a blonde”, and everybody just accepts it without several essays of justification and a court order. Movies can’t do that, and if you disagree, please consider researching one James Bond.

The most important part about retconning something in a game isn’t how to do it, but when. Obviously it’s a terrible idea to retcon away a player’s moment of glory or something that ended up being incredibly fun. But a lot of systems also track character resources, and the more resources there are, the harder it is to back up to a time before they were expended. Fate systems may only need you to remember a few stress tracks or fate points, but in D&D you have to remember hit points, cast spells, used powers, consumed items, spent ammunition, and the like, on top of forgetting all the story changes. This is why it’s generally a bad idea to retcon a fight; too much goes into the result to change the result on the fly.* It takes a very specific kind of player or software to remember every action that occurred in a fight and set things back to how they were beforehand. The same may be true of a long skill challenge, or even exploration or interaction encounters. After all, suspense and emotional investment are resources you can spend, just ones you can’t track numerically or roll back.

There’s one exception to this: retconning a TPK. I completely understand if a party decides a particular fight was objectively bad and letting it stand ruins the campaign entirely. I usually feel there are story-based ways to get around it, to let the failure stand and roll into something else, but in some situations that just isn’t possible. Here a retcon still isn’t a good idea, but it may be the best idea among bad ones.

The DMing books I’ve read actually do recommending using a retcon for one specific case: rules questions. There’s an oft-cited example of the player who tries to do something ludicrous and it’s not immediately clear whether the rules allow or even cover it. Consider the player who wants to summon an elephant ten feet above the wizard. A DM may not know whether this is legal, and finding the section in the rules that says “you can’t summon a creature into an environment that can’t support it” can take forever, especially in the days of paper books without search functions. The books recommend the DM make a decision and use it until he has a chance to verify, usually between sessions, then use the correct rules. This is sort of a retcon in that technically things didn’t happen that way and the players are free to invent another reason the wizard suddenly took two hundred damage and became flat, but again, nobody is recommending re-running the fight without that specific action.

My retcons tend to be story-based, and for me the “when” is “whenever the decision I made affects how the game will play out for the worse.” Sometimes it’s small, like when I changed the caliphate in my current campaign to a sultanate. Between sessions one of my players pointed out how strongly caliphates were aligned to a religion, and I didn’t want that, so I changed it. Except to an especially historically, religiously, or etymologically aligned player both words mean basically the same thing, so the context shift wasn’t a problem.

The other big one we did was when the players suffered a planar travel mishap and ended up on Carceri, the plane of prisons. Carceri was the result of a random roll at the end of the session, and as we were packing up I tried to think about what I would do with Carceri next week. The answer was, unfortunately, nothing. There was no fun idea I had related to the campaign plot, and doing a side quest there would derail things just as they were starting to pick up. Instead I told the players on their way out the door that they had actually landed on Utopia, the plane of city. They had allies there, it was relatively non-hostile, and there was any chance something would happen that could lead into this campaign or another one later. By all metrics it was a better landing spot, so that’s what happened. At the next session the players faffed about on Utopia, purchased some perfectly cubic pastries, and went on their way without problem. I didn’t have to scramble for a meaningful plot, the players got to do what they wanted without random interference, and we got everything back on track much faster than we would have if we had adhered to the will of the dice.

The point of both is that they made the game better, one because it helped the plot and the other because it helped our impression of the world. That’s ostensibly the point of the retcon: something wasn’t working, so you fix it. It’s a delicate task to decide what’s worth changing and what isn’t. Most parties might decide changing “caliphate” to “sultanate” isn’t worth learning how to pronounce a second word, or they like the feel of the word better regardless of the religious connotations, or they like the connotations and want to adopt them. They can decide what they want to do with a ninety-second conversation. Bigger changes require more consideration and perhaps table-wide agreement, but the point is that it’s possible. You’re not beholden to your past decisions any more than the players are, and it’s well within your rights to change anything from an NPC’s name to the entire grappling ruleset if it’s affecting how you feel about the game.

I guess my suggestion is the same as always: talk to your players if you’re going to change anything major. There’s a chance any change that enhances your game will hurt theirs and there has to be a balance, if not a third option that makes everybody happy. But for small changes, you’re probably fine. Your players aren’t likely to judge you harshly if an NPC’s race or accent changes. The ability to adjust continuity like that is one of the advantages of this type of media, and it would be a shame to ignore it when it’s needed.

* — This said, I kind of like the idea of a campaign where the battles are intentionally ludicrously hard, where the players have to try them again and again until they succeed. Maybe something like Edge of Tomorrow where the players are stuck in a time loop, or something that mimics a video game where the players take snapshots of their characters at “save points”. But since repeating fights is intentional, those aren’t really retcons, and that’s beyond the scope of this article.

Posted in DMing | 3 Comments