DMing with Charisma

If the premise of the Great Tower of Oldechi had one problem, it was that every floor was separate from each other. The players might meet other characters who climbed the tower, but an NPC from Floor 1 would not show up on Floor 2, Floor 3, or ever again. On the one hand it let me focus on the important recurring NPCs, like other climbers and the tower-spanning organization, but on the other it quietly encouraged the players to do anything they wanted to anyone they wanted. If they burned down a town for giggles, that’s fine! Nobody in later floors would know.

This became most clear in Diana’s floors because they were divorced not only from each other but from D&D. Alex’s floors felt like a campaign or video game where you killed the boss and moved to the next area, but vanilla D&D is kind of like that anyway. Diana didn’t even have that structure because she told the players “this is a puzzle, and when you solve it you advance”. Her area of the tower was even less real than Alex’s and it showed even fewer long-term consequences because it lacked even the same threat of combat death.

The players pointed this out to me somewhere in Act 2 and early Act 3, and once I looked at it I realized they were right. So for the sixth section of the tower I wanted floors that connected to each other even if it wasn’t obvious at first. And I needed a tower guardian who made sense for this concept and represented an escalation from Diana as we approached the end of the campaign.

Sennacherib was probably the longest-reigning tower guardian. He’d been around to see what hundreds if not thousands of parties had done with and to the tower and the NPCs that lived in it. By the time the party made it to him he had long ago become sick of the nonchalance with which climbers fought, threatened, insulted, demeaned, and murdered people when they thought there would be no repercussions. So he inflicted consequence on the players.

All of Senna’s floors were set in the same area at different points in time, a fact that players didn’t learn until they had already beaten the first floor. His section of the tower was a branching timeline where a party’s actions in Floor 23 would create a Floor 24 showing the world some hundreds of years later. If the party set themselves up as heroes, they might still be revered in the future and asked to save the town from demons. But if they murdered everybody they saw, they would be history’s greatest villains and hunted as soon as they arrived in the bleak ruins of the future.

To my players’ credit they did not kill everyone. But they did try to kill the entire ruling family, so let’s not give them too much credit.

But that’s where the puzzle got interesting. The players could very well have opted to burn down the entire city in Floor 23, which would have triggered a level-up because it sufficiently affected the future. If they had, Floor 24 would have been a barren wasteland with no buildings, creatures, or activity of any sort. I trusted the party not to do this because they’re good players, but the possibility existed (not the first time I put an instant-win or instant-lose opportunity into the campaign that I trusted the party to either not recognize or consciously reject). Because of the way the floors branched it was unlikely that another party would ever make it to the same version of Floor 24, so anybody who made it there would either starve to death, die of old age, or spent eternity trapped in a prison they accidentally built. Diana had cute little puzzle floors with rules and an outcome, like Professor Layton. Senna was more like a Sierra game, where a choice in the first room could lead to guaranteed death at the very end and there was no way to know until you got there.

The floors themselves were a D&D-based, pseudo-, I’ve-never-actually-been-so-please-don’t-hit-me version of Tokyo. Floor 23 was the city in the feudal era, Floor 24 was modern, and Floor 25 was an unspecified time in the future. Finally I got to use all those ninja minis, but more than that we broke with the D&D tech level and introduced guns and cars, then laser guns and flying cars. At one point the players fought a pitched battle against (and eventually with, because my players are like that) a platoon of tanks in an intersection in the middle of the day. It’s the sort of combat you just don’t expect to have when you’re used to a campaign where you play goatball.

Because of this Senna set himself up in a weird way to the players. He was happy to give them ways to show off, like Jay, but he also was happy to test their mental mettle, like Diana. What made him different is that he made everything into a consequence. Jay and Diana didn’t care what the players did as long as it fit some specific idea of success. Senna’s vision of success was a lot more fluid and murky, and it’s more accurate to say he had a lot of visions of failure and a few gaps in between*. This gave him a different relationship with the players and the campaign. With the other guardians, the players either liked or didn’t like them. Approval wasn’t a factor with Senna. I think he was the first guardian that actually scared them, and that speaks volumes about his DMing style.

In terms of style Senna was somewhere between a simulationist and a narrativist. He felt that the world should progress without the players and that anything they do should have consequences like it would in a non-tower environment. But his world wasn’t the star of his story, it was the hero, and the villains were the PCs. He wanted them to look at themselves, think about what they’d done and what they intended to do, and change for the better. His floors had the first real party conflict, with players drawing swords against each other and eventually going through redemption arcs. He forced them to react rather than act, putting them on their back foot for the better part of his section, and he gave them (relatively) long-term villains who went through character growth that led to their own destruction.

Which was sort of what his section of the tower was about. The capstone for Senna’s section was Floor 26, a literal junkyard no matter what the players had done previously. At his core Senna felt that players by their nature cause ruin and decay. It doesn’t matter if they think they’re heroes or world-savers because they function by death, destruction, trickery, theft, and every other action that deserved to be punished. And he made advancement from Floor 26 the ultimate representation of that goal: the only way to advance was PvP. If a party wasn’t willing to turn on each other or kill another party they couldn’t advance, and if they were willing they faced Senna and his hanging tree in combat. Because regardless of what the PC’s intentions were, they had killed a real person with hopes and dreams, perhaps even a friend of theirs, and that merited a judge and an executioner.

The players rightly viewed this as the ultimate kick-the-dog moment, and they didn’t take kindly to being forced to kill another climber and then being punished for it. In the same way that players don’t appreciate being forced into a corner and punished for getting out of it, the party didn’t appreciate that the only solution to Floor 26 was the only thing they had managed to avoid for the entire campaign, and in fact the things that differentiated them from many of the campaign villains. But more than just giving the party somebody to punch, Senna gave them a transition to the final section of the tower, where the players and the characters got their chance to prove whether or not they were good at heart. But that’s another post.

