Session zero 3: Case studies

Now that I’ve outlined the foundational taxonomies of fun, and gave some advice on how to discover them during session zero, let’s get to the fun part: judging people!

Well, okay, if you want to be polite we’d call them case studies on why this method solves a lot of problems that don’t seem to be resolved by traditional session zeros. This is not to say that a “traditional” session zero is incapable of achieving the same effect: a sufficiently competent GM can achieve the same result with any method, because for them these things will be instinctual. It’s just that with certain methods make some things more obvious than others, and as a result some methods are more prone to avoidable mistakes than others.

As usual, if you are confused about any terms, make sure to check out the glossary, and if you are looking for other posts, check out the blog map.

I’ll use the rpg horror stories subreddit as a source for most of the case studies: it is specifically dedicated to people posting about their horrible experiences with some gaming groups. I’ll go through the posts summarizing key parts: some of them can get rather long, so I won’t make you read the originals.

Without further ado, let’s get to it.

Case Study 1: Rogue GF.

Poster (person who made that post) is a GM in a game with all new players. Among the player cast are his girlfriend (GF) playing a Rogue, Bill playing a Wizard, Gajiro playing a Barbarian, and Marine playing a Fighter. They are all playing DnD 5th edition.

Now, it is important to point out that DnD is among the crunchiest PNP RPG systems out there - that is to say, it has a lot of rules that interact with one another in complex ways. 5th edition attempted to simplify the rules a whole lot, which (as far as I am aware) succeeded to some extent. But it is still quite complicated.

This has the consequence that for a new player, one who isn’t at all familiar with the rules, it is quite difficult to tell what parts of the system will serve their desired character feel and which ones will hinder them. As a result, it isn’t uncommon for players to want to do one thing, but end up building a character who is capable of something completely different. This happens because of legibility.

Legibility refers to how clear and understandable something is. For example, let’s take two classes in DnD 5e and consider how much damage they do. First we have Barbarian, whose class table looks like this:

Down the page we can see that Barbarian has a feature that lets them add some damage to their attacks in combat, as indicated on one of the columns in the class table. This addition starts at 2 and goes up to 4 at later levels. Now let’s look at the Rogue:

Wow, that’s a lot of dice highlighted by that arrow! What is that feature? A little down the page we get a description:

...Once per turn, you can deal .. extra .. damage to one creature you hit with an attack if you have advantage on the attack roll. The attack must use a finesse or a ranged weapon.

You don’t need advantage on the attack roll if another enemy of the target is within 5 feet of it, that enemy isn’t incapacitated, and you don’t have disadvantage on the attack roll.

At first glance, it seems that a rogue can deal way more damage than a barbarian. 10d6 (averages to 35) is way more than a measly 4, after all! 

And herein lies the legibility trap, because in actual games barbarians deal more damage than rogues. The problem is that rogue’s seemingly high damage is extremely legible: it is just a huge number. Meanwhile, the conditions when they can apply that damage are illegible: there is all that stuff about having advantage, which a new player can not understand at all. Even if they went and read the rules for what advantage is, how are they supposed to know how common are situations that grant this advantage in actual play? That would depend on dozens of factors, from types of monsters typically found in the monster manual, to other players, to interactions between different rules of the system.

In much the same way, the amount of damage barbarians deal is hidden within all the rules within their class, not just a single one. Another class feature lets them do two attacks at once: that would double their damage output right there, but would be invisible to someone who missed that feature when skimming the class description.

Why am I telling you this huge prelude? Because the poster in this case study writes:

Right off the bat ... she(GF) made comments [about] how she wasn't doing the biggest damage of the group. I simply told her that while she's a [level 1 Rogue] with only a dagger [she wasn’t] going to be doing really big damage yet. That, she was sorta fine with but when I then told her that with a Fighter, Barbarian and a Wizard beside you you'll probably be doing the [least] amount of damage out of the group [even at] later levels. Then she really wasn't pleased.

(square brackets and dots added where I edited the text to summarize or correct grammar)

What can we get from this one paragraph?

First of all, I hope that any reader of this blog would have lights flashing in their head: this player wants a power fantasy. They all but come out and say it. Though this is still only a suspicion, it is a very important one.

Second, we can tell that the GM knows that Rogues deal less damage than Barbarians, and already knew it at the time. This will become relevant later.

But let’s keep reading.

I asked her to give it a little bit as Rogues are really good at not getting hit and doing things outside of combat. She still wasn't happy but kept going. This continued for several more sessions…

What do we get from this? Very simple: this wasn’t a single isolated signal that you could miss, but multiple signals, over a prolonged period of time. After this point, GF asks the GM for some magic items, heavily implying they would help her catch up in damage potential. GM sends them on an adventure to get some.