* — I’m willing to bet that a plurality of my players will say this is closest to my style, and that of course I would put myself into the campaign as the grumpy old man who’s resigned himself to disappointment. I’m more inclined to think that I have elements of all seven guardians, or more accurately they all have elements of me.

I said that I intended the Great Tower of Oldechi as a three-act campaign, and that was true even with the Gaiden adding an extra thirteen or so sessions into the middle of it. My original intent was to have the acts transition midway through Haelyn and again midway through the fifth tower guardian. But with the Gaiden lengthening Jay’s section of the tower I could make floors 15 through 18 be Act 2. Now the party knew that the tower was in trouble, they knew what was at stake if the maybe-enemies-maybe-allies group continued unabated, and they were ready to do something about it.

So while I originally had an idea for the fifth tower guardian, pushing forward the act transition let me turn the character up to eleven because I didn’t need them to execute a shift, just be emblematic of it. I was already planning on a guardian who allowed as much player agency as Jay but did it differently, and somebody who was a lot less glib about it along the way. The campaign was serious and difficult, so the story needed a guardian whose floors were serious and difficult.

Diana was the opposite of Jay in a numbers of ways. She found humor and lies frivolous, and she didn’t believe in advancing via awesomeness because it was too subjective. Her floors had singular, objective measures for success. The method by which that success would be obtained, however, varied significantly within the boundaries of rules that the party discovered along the way. Combat as a solution was almost universally unacceptable, and her floors probably had the fewest combats, including a couple of sessions (in D&D 4E!) with no fights or threat of fights at all.

This is because Diana’s floors were all puzzle floors. The environment only existed in so far as it set up the mental challenge that the players, not the characters, had to solve. Basically Diana’s theory was “If you solve all your problems by violence, you don’t deserve to go to epic tier play.”

The main problem with this is that D&D, especially 4E, is a violence-based system. The fifth section of the tower worked mostly by being things that weren’t D&D. One floor had the players interviewing people, putting together a schedule of events, and trying to find a killer, so it was more Phoenix Wright. Another floor was set inside a body dying from a sickness, so it was more Trauma Center. A third floor was explicitly and unabashedly based on one of my favorite flash games, Archipelago, but with less swimming and inventory management and more looming specters of world-devouring leviathans.

I suspect the players really liked Diana’s floors. I suspect this mostly because near the end of the campaign I asked them which floor they liked the most and every player chose one of Diana’s floors. In fact, among the five players all four of her floors were represented, so it seemed they were pretty universally beloved.

But speaking of the players, we lost three and gained four between the beginning of Jay’s floors and the beginning of Diana’s floors. By the time floor 19 got rolling the party was:

• Tela, goliath warden, as before.
• Lao!ze, thri-kreen ranger, a creature born in the tower but to a retired climber, so he retained the ability to advance in floors. His was a fairly interesting mystery because he had no pre-tower life to discover, and the question of what would happen to him was an occasional recurring issue. (For those of you playing at home, Lao, Laotzu, and Lao!ze are all from the same player. See if you can figure out his naming scheme.)
• Cid Viscous, slime man reskinned from a shardmind druid. His story is interesting enough to save for a podcast, but suffice it to say that this is one of the characters I reference whenever I talk about how powerful reskinning is.
• Rousseau, dragonborn fighter/cleric. A devotee of the god of tyranny, Rousseau did everything in his power to be nothing like the philosopher who was his namesake. But he did flex his way out of an obdurium cage, so that’s pretty serious.
• Miesha, succubus stripper reskinned from a dragonborn sorcerer/warlord. This is the kind of reskin where I would normally choke out a player for trying it, but Miesha’s player was not only female but also, like, three feet taller than I am and trained in hand-to-hand combat.
• Plague, changeling warlock pretending to be a tiefling. He had a tendency to debuff the party accidentally and damage the party on purpose. Normally this would not endear him to other players but on the first floor with Plague he sacrificed himself in a nuclear explosion to destroy thousands of monsters simultaneously. He got better, but that kind of gave him a pass on “I deal 1d10 damage to you sometimes”.

With one exception this party made it to the end of the campaign, and they worked startlingly well with each other considering their disparate backgrounds and playstyles. By this point the campaign design was much more freeform, with heavy reskinning of monsters and powers, weird environments, unclear and mutable goals, and playing fast-and-loose with the conventions of 4E in general. This party was really good at saying “I do X” and trusting that we would come up with some sort of rule or check to adjudicate it, and they spent as much time subverting the tower as it tried to spend subverting them. The Act 3 party was nothing like the Act 1 party, but it was perfectly suited to the situation and the opponent.

It’s worth noting that the payers seemed to be much happier near then end of the campaign than they were at the beginning. We can chalk this up to the players who rotated until we found an amazing group, but this is a blog and post series about DMing, so I think we need to focus on the change to this more freeform style of design and play. It’s as though players like a campaign best when they get to decide their own way. Imagine.

In fact, at no point did the players seem to resent or disagree with the “there’s only one solution” approach Diana used because the method by which they could approach that solution was variable. They could use any tactics they wanted to treat the disease eating away at the body, navigate the library maze, or interview suspects during the murder mystery (yes, even violence, though the players determined that beating up one member of a close-knit family was not going to aid them in interviewing the rest). They could also gather information however they wanted—at one point a player tapped directly into the body’s nervous system to try to communicate with the brain. The campaign had transitioned from “this is awesome, so you level” to “this is awesome and makes sense, so you make progress toward your goal” even if that goal was set in stone. The players could make their own decisions, the DM got to design around known quantities, the world could move at its own pace, and the action-success-reward pacing of D&D remained intact. For Diana’s floors, at least, everybody was happy.

The biggest takeaway I got from this section of the tower and perhaps the campaign as a whole, besides that players like puzzles, is that I tend to best put together things players like when I have a specific end goal but provide some variance in getting there. We reinforced this with Wrath of the Cosmic Accountant, where I had no specific end goal and the campaign was terrible, and the Eight Arms and the Empire of Sin, where I followed Diana’s formula and the campaign was great even if I did kill nearly everybody. A good campaign is lot like a puzzle in that you start by explaining the rules and restrictions then pace feedback and rewards based on experimentation and success. Perhaps it was just a confluence of play style, players, characters, system, tides, blood sugar, and any number of other things, but I’ve been trying to duplicate that magic ever since.

Diana did end up serving as a transition, but to a much harder style of puzzle in the next section, the type that doesn’t explain the rules at all while still letting you get the game into an unwinnable state. That was when some cracks in the party started to show. But that’s another post.

So here’s the thing.

The summer between floors 16 and 17 was rough for the campaign. I lost players to the summer break, more than I was expecting. I had a falling-out with another player. Of the players who were still there, all but one expressed an interest in changing their characters. In a campaign based entirely around “the party is a unified group and cannot split up”, the players themselves split up instead.

At a loss, I decided to put together a gaiden. The campaign proper went on hiatus for three months while we waited for things to reassemble and to give me some time to sort out what was happening. In-story, right before the summer break Tela’s player missed the only session she missed that entire campaign. As a reward for her attendance and dedication, in the middle of that session a hole opened in reality and stole her soul. The players carried her lifeless body for the rest of the floor, and once they finished floor 16 they met with Jay to ask what was happening.

He told the party that Tela was a victim of the tower’s increasing instability, caused by that maybe-allies-maybe-enemies group I mentioned last post. Her soul still existed but it existed in the Space Between Floors, the parts of the tower that climbers never saw or, ideally, knew existed. Creatures bitter at being abandoned by their creators had stolen the soul and split it into six pieces hidden in the detritus of forgotten, broken, and incomplete floors that had never been destroyed by the guardians.

Luckily, the guardians were prepared for something like this. Occasionally climbers would express an interest not in completing the tower or living there but joining it with the eventual goal of becoming a guardian themselves. These former climbers were able to navigate the Space Between Floors, and Jay tasked them with finding Tela’s soul. The problem was their time limit; if they didn’t reassemble the whole thing in one week parts of it would be lost forever.

And so we started a side story following a new party as they tried to find and destroy the offending creatures. They included:

• Varon, half-elf bard, who you may remember as a ranger. Arcane Power had come out since the campaign started, and it included a Charisma-based archer. This was a redesigned version of Varon, changed from a striker to a leader, and it explained what had happened to him once he left the party.
• Bjarn, dwarf fighter/runepriest, who spent most of his time complaining about how the runepriest was terrible. This was the same player as Borris, continuing his tradition of dwarves with startlingly high-damage builds whose names begin with B.
• Midnorton Jones, wilden monk, an archaeologist who loved the forgotten bits of the tower. His eventual goal was to discover its origin.
• Thor, dwarf invoker, who joined midway through the summer. I don’t remember too much about this character, mostly because I was in another campaign with the same player also playing an invoker. That character was interesting enough that it largely wiped this one from my memory. I do remember his “reroll any attack that targets Will and misses, but if the reroll also misses I’m stunned” ability, though. He missed a lot.
• Siven, rogue, who wore a mask and only joined the party for a few sessions. I recall him spending a lot of time pointedly not interacting with the party. Maybe he stabbed them once? It’s vague.

The neat thing about the Space Between Floors was that it was made of scrapped floors, so I made it of scrapped floors. I used it as a way to throw things at the players that I had considered for the real campaign but discarded for one reason or another, usually because I couldn’t justify making an entire floor out of them. So the players were literally going through a set of levels that the tower designer had created but was not using. It was a pretty meta adventure.

The Space Between Floors was set up like a fragmented computer, with various bits of floors tossed around, often abutting each other in weird ways. The players started in a prehistoric land fighting dinosaurs (scrapped because I had insufficient minis to make it as exciting as I wanted), then went to Ravenloft (scrapped because it wasn’t interesting on its own given Haelyn’s entire section), and eventually went to one-fight-large floors based on concepts that didn’t make for interesting enough plots (sewers, gravity, chaos, music, etc.). Along the way they found that Tela’s soul was held by six pumpkin-men (because I like pumpkin-men for reasons that probably have something to do with The Nightmare before Christmas, but I couldn’t pack six of them into a single floor), one of whom they actually befriended in what I guess I should now interpret as foreshadowing for all later campaigns.

Given the computer milieu, it’s unsurprising that we played with the tech level of D&D. The players got lightcycles from Tron when they started, which were very fast but completely unable to handle anything but the most gradual slopes. Eventually they made it into a Spelljammer spaceship, sabotaged its wireframe control crystal, and fought the MCP. The final boss was Abyss from Marvel vs. Capcom 2 (because he was technically made of Tron lines), for which our local game store owner made some fantastic custom miniatures.

(During this time it became clear to my wife that I had never seen Tron. She remedied that, then took me to see the sequel coming out later that year. So at least now I have an excuse.)

Since there wasn’t a tower guardian to act as a DM this section of the tower doesn’t have a lot of unifying themes. If there was anything I learned from it, it was “don’t throw out anything”, which I already knew. I do think that the pumpkin-men were a discrete leap in my understanding of 4E monsters, because they were the first time I really designed weird monster powers from scratch to fit a narrative goal rather than using powers I knew were reasonable and balanced given the party’s level and abilities. The result was raucously successful, enough that my monster design escalated almost immediately. It’s safe to say the warden never would have had his aura if I hadn’t changed how I designed monsters, and that’s a story too good to not exist.

Eventually the gaiden party succeeded, reassembled Tela’s soul, and rode off into the sunset. Tela woke up after a week asleep to find a note pinned to her armor that said “Got bored waiting, sorry!”. The tower assigned her a brand-new party, and together they completed Jay’s section and moved on to a tower guardian who was less inclined to help them. But that’s another post.

I kind of assume everybody is familiar with some version of the Rule of Three because it’s so ubiquitous, but in case you’re not familiar with it, in general it means that audiences like it when things come in threes. Movies come in trilogies, books have three-act structures, jokes and fairy tales repeat lines and setups three times, and so on. There are enough examples online that I won’t bore you by repeating them.

I lean on this every once in a while in campaigns, and I try to split campaigns into three acts: understanding the world and the central conflict, escalation of the conflict through the actions of the antagonist, reaction from and resolution by the players. Given the nature of a medium where the designer is only responsible for some of the storyline the length and content of a given arc is subject to change. The Monster Campaign followed the rule well, with enemies that were at their strongest in Disc 2 but passive for most of Disc 3. The Eight Arms and the Empire Sin sin was also very good, as the players chased an unrelated plot in Act 1, dealt with immediate problems in Act 2, and spent Act 3 chasing the bad guy and invading his home. The Eight Arms and the Memento Mori wasn’t quite as perfect because the third act ended up heavily rushed, but it was there.

Given this eye on the Great Tower of Oldechi, The first fourteen floors were Act 1. They got the players together and functioning as a unit, introduced many of the major players, and gave the party a feel for how the tower itself worked. Arithmetically-inclined readers may note that this is half of the way to Level 30, so Act 1 was half the campaign. Well, sort of. I expected the early levels to be simple while the later levels were complex, so I designed for each floor in heroic tier to take two sessions, paragon three, and epic four. When all was said and done the party beat Haelyn in session thirty-four of a 108-session campaign, so we were right on.

For the fourth section of the tower I wanted the conflict to ramp up. I wanted to force the players’ hands and have them decide whether to ally with or fight against a looming organization that may or may not have been evil, either setting them up as campaign villains or the party’s only lifeline. I wanted the floors to get to the point where players really felt like they could die, which in this campaign was a permanent death and a new character. And I wanted a tower guardian who moved the campaign at a right angle to where the players thought it had been heading, who knew he was in charge and leveraged it instead of letting them go about their business, and who could in the right light be seen as a legitimate threat instead of just a rough-around-the-edges moderator.

Jay made it clear from minute one that he was messing with the party. One of the first things he said was that “Jay” wasn’t his real name to get across that his reality was exactly as he intended it. He also actively trolled the players, some more than others, with lies and jokes. He was an active participant in his floors, not taking up arms as such against the party but certainly willing to step in to make their lives a little more interesting.

But what really set Jay apart from his predecessors was his advancement mechanics. Alex said, “Beat this guy and you level”. Rody said, “Beat one of these guys and you level”. Haelyn said, “Find the guy to beat, then beat him, and you level”. But this was all too concrete for Jay. Jay said, “Impress me.”

He went on to explain that sometimes all it takes to impress him was finding a big guy and killing him. But that was because Jay cheated wholeheartedly. The most powerful guy on each floor was ten levels above the party; killing something like that was legitimately impressive and merited a reward. But killing a guy who merely looked powerful was not sufficient. Nor was steamrolling a bunch of enemies with the raw power of an excellent character build. Jay didn’t want to see the party strong-arm an equivalent-level solo, he wanted to see something that made him sit up and take notice. And except for “feel free to get yourselves killed fighting a demigod or something” he provided no further direction.

For the first time the players were on their own. They had to explore the floor, think of something to do, try it, and look at the sky expectantly to see if that was enough to make Jay happy. Often it wasn’t. Jay wasn’t treating the tower like a gamist system any more. He was more narrativist: “Get yourself in a ton of trouble, find a way to get yourself out, and do it in an emotionally satisfying way”. Combat still happened on his floors but was no longer encouraged or required. When the party did kill a scary guy, it was only after that guy harassed them for the better part of three sessions by virtue of being largely unstoppable. Instead the party looked to other options, like stealing artifacts, ending ancient rituals, and collapsing entire societies.

The floors themselves weren’t that impressive. Jay based them on the four classical elements, so the party dealt with lava floes, an ancient jungle, incredibly windy cliffs, and undersea ruins. But the story wasn’t about the floors. Jay didn’t put them out and say “and now you will love them”, he put them out and said “find a way to wreck these for me.” In Act 2 the campaign started being more about the players and less about the world they inhabited.

Speaking of the players, there was a fairly radical shift in party makeup midway through Jay’s floors. I’ll mention some of the changes in a future post, but there’s one new character I won’t get another chance to mention:

• Cletus Hightower, redneck ranger, and his animal companion The President, squig, which is a pretty weird confluence of words for a character in what until that point was mostly a high-fantasy campaign. Cletus was only with the party for two floors, so there’s not a lot I can say about him.

I think the characters don’t have many good memroies about Jay, but the players really liked him. He kept things light, he gave them some agency to decide their own path, he hit them with some hard and some fun challenges, and he seemed more like a regular person than a standoffish arbiter. Running Jay reinforced a few things I wanted to confirm about DMing: players like present opponents more than absent ones, good worlds can stand up to a little meddling without falling apart, and players are happiest when they can make their own way, especially when that way allows them to do things they think the DM hadn’t planned for. It made for a good transition between the serviceable but uninspiring tower guardians of the Act 1 and the challenge-the-player, more freeform floors of Act 3 where the guardians would start really coming down on the players for the first time.

Of course, a lot of Act 2 only worked as Act 2 because of a side-arc that happened between floors 16 and 17. This not only lengthened the campaign but also explained some of the mysteries of the tower itself, letting the campaign focus more on the people in it than its own ontological mystery. But that’s another post.

Alex and Rody were part of a common but not mandatory progression in DMing style. Alex was mundane, by-the-book, and serviceable but ultimately forgettable. Rody pushed boundaries and tested the system, but made mistakes as he defined his style. Neither was a bad DM, and both needed to work a bit harder and grow into something else.

The rest of the tower guardians do not fit as part of an obvious flowchart of style. I think all of them are recognizable, but there’s no guarantee that you’ve played with, been, or even met a DM that fits the remaining styles. I wasn’t about to make a global statement like “these are the seven stages of DMs”, not least because I can’t come up with styles that would fit that tree. I also didn’t want to tie the remaining floors to some indefensible notion of “these are the kinds of levels you create as you get better and better at DMing”. That’s not what the campaign was about. Though the tower guardians were a big part of the campaign the story was ultimately about the players and, as they would eventually learn, a series of tests they passed on the way to deification. It would have been a disservice to them and their narrative to restrict their adventures so I could make a statement about good DMing.

So when it came time for to to create the third tower guardian I didn’t sit and think “what would an evolved form of Rody look like?” I thought “what would be a neat antagonist, and what would be a neat series of floors, and how can I marry those into something appropriate for low paragon tier?”

Haelyn could pithily be described as “Wednesday Addams with magic powers.” Her goal was to instill emotion in the players, and the emotion she chose was dread. Her floors were desolate, dark, and foreboding, full of threats lurking just beyond the corners of the players’ eyes. Though still gave them their share of combat and triumph she really wanted them to get the sense that things were bad and were only getting worse, that they were in constant danger regardless of their actions or successes, and that the only reward they deserved was a slow, maddening fall into a mass of despair.

So it was a lot like a story game, really.

I partly chose Haelyn because I wanted somebody to contrast with the party. They were a fairly comedic and lovable group of scamps: the handsome and charming but self-important bard, the highly capable and startlingly tough sorcerer who “accidentally” caught his allies in his blast radii, the brilliant and powerful but single-minded wizard, and the unkillable but exasperated warden who tried keep them all safe from themselves and each other. The astute reader will notice a party change, as the ever-shifting dwarf left the party and was replaced by a new character from the same player:

• Thump, goliath barbarian, whose love for the world and its meats was matched only by his hate for bees. I don’t think I can describe this character any better than his player already did.

You may notice that this is not a more serious character than the dwarf.

Haelyn’s style was meant to put the party out of their element and force them to deal with a world that was more actively hostile than the previous floors. She didn’t waste any time either; as soon as they stepped onto floor 11 and saw it was a graveyard reaching to every horizon I think they understood what that section of the tower was about. The following floors were an abandoned pseudo-futuristic mining base filled with scattered machinery and mindless robots, a mazelike set of caverns on the brink of inevitable war, and a lifeless expanse of ice. All were designed to rob the party of potential allies and show them that not only were they alone but the land itself was gradually growing more dangerous with each floor.

To the party’s credit they treated each new floor with the deference the environment initially deserved (and no more). They stayed light-hearted and humorous throughout but saved the true wackiness for downtime. Any time I gave them a real threat like a high-level monster or angry terrain, they treated it like a real threat. Basically they kept the goofiness on hold for times when they could get away with it, because they were starting to see that they couldn’t blithely romp through the world like they had before.

The problem was that they had to create their own opportunities for comedy because the floors provided none. Haelyn knew how to increase dread but didn’t know when to hold back. Her pacing didn’t have the ebb and flow that it should, where the players stop to breathe only to have things escalate as soon as they think they’re safe. Because the players were never safe, they never found allies, they never caught a break unless they grabbed the world by the throat and shook it until it left them alone for five minutes (but never rewarded them for it). It didn’t build anticipation and pressure as much as it drained the players.

The only person on the players’ side was actually Haelyn herself. She would heal the party between floors, give them hints about where to go, and be generally supportive. The players didn’t seem to know whether it was out of genuine concern or as part of some elaborate plot to get them to lower their guard. Regardless it made the pacing even weirder.

After floor 14 Haelyn didn’t actually fight the players directly. She knew that she wasn’t the scariest part of her floors and any direct confrontation would remove any lingering dread because it meant any horror could be destroyed by hit point damage. I should have sent the players into some terrifying set piece to tear them apart until they barely clawed themselves free in the end. Instead I just had her throw summoned creatures at them, one a round, so they could see the escalating difficulty but knew they had the power to stop it. It worked on paper and the players did like stabbing things, but looking back I feel it was a missed opportunity.

Haelyn is a good example of why trying to run a horror campaign, or any adventure based on fear, is hard. Yes, the DM has to been good enough to manipulate emotion with a delicate enough touch that it doesn’t look heavy-handed. That’s a skill worth learning regardless of the system or even the emotion in question, and I’m not convinced I had it. But it works best when the rules work with the intent, and 4E’s combat-happy mechanics and “all problems can be solved by daily powers and trained skills” feel set the DC for running a horror adventure very high. And even with a good DM and an appropriate system, it’s all for naught if the players aren’t bought in. Horror campaigns require a certain restriction on player agency that some parties can’t or won’t accept. No matter how scary and unbeatable you make the monster it loses some mystique when the players start calling it “clownshoes” and critting it. At least movies only have one funny guy they get to kill in the second act, not five funny guys who make it to the end through the power of flexing.

The next tower guardian instead took the players’ style and ran with it, giving them incredible freedom but requiring them to continually top their own successes to advance. But that’s another post.

By level 6 the players were starting to get into a groove, both in party style and in how they worked in the campaign. They’d paid their dues in the early levels, fought some goblins and hobgoblins and such, and beaten their first tower guardian. They were ready for something more exciting than forests and hills. But they were still heroic tier, not high enough level to get to the really fun parts. They needed to go through places and challenges more exciting than starting quests but not so extreme that they couldn’t handle it. In a sense, they needed to find their limitations and how to get past them so they could move on. It was only fitting that their tower guardian was the same.

To be as pithy as possible, Rody was Alex with two campaigns’ more experience and three more rulebooks. While Alex was limited to the rules as written, Rody was working on finding his own voice separate from the vanilla DM the books described. He was very willing and at least partially ready to explore things that the rules didn’t cover to see how he, the players, and the system could handle it. At the same time he held some disdain for the mundane environments of the lower floors and for the party itself.

Rody fell perfectly into the escalation of the tower floors. He still used mundane environments, but he tweaked them in some way to make them a little more fantastic. He sent the players into caves, but with a whole ecosystem that threatened the party instead of a single boss leading a group of humanoids. He gave them a swamp, but one made of poison. He gave them ruins, but with competing players stalking and attacking them more than hostile monsters. In each case he kept the framework of ordinary D&D but modified it a bit to work with his own voice and with the player’s desire for more interesting challenges.

The hardest part of running Rody was figure out how to get some things right and some things wrong intentionally. I wanted Rody to look like somebody who knew what he was doing, but not entirely. Luckily I didn’t really know what I was doing either, so it worked out, though I certainly made mistakes I hadn’t intended.

Some things that went well deserved it. Floor 6 had four groups warring for dominance over the caves, and the players could choose any one of them to fight. Each of them was a sufficient challenge for the players to advance, so they got to measure the threats and pick the one they thought they were best able to stop. Instead of a simple “there are monsters, kill they” plot there were elements of fighting, stealth, and potentially diplomacy in each option. Rody had progressed beyond a video-gaming-style idea of “challenge the players” and was giving them choice and more varied activities.

Some things that went well didn’t deserve it. On floor 10 the party met a group of goliaths, the same race as Rody. They challenged the party to a game of goatball, which Races of Stone tells me is a common goliath pastime. This was supposed to be a bit of racial arrogance on Rody’s part; the last allies the party would meet are goliaths, and to get the golaiths’ help they had to win at a goliath game, because of course goatball is an appropriate measure of heroism. But the party took to it like fish to water. Or, to be more exact, they took to cheating at it. The wizard managed to score at least two goals with a combination of mage hand and the referee’s terrible Insight checks, and there were many skill checks to advance the game in “ain’t no rule” ways. The players won easily, and not by following the rules of the game but by inflicting themselves upon it. We retold stories from that session for months.

Some things didn’t go well like hexes, which deserved success, and flying, which didn’t (though I suppose I’m not the best judge of which of my ideas deserved to succeed). Except for the area of blasts (which I think I understand now, partially, perhaps) 4E seems made for hex-based movement: heavy combat focus, no facing, non-Euclidian distances, probably other reasons. So I tried a combat using hexes. But in what I promise is an unusual display of terrible foresight I also put the players on flying vessels that had a maximum turning radius and minimum forward speed. In my mind it made sense because I cut my teeth on hexes with airplane-based board games. But to the players it was two new, weird mechanics at once, one of which limited their options in combat. I think since both happened at once the flying soured them on hexes. We never tried either again.

But that’s what Rody was about, trying things and seeing what happened. He was a DM with a few campaigns under his belt and he wanted to push the limits of the system to see what bent and what broke. And it helped a lot that I was doing the same. The session between goatball and hexes was the session where I first tested my system for skill challenges, finding the original system lacking. I was designing monsters with new, weird powers. I was giving the players combats made entirely of minions (fun fact, players can handle a lot more minions than the experience point budget thinks they can). I was finally comfortable enough with the game to give the players my favorite sort of puzzle, the type where I don’t know the solution but I expect them to figure something out while I adjudicate it.

At the same time Rody and I still fell into some of the traps of beginning DMs, mostly around “if I like X, my players will.” Mechanically he liked ranged attackers, traps, and damaging terrain (that is, terrain that damages players, not destructible environments). Besides the aforementioned love of goliaths he also liked humanoid enemies rather than monstrous creatures (another trait he shares with me). He liked self-important simultaneously-endearing-and-annoying enemies (a trait my players assure me he shares with me), like the shadar-kai who insisted his name was “Sir” and The Orb who spoke only in third person, who could be beaten with diplomacy. He preferred smaller environments to lands that could conceivably expand infinitely. His opportunity for sending giant constructs at the party was limited by the party level, but he still managed to hit them with one when the players fought him.

I don’t think the players ever picked up on Rody’s list of likes or dislikes. I guess that’s good in that they didn’t roll their eyes and say “oh, a gnoll archer behind a pit of acid, how surprising.” At the time they may have been feeling out their own characters as well, focusing more on how to work with each other and themselves more than the scenery of their story. So Rody was largely left to do whatever he wanted as long as it wasn’t disruptive to the game. It was probably for the best; when a DM is trying to find his voice, nothing shuts him down like the players telling him to shut up.

They did quickly figure out that he was their enemy and he knew it. He had been DMing long enough to know that the players were going to do everything they could to ruin the world he had made, but not long enough to plan for or encourage it. To him the floors were the hero and the players were the villains. They deserved whatever challenges he threw at them (hence a floor made mostly of poison) and if they couldn’t handle it that was their own fault. But he still knew enough to keep things fair and always give them a way out whether they took it or found their own. And, of course, his section of the floor ended with the players beating him and his giant construct in combat. A DM is allowed to be antagonistic if that’s what everybody wants, but he should expect to lose.

The party wasn’t quite as forgiving for the next DM. While they accepted her style, I think they grew weary of it over time despite her attempts to keep things fresh. But that’s another post.

There’s a lot I’ve thought about saying about 5E, mostly about things I don’t want. Things like the focus on Faerun as the default setting. Or using published adventures as a major, if not primary, part of the release cycle. Or the use of dragons as the be-all-to-end-all monster in a game that’s ostensibly about player characters. Or, heck, even the name. (Why not call it “The Dungeons and Dragons”? “2 Dungeons 2 Dragons?” “Live Dragon or Die Dungeon”? At least then the dragons get top billing.)

But there’s not a lot of point. 5E will be what it is whether I like it or not. It’s going to kill the company, save us all, or perform exactly as previous editions have no matter what a blog says and regardless of how much I play it. And, who knows, I might love it. I certainly warmed up to 4E once I figured out what it did, what it didn’t do, and how to accept and correct what I could. And there are parts of 5E I’m really excited about, parts that I hope make it into the final product (the basic rules don’t tell me a lot about monster design, and there’s no way I’m springing for the starter pack). It’s like judging a movie by the trailer; sometimes you get exactly what you expected, and sometimes you get The Mummy.

So instead of looking forward and shrugging, this might be a good time for looking backward and commenting, specifically about a campaign I ran some time ago. The Great Tower of Oldechi was an ontological mystery: the players knew they were in a tower and that they could advance in levels by performing tasks, but the exact nature of the tower and the way they advanced were questions they had to solve along the way. During their trip they learned more about their lives before entering the tower, explored a wide variety of worlds, and fought monsters, other explorers, and eventually the tower’s creator on the way to the top.

The campaign was somewhat special in that it was the first thing I ever did in 4E, and when I started I decided that it would be a full campaign from levels 1 to 30 corresponding to the thirty levels of the tower itself. It ran for 108 sessions over three years (you miss a lot of weeks in a college town) and ended with the players ascending to godhood so hard they were added to the pantheon in other campaigns I ran using other systems.

There are a lot of great stories I’m sure we’ll get into during a podcast, but that’s not what I want to talk about. The tower, and thus the campaign, was divided into seven sections, each controlled by a tower guardian. Each guardian had total control over his or her section of the tower, including deciding how the floors (demiplanes, really) looked, what monsters and allies lived there, what challenges the players might face, and the criteria by which they were judged. The seven guardians had seven distinct styles in how they designed their floors and how they treated both the players and the tower itself. Essentially, each tower guardian was a DM, and the PCs were players in a world they created.

I didn’t set out for the tower guardians to be different DM archetypes, but partway through the campaign I looked at what I was doing and worked with it from that point onward. In a sense it allowed me and the players to explore various styles while sticking with the same system, characters, and people, and it told me a lot about the way players react to DMs and vice versa. None of the guardians was objectively good or bad, but each had his or her points that fit or didn’t fit with the players specifically and the campaign at large.

So for this and the next few posts I’d like to look at each of the seven tower guardians to discuss their DMing styles and what that meant for the players in each section of the tower. In doing so I’ll go through the campaign retrospective that a single campaign writeup couldn’t have covered. That should cover us while I’m waiting for a 5E campaign to start.

Along the way, we’re going to play a game. The seven tower guardians share a unifying theme. When you figure it out, post in the comments that you did (but don’t say what it is, of course; that would be a spoiler). If you see the theme before the seventh guardian, congratulations! You managed to get it before every single player in the campaign.

The campaign started at the beginning in more ways than one. Of course the players were level 1 and it was the start of my first 4E campaign, but it was also the bottom of my normal escalation path. My recent campaigns tend to start with easy fights and challenges while the players get their bearings and feel out the world and their characters, and over time problems ramp up as the plot thickens. By the end of a campaign the players are barely hanging on, the villains are poised to win, and character death is unsurprising if not expected.

The Great Tower of Oldechi was no different. I knew that I wanted to start with simple floors and gradually get more and more complicated and unlike standard D&D. I had some great ideas for endgame floors, most of which I used. But to get there I first had to go through the simple, easy parts of the campaign so the later parts would seem that much better. There’s no point in hurling the players right into a demiplane set on the back of a city-sized, flying dragon at level 1. So I needed a tower guardian who was fine with relatively simple worlds and encounters.

Before I talk about the mundane guardian and his mundane world, I just want to talk about the party, which was anything but mundane. We started with:

• Varon, half-elf ranger, who was an incredibly curious case for 4E because his highest ability score was not his primary attack stat. Varon was a Charisma-based ranger, which is not a thing. It made him a decent party leader but not very good in combat, especially with his “lead from the rear” style.
• Rascon, tiefling bard, who managed to be everything Varon wanted to be but better because his class was actually build for it. Rascon was very nearly the “as likely to jump off a bridge as cross it” version of chaotic neutral but had some golden interactions with NPCs over the course of the campaign.
• Lao, bladeling wizard, another race/class combination you might recognize as not mathematically ideal. I don’t have a strong memory of Lao for reasons that will become clear shortly.
• Jaffar, elven avenger. Not even the player of this character remembers his name. We mostly remember the character because the player tended to choose powers he had a hard time saying. For that reason, to this day he remembers the name, text, and pronunciation of “angelic alacrity”. (Edit: turns out the character’s name was hidden on the campaign wiki, further proof that nobody reads those things.)
• Baerd, dwarven fighter, who was exactly what you would expect from a dwarven fighter.

…we think. It quickly became clear that the party wasn’t very fun for the players. 4E is not designed to handle suboptimal builds so half of the characters were meaningfully underpowered for such a low level, and before long we had a player unexpectedly leaving and another player unexpectedly joining. So on Floor 3 the party split up. One half of the party freed some captives, who happened to be characters designed by the members of the other half of the party. Some further shuffling happened between floors to work with everybody’s schedule. We ended up with this party, which we remember much better:

• Rascon, still.
• Laotzu, dragonborn sorcerer, who love spitting energy and damaging allies more than life itself. Laotzu was very effective in combat and very frustrating for me because of his high damage output and ridiculous defenses. But he is also the reason War Wizardry is the only feat we’ve banned in our local metagame; any mechanic that encourages a player to damage their allies, regardless of the scale of the damage or the benefit against enemies, is a recipe for trouble.
• Tela, goliath warden, who started as a mundane warden but slowly reskinned her powers over time to become more of a weather-focused defender. Unlike most characters she picked a deity available to PCs and stuck with it, and she ended up being the party’s most moral member by omission. Spoilers: Tela is the only character in this post who survived to the end of the campaign.
• Borris, dwarf chameleon. The thing I most remember about him is when his player explained that Borris had changed every floor. That is, more than 4E traditionally allowed. He had been creating a new Borris every level, complete with different classes, to see if the other players were paying attention. They weren’t, at least not enough to see through his heavy reskinning. This might have been a strategy I respected except I found out about it at the same time everyone else did.
• Hadarai, elven wizard. There’s a feat tree that allows wizards to hit a creature with cold damage to make them vulnerable to later cold damage, which put their damage at low levels nearly on par with strikers. Hadarai took that tree and made it into an art form. He had a tendency to solve all problems with the same at-will attack, even in skill challenges. “An air vent spouts poisonous gas!” “Arcana check to cover the vent with ice using ray of frost.” “The door is locked!” “Thievery check to disable the lock mechanism with ray of frost“. “The orcs are scary!” “Intimidate check to awe them with ray of frost.” At the time I was irritated. Now I’d probably allow it if all his other attacks did cold damage as well. Cryomancy is somewhat underrepresented in D&D.

This is the group that really started to play off each other, and they had some great stories. But we’re here to talk about their DM and villain, of sorts.

Alex was a fairly bookish human (and I mean that literally; he carried around a giant tome that he used as his primary method of attack). He interacted with the party between floors, during the downtime they used to gain levels, and no other time, but during those times he was fairly understanding and a little affable. His floors were incredibly boring compared to the later parts of the tower: plains, a forest, hills, a tundra, and a desert. On each floor the party had to kill some monster to progress; once they beat the boss, they could return to the beginning of the floor and enter a magical door that took them to the next level.

Again, I did not plan on having each tower guardian represent a DMing style when the campaign started. But looking back Alex is a perfect example of the beginning DM. His environments were the ones described in the books, his monsters almost all came from the monster manual, his puzzles were small and didn’t always mesh narratively, and he was friendly with the party but standoffish about the floors themselves. It’s the sort of DM a person thinks they should be if they’ve read the examples in the core books but not thought about the softer advice like “read the table” and “give the players control”.

This showed best in his mechanic for advancement. Each tower guardian decided whether a party merited admission to the next floor, and since they controlled the world there was no other option. Alex always had players advance by killing a boss, because that’s how D&D works: kill monster, get experience points, make numbers go up. Alex legitimately didn’t understand other methods of advancement or experience gain because the books didn’t explain them. He DMed with Intelligence but not the sort with carefully-balanced encounters and meticulous plots. He followed the manual for DMing in as much as it could teach him, and anything he couldn’t learn from reading wasn’t worth learning.

When I explained Alex to two of our local GMs, one responded “That sounds boring.” Before I could respond, the other said “Well, it was the beginning of the campaign. I think the players expected boring.” That’s a pretty good representation of how the players reacted to Alex. They tolerated him and his floors, but only because it was a way to get to later floors. Nobody loved the mundane levels, even if they did love some of the things they did there, and once they were past Alex they barely referred to him or his section of the tower again.

And that’s kind of how beginning DMs work, or at least the kind they teach in the core books. The rules can’t teach good DMing because that comes from other sources: experience, communication, theater, pacing, all bits of a whole that aren’t addressed by charts or text. The rules do teach what they can, but on their own they are insufficient for a good DM. The best they can do is create a tolerable DM who creates tolerable campaigns. Nobody watches the clock waiting for it to end, but nobody talks about it the next day. Alex was a larval DM who needed work, and his problem was that he had no interest in putting that work in.

When the players finally fought and beat him at his lighthouse, I don’t think they were glad that they were done with him per se. But they were so interested in what came next that his whole arc became an afterthought. The next DM was a little more evolved, and he played with the rules in ways both good and bad. But that’s another post.