They go over and notice a giant chess board [guarded by a wizard]. It was a .. Harry Potter style chess game where they were all pieces that could only move on their turn [and you had to attack pieces to take them out] Marine was excited, ... Gajiro didn't know chess but still tried his best. As for GF and Bill .., they didn't want any part of this. They saw this as a brain challenge and .. killed their piece early and [sat on their phones until the end]

Next piece of information: GF and Bill don’t like puzzles. Finally, they all get treasure:

Gajiro ... got the Horn of Valhalla ..., and Wave - [a sentient trident]... Marine .. got the Defender Sword and Black Razor - another sentient longsword... Bill .. got a Fireball necklace and [a rod that turns into a fire sword] ... GF got a cloak that .. let her cast disguise self .. 8 times a day and a lockpick [with a large bonus] that never broke .. and boots of spider climb. In my head I thought she could do more Rogue things and be sneaky and be a really good party player.

Remember why GM was doing this in the first place? One of their players complained about feeling they weren’t doing enough damage. How does being “sneaky” address a concern about damage?! This is an error that should be blatantly obvious to anyone who pays attention to what types of fun their players find engaging (or to, frankly, anyone with a pair of eyeballs). Sure enough, the consequences aren’t pretty:

She immediately made a comment of how [everyone else] got weapons and she didn't and that [her] items were useless. [She then] said [that] the Wizard's name was terrible. I already felt bad that they didn't like this encounter and now that she didn't like the items [it] made me feel worse.

Outside of the game she explains again how she thought it was unfair she wasn't doing the biggest damage and I tried to calm her down by telling her that now with these items she'll basically never get hit and [will] be able to sneak everywhere, [but] she wasn't having it.

At this point the player’s dissatisfaction grew to the point that they are lashing out at features of GM world, such as NPC names. Yet the GM continues to ignore the signals about what type of fun the player prefers. There is no excuse for this at this point: they should be absolutely impossible to miss. This pattern continues:

They make [it] to the end of HotDQ all still alive and GF commenting literally every single session [about the damage issue]. She would go out of her way to fight enemies herself to feel powerful, sleep with NPCs just to kill them in their sleep and split from the party to start combat. In this case I really did try my best to spoon feed her encounters that a sneaky Rogue would excel at but she didn't want any part in those as they weren't combat.

The post goes on in the same fashion, describing more times when GM thew “rogue-styled” encounters at GF, as well as puzzles (which, naturally, both GF and Bill hated). Her concern about damage was never addressed.

Finally, we come to the kicker: 

There was a session zero ... and she went through the classes and choose the Rogue herself

There was a session zero! This GM did it! Now, obviously I can’t peer into their soul and find out what in Hell’s name they discussed there. But whatever it is, it wasn’t what would have prevented this entire problem in the first place.

If this GM started out by asking themselves what types of fun their players enjoyed the most, this problem would never even come up. They would either realize that they should avoid puzzle encounters altogether, or find another group to play with puzzle encounters. Likewise, they would have instantly realized they should not let a player who knows nothing about the game system pick a class: how could they possibly do so? They should have asked them what feel they were going for from their character, and then actively suggested classes to them. Alternatively, if they let their players pick classes, they should have immediately seen the problem coming up: they knew rogue would be sup-par in damage, so if someone who wanted a power fantasy picked it, they would have told them it was a bad pick!

Now obviously not all problems with the game go this deep and get this bad: but I do not see why you should settle for having merely tolerable problems if you can have no problems at all. Use taxonomies of fun: they will help you!

Now let’s get to our second case study.

Case Study 2: Stepdad Stepping Out

This one isn’t a story - it is just a couple sentences, with a screenshot backing up the truth of it, but it illustrates a pretty common failure point of typical approaches to PNP RPGs.

Family DND night ends in shambles. My stepdad spent about a week making this world, only for it to be ruined by the murder hobo which is my brother and my (honestly didn't seem interested) mom. He later sent me this heart breaking msg.

Stepdad later sent a message saying he was done with DnD as a result of this. What mistake did he make? Very simple - he should have first found out what types of fun the players were interested in, and only then started worldbuilding. Of course trying to use a world that doesn’t fit the preferences of the players will end in disaster!

Now, it may be the case that the players you have access to don’t want to play in the world you want to design and run - their personalities will simply conflict with yours too much, in terms of the types of fun you find enjoyable. Too bad! Life isn’t fair. You can not lead a horse to water, and you can not force players to like your world. In that situation, you should either content yourself with running a game your players want to play, or try to find another group of players. If you aren’t willing (or able) to pick either of these options, you can choose to not play PNP RPGs until that changes, or risk your game ending in a lot of drama. There is no fifth alternative!

Case Study 3: Man’s Best Friend

This one is quite a tear-jerker. One of the players asked the GM if his character could have a dog - same one he had in previous campaigns with other GMs. GM agreed, but after a while, started to find the dog “overpowered”, and in the end set the story up so the player found their dog dead in a pretty disturbing scene.

The catch? The player originally named this dog after their dog in real life. Furthermore, their dog died one month after they started a campaign where they had a dog pet. They have been moving their pet from campaign to campaign as a memento of their pet in real life. And then they find their pet dead in a disturbing scene!

Now, there are several issues with what the GM did here, in terms of pure GMing strategy. First of all, it’s quite obvious that the player wanted to have a dog - removing it in any fashion would obviously go against one of their sources of fun, which is contrary to the point of the entire endeavor. Secondly, I am unconvinced that the dog really was “overpowered”: typically, in my experience, it is quite easy to modify encounters and stories to account for the level of power I’d expect a dog to have. It is not at all clear to me that the GM was genuinely having trouble with deriving fun from running encounters, or from writing plot, or that they were forced to spend a lot more time preparing than previously. Combined, these two factors raise an obvious question: why remove the dog at all?

But there is a much larger problem with GM strategy here. Let’s say that the dog really was “overpowered” and GM was having trouble writing interesting and challenging antagonists who wouldn’t be instantly discovered by the dog. Let’s say that something really has to be done about the dog specifically. But there are dozens of ways to “nerf” the dog: reducing stats, asking the player to leave it at the base, et cetera. It’s obvious that out of these ways, GM has to pick the one that maximizes fun for everyone (fun maximization being the fundamental point of this whole PNP RPG activity). GM here picked the single worst way to achieve this objective, judging by the reaction of the player.

And that’s the fundamental failure: lack of information! If the GM followed my advice and used previous campaigns of the player as a valuable source of information, they would have asked the player about them, and quickly found out the absolutely crucial history of the dog. If they would have tried to fill in a profile for the player, they would have hit the section for previous campaigns and key items, and would have likewise been led to asking the same questions.

I consider this failure to be much worse than the actual desire to kill off a player’s pet. Latter is also bad - obviously - but on it’s own, it’s a tolerable sin. On the other hand, not seeking out crucial information would be unacceptable even if GM harbored no malice towards the dog. What if the dog gets killed in combat? I imagine this would have been a lesser, but still traumatic experience for the player. 

Any GM who doesn’t actively try to find these crucial pieces of information about their players is like a careless fool walking on a minefield, not knowing where the mines are. An explosion is inevitable!

Case study 4: Bait And Witch

This story comes by way of another player in one of the games I play, whom we will call Wolfgang. Wolfgang’s GM tells him and some other players during session zero that he will be making a campaign focusing on social interaction and intrigue, and Wolfgang made a caster focused on non-combat spells, subterfuge and utility. Several days after, GM switches tracks: now he says that all magic is illegal in this world. During the first session police attempts to capture Wolfgang’s character, and the party is forced to flee the city into the wilderness, where Wolfgang’s subterfuge spells are largely useless. Campaign then turns into a dungeon crawler: a PNP RPG primarily focused on combat encounters and looting dungeons.

What is the problem at hand here? It should be quite obvious: if the GM wanted to run a dungeon crawler, why was he recruiting players for a game about social intrigue? That seems like the exact opposite of what you would want to do if you were focusing on your own fun and the fun of your players! That would be like trying to build a house by hiring starving literature students instead of construction workers.

On the other hand, if the GM didn’t want to run a dungeon crawler game, why did he put the party in a situation where this was the most likely outcome?

If the GM didn’t want magic, why would he let one of the players make a mage? If he did like magic, why the decision to make it illegal?

There is only one explanation that doesn’t make the GM look like a schizophrenic, blatantly acting against their own self-interests: GM simply didn’t know what they wanted, and only reflected on their own desires after the session zero. But that is exactly why figuring out your own preferences is the most fundamental and foundational part of GMing, that has to be done far in advance of even finding your players, let alone doing a session zero! Of course if you do not know what you want to do, then any action is unlikely to bring you closer to achieving it.

These case studies aren’t exhaustive by any means: there are hundreds of possible errors GMs can and do make over the course of both running games and doing session zeros. However, I think they demonstrate two crucial points. First of all, it is essential to keep the fun of you and your players in mind, and plan around various pitfalls which will reduce your fun. Secondly, typical approaches to doing session zeros are not sufficiently robust in terms of helping GMs avoid these pitfalls. Their limitations are predictable, and can be avoided with better methods, such as the ones described on this blog.

Next we will discuss some of the typical patterns you see in conflicts between different types of fun, and ways of dealing with them.

Meanwhile, if you enjoy what I write, you can subscribe to receive updates by email: