Saturday, January 30, 2021

Predictability and Opacity in Advanced Dungeons and Dragons




I came across this series of Tweets today from @rdonoghue and it got me thinking about game design. 


“Random aside:  One design ethic that i do not particularly enjoy in D&D is what i would describe as the "Gygaxian No" - that is, the players have legitimately earned certain abilities and spells, and adventure designers explicitly negate them for simplicity or effect

Two most common examples being higher level adventures that take away mobility effects (like flight), and effects which explicitly pierce immunities.

100% understand *why* adventures remove mobility, I just find it lazy and sloppy in almost every situation, especially because it's almost never "This is harder", but rather "This is FORBIDDEN”

And I just find the immunity thing a jerk move.  Player immunities are reasonably rare in D&D, and are often against rarely-encountered things like disease, and the result is often the only time disease shows up is when it ignores immunity, and that's just crappy.”


What I find fascinating is the idea that having abilities restricted could be seen as a case of flawed game design. And it is certainly seen that way here, it is not “just different”, it is lazy, sloppy and a “jerk move”. There was pretty much universal agreement about it in the comments as well. That surprised me, what is “jerky” about restricting PC abilities in certain cases?

So let’s take an example of one of these cases of bad game design, it’s always better with an example. Here is one from Descent into the Depths of the Earth, one of those high level AD&D modules referred to, and one that restricts movement, something mentioned in the Tweets. 


So note a few things. First off, it tells the DM not to penalize the players if they plan ahead for the difficult travel. And it then says to neither help nor harm them. I think this is an important point to bring up the trope of Gygaxian D&D being “unfair” or “punitive”, the DM’s job isn’t to be unfair and screw over the PCs, but to neither help them too much or harm them too much. 

But on to the “Gygaxian no”. It forbids teleportation. Not all teleportation, but teleportation beyond a half mile. It does not forbid flight, or even spells like Dimension Door, which have a reach of much less than a half mile anyway. PCs have a lot of ways to fly, and to move across on the ground but faster or more easily. The pre-gen characters provided have two ways to fly (an ebony fly and a potion of flying) and two ways to move across the ground faster (boots of striding and springing and boots of speed). They also have access to a stone of controlling earth elementals, which can have all sorts of interesting effects on travel.

I don’t see how this lies under either of the assumed reasons for “nerfing” things, e.g. “simplicity or effect”. Nerfing teleportation doesn’t simplify the module, it makes it more involved, as the party can’t skip the land based travel. Neither does it produce much of an effect. It nerfs one particular spell, possessed by one of the pregens. That’s it. 

Indeed, the very presence of pregenerated characters in the module suggest it was designed for use either in someone’s campaign or as a one shot adventure. So to suggest that it is unfair to players who have ‘earned’ their abilities, is a bit odd. For many this will be a tournament game or one shot adventure, so they haven’t “earned” anything. But even if it is for a long term game, nerfing a power in a limited way like this is not bad game design. 

The player is not reduced to that ability or spell power, certainly in a AD&D game where the PCs were high level (as they are in this case) they would have many more arrows in their quiver. Here, for example, are the spell lists and some of the magic items:


Also, the ability to teleport is removed, but so too is it removed for the locals. Given the environment (underdark, drow, kua-toa, etc.) one might be thankful at this restriction. 

The real issue here is not nerfing an ability, it’s the frequency of nerfing an ability. The baseline here is not “once I get an ability/spell it will always give me an advantage” it’s “once I get an ability/spell it will give me an advantage most of the time”. This is an important difference, as the latter implies some opacity in the tactical aspects of the game. 

I think the concern about nerfing in this case also lacks context, or perhaps is mired in the wrong context. Gygax designed the game to challenge the players, not the characters. So this meant putting them in challenging situations where they would have to be creative to overcome, not just applying abilities in a rote way. 

So why do these things happen so often, why does Gygax put so many nerfs in his modules? I don’t know if this is actually the case. I know there are modules where particular abilities are “nerfed”, but I don’t know if it was more common for Gygax to do it, and to what degree it is the case, for example, that disease immunities most often come up in modules where they are nerfed, as was claimed in the same thread. 

If it is the case that immunity exceptions are common, then this is where I agree, choosing to do this sometimes is good game design, choosing to do this a lot is not. But I don't recall Gygax suggesting to do this a lot, he does discuss some specific examples, e.g. invisibilty, but there is no blanket call to nerf powers regularly or to ensure that players abilities are commonly nerfed. It's always contextual.

This is a fine line to walk from a design perspective, because it straddles two issues, predictability and opacity.

Some degree of predictability is needed for a game, as otherwise the players can’t make informed decisions, and you rob them of agency in the game. However, some degree of opacity is also needed, otherwise the game becomes a rote exercise and the players get bored. 

You can get this opacity in a few ways, one is to randomize things, so you can’t know if any given instance of using an ability or spell will work, just the odds over time. Another way is to use counters, like the restriction in Descent, but forget the case of teleport, let’s try a low level spell. 

For example, there are spells like shield that counters magic missile, how often you give enemy magic-users this spell thus matters if one of the party members has magic missile. If shield is a common spell, then magic missile will be nerfed in many cases. But of course it makes sense that a spell like shield exists, when a low level spell like magic missile exists that is so efficient, a counter to it would likely be made.

So again, it isn’t the counter to the ability or spell that matters, it is the frequency. But even that is contextual. In the case of shield and magic-missile, since the DMG method for choosing spells has a 1 in 10 chance of getting either spell, 2 in 10 if you want the spell as one option is to pick. So shield will come up often as the game is designed, and it is relatively common, and BTB, for enemy spell casters to have good odds of having shield, a spell that nerfs magic missile. There are other examples in spells, magic items and monsters. Contextual nerfs are all over the place in AD&D. 

People assume these things are lazy, unimaginative and punitive as they don’t see their purpose in the game, and they don't understand the evolution of D&D from war games. A nerf is a tactical tool, it is no different than manouvering an enemy force into a place where they have some disadvantage, e.g. attackikng your enemy in a forest where there bows are less useful. Contextual nerfs are a big part of war games.

The purpose of counters like this is threefold, particularly in a long-playing game. They keep things from being too predictable, they give the players tactical/resource management choices, and they keep the challenge and the threat of lethality in the game. 

Playing AD&D (the game in question as Gygax was being criticised for this) long term means balancing predictability and opacity, both to preserve lethality and challenge in the game, to give the players interesting choices, and to keep the game from becoming boring. These are just as important goals as “allowing the players to enjoy their ability/spell” in a long term game.

If there are counters to your abilities and spells then you can't assume that they will be sufficient for victory, and you will play differently. You can assume they won't be there all the time, so your abilites are still useful, but they might be there in certain cases. This means the game stays challenging, and carries the threat of lethality. I hear people all the time talk about how flight or some particular spell "breaks the game", but if you can't guarantee that the advantages will always apply, you will end up using things differently. 

That’s another thing I don’t get about the critique. This is a problem for long term games as the PCs have earned their abilities over the campaign. But if it is long term, then any particular “nerf” is for one adventure or encounter, and since the PC will be on many adventures, it isn’t that big of a deal. The example of Descent is a good one to show this. The "nerf" on teleport in this module, if you were to apply it to a group using the pregens for a party, applies to one of the spells from one of the pregen characters of nine. That character's spells and magic items, just for reference:



So taking away teleport is a bummer, for sure, but its not like the PC is now dead in the water and unintersting to play. 

All of this is fascinating to me, given the historical discourse about D&D. So many people complain about the fact that PCs can become too powerful, but balk at situations when their immunities or powers don’t work. They also complain about the oversimplified ‘pass fail’ mechanics of the game, but balk when exceptions to a rule appear. 

They also assume playstyles, in a game where the primary goal is to challenge the players, removing situational bonuses/immunities on occasion in a case by case way is good game design. Not all the time of course, or it would get frustrating. But a quick look at a classic case from the big bad Gary Gygax shows that the nerf was very contextual, very narrow, and he encouraged the DM to allow the players to plan and get around the problems with ingenuity, to be fair, not help or unduly restrict them. It isn’t a “nerf a ton of abilities to crush the players” idea, it isn’t oversimplification or for “effect”, it’s designed to challenge the players.

In a game where the focus is on expressing and playing the character, removing situational bonuses/immunities seems unfair. That’s fine, if the goal is to maximize the player’s ability to play the character, to express them, to use their various abilities, then it would seem unfair to remove them. When your model is “give everyone a chance to shine” then taking away an ability would be taking the shine off. 

What I object to is the idea that what Gygax is doing is lazy, or boring, or jerk behaviour, e.g., bad design. When I read things like this I just think that it’s seeing things without their context in a mean spirited way. 

This difference in stance, between challenging the players and expressing the characters is starting to look like one of the major differences in playstyle in the hobby today. The important point is that neither approach is “lazy, sloppy or jerky”, they are simply different ways to approach the game and enjoy it.


Saturday, January 23, 2021

 The World in a Grain of Sand - Tables, Randomizationa and Generation in a City Setting





A few years back I decided to create a setting for my home D&D campaign. I wanted to create a fantasy city, so I went out and read a ton of published fantasy cities to see how they handled this. Most of them were like the old Judges Guild products, short entries for the majority of buildings in the city, usually accompanied by brief stats for the people within, and a number of buildings that were not given any sort of stats at all. 

This was a tempting approach, but my problem was that I was planning a BIG city environment, REALLY BIG. To “stat out” something that large was not going to happen, and would make the setting too unwieldy to use. I could partially stat out the setting, but that would leave hundreds of unkeyed entries. That’s not necessarily a problem, but what happens in play at the table when you have to provide stats for the unkeyed entries on the spot. I’ve done this before, but it can be a challenge.

So I fell back on what I know: tables and bounded randomness. 

If there was one thing I learned from 1e AD&D is that curated tables can be used to create a world relatively simply. So the idea was to create tables for each ward in the city, and use these, rather than explicit descriptions, to populate things. The “insight”, such as it was, was that AS THE PLAYERS EXPLORED THE SETTING, THE LOCATIONS WOULD BE SPECIFIED.

I have come to think of this as “Schrodinger’s Setting”, e.g. everything is in flux until it is interacted with, then the waveform collapses and you get an entry. It give the players a real role in quite literally CREATING THE WORLD THEY PLAY IN, and it gave me two important gifts. First, I didn’t have to stat everything, so I could create a huge city setting that wasn’t gigantic, and two, I would have a system for providing stats for everything, rather than just leaving things blank and filling them in on the spot. 

With this in mind, I thought it would be a good idea to give an example of one Ward from the setting to show how it works. 





13. Clothing Ward – Crest: Yellow knife on a blue background
House Kesht - Magus Warlock Nasim Tellbinder (H)*  – MU14/Il12 – Ur Lord Omari Ekon (S)  – F12/C10
PP: 1 in 6, FC: 4 in 6
Establishments
1-2: Dyers
3-4: Bootmaker
5-6: Tavern
7-8: Private Residence
9-10: Seamstress/tailor (exotic materials)
11-12: Clothworkers/ Clothier
13-14: Cordwainers (workers in fine leather)
15-16: Jeweler
17-18: Furrier
19-20: Restaurant
21-22: Hat Maker
23-24: Plumer - a dealer in feathers
25-26: Tailor
27-28: Private residence
29-30: Dressmaker
31-32: Laundry/ Launderer
33-34: Tanner
35-36: Mask Maker
37-38: Outdoor stable
39-40: Alchemist
41-42: Inn
43-44: Jeweler
45-46: Tavern 
47-48: Seer
49-50: Cordwainer (unusual leathers)
51-52: Seamstress/tailor (non-standard humanoids)
53-54: Glover
55-56: Restaurant
57-58: Dress maker
59-60: Alchemist 
61-62: Fine Clothier (custom- small humanoid races)
63-64: Cloak and Robe Maker
65-66: Dyer
67-68: Bonesmith
69-70: Shoemaker
71-72: Tavern
73-74: Fine Clothier
75-76: Furrier
77-78: Private Residence
79-80: Seamstress/Tailor (custom small humanoids)
81-82: Inn/tavern
83-84: Bootmaker
85-86: coatmaker
87-88: Beltmaker
89-90: Backpacks, Pouches, Satchels
91-92: Tailor
93-94: Private Residence 
95-96: Restaurant 
97-98: Glover (exotic materials)
99-100: Private Residence

This is Ward 13, it is primarily filled with establishments related to the trade and creation of clothing. It is the domain of House Kesht. 

There is an establishments list. There are around 30 different establishments, some come up more than once in the list. There are 49 entries on the table, and every ward has a warlock tower, garrison and temple that are placed by the ref. As the PCs explore the ward when they ask about a particular building I roll on the list. Every list has generic listings like inns, taverns, stables, seers and alchemists, and entries keyed to the guild/trade associated with it. 

There are around 100 buildings in this ward. So let’s do a thought experiment, say the party took up residence in this ward and adventured in the campaign for a few years. They would slowly but surely encounter the buildings in the Ward and they would be specified. 

So as a thought experiment, what would the ward look like if the party went to EVERY building in the Ward? Each time you would roll for the building in question, so just for fun I went to Discord and rolled percentage dice 100 times then noted the results. 

This produced:

Private Residences - 9
Bootmakers -7
Dressmakers -7
Tanners - 6
Jewelers - 6
Clothworkers/Clothiers - 5
Tailors - 5
Taverns - 5
Dyers - 4
Inns/taverns - 4
Alchemists - 3
Bonesmiths - 3
Fine Clothiers (custom- small humanoids) - 3
Furriers - 3
Laundries/Launderers - 3
Outdoor Stables - 3
Restaurants - 3
Shoemakers - 3
Backpacks, Pouches, Satchels - 2
Cloaks and Robes - 2
Seamstresses/tailors (exotic materials) - 2
Seers - 2
Beltmakers
Coatmakers
Cordwainers (workers in fine leather)
Glovers
Glovers (exotic materials)
Hat Makers
Inns
Mask Makers
Plumers
Seamstress/tailors (non-standard humanoids)

There are 53 wards in the city, a small number of them are mostly residential, mostly parks or mostly agricultural, but the majority have mixed use buildings. This system allows me to instantly determine what any particular building is for, without having to list them out ahead of time. 

But almost as important as that, this system allows the ref to explore the city right along with the players. I don’t have all the answers when we start, I discover them as we play the game. I find that this makes the setting uniquely alive in a way that a heavily detailed, fully specified system would not.

One of the brilliant ways that 1e AD&D handled the creation of entire game world’s was exactly this, using bounded randomness and tables to generate material on the fly. What I have done here is simply an extrapolation of that on a bigger scale. Good game design scales up well. 







Wednesday, January 6, 2021

Character Investment in D&D 


Over the last year or two I have informally absorbed a lot of details about current TTRPG trends and practices, Twitter is, of course, not a fully representative medium, but it does manage to distill a lot of things in plain view. I have also had more D&D related conversations with people in real life this past year than ever before, and what I’m seeing on Twitter is matching what I hear from those I speak to about D&D.

The most obvious example of what I’m thinking about here can be found in dozens of Twitter threads where people gush about their characters, I’ve seen people talk about:

Their character’s favorite foods

Their character’s favorite music

Their character’s relationships with other PCs or NPCs

Minute and numerous details about their character’s appearance, habits, etc.

That sort of stuff. 

There is a degree of investment in characters that is fairly intense. It is clear that these people are thinking about their characters… a lot. Creating details to flesh them out and make them “real” to the players. 

“Backstory” is a thing. I have seen endless Twitter threads talking about it, sharing examples, discussing how to use backstory in the game to create player buy-in and interest, and threads where new players express anxiety and stress over “getting the backstory right”, or “not having any idea what sort of backstory my PC would have”.

Now, just to be crystal clear, THERE IS NOTHING WRONG WITH INVESTING IN A PC IN THIS WAY. It’s perfectly fine to play D&D this way, and it clearly resonates with a lot of people.

And of course, that makes sense, role-playing has some powerful impacts on people, assuming the role of a character can be transformative, or at least self-revealing, there is a reason psychiatrists use role-play to build empathy and understanding of self and others. When you roll to hit, when you are talking in character, when your PC almost dies, these things create a strange, gestalt experience where your PCs actions feel like your own.

It’s a kind of magic, and something to be treasured. 

However, there is a shift in perspective here that is worth noting.

For many years TTRPGs like D&D were played in such a way that you didn’t have elaborate backstories as sort of the default. That doesn’t mean they didn’t happen, over the last 40 years I have gamed with a number of people who came to the table with multi-page backstories in tow. But it wasn’t thought of as necessary, and there were a large number of people who simply didn’t bother at all. 

So this isn’t entirely new, but the focus has certainly shifted.

What is the shift? Backstories are not new. Using them to inform the direction of the adventure is not new. Getting interested in the “fluff” of a PC (their favorite food, songs, etc.) isn’t new either.

I think there are two primary changes between now and “BITD”

First, the frequency and scope of this has changed significantly, more people do it, and they are doing more of it. When I was speaking with a friend of mine who runs weekly 5e games he told me that all of his players had backstories for their PCs, and all of them were big, not two sentence summaries like I used to see BITD, but multi-page, elaborate discussions with enough detail to significantly flesh out their PCs.

I think this is far more common today than it was in the past.

The other big difference, and I think the important one, is the focus of the game, on the character or on the player. 

In old school games, the PC is your avatar, your “skin” if you like, but the DM is not challenging your character, they are challenging YOU. 

Yes, you do things as your PC that you couldn’t and wouldn’t do as a person, and yes, to a degree you “become” your PC when you play. But the point of old school systems is to challenge THE PLAYER through the PC. The PC is YOU, but you filtered through the lens of a fantasy setting.

The reason for this is simple, D&D is primarily a game, the game challenges the PLAYER with a dangerous and exciting environment and if they succeed their PC will survive and amass wealth and power. The game is primarily player focused, not character focused. 

This has a number of well known knock-on effects, so in old school games you are told not to get too invested in your PC until they have been around for a few levels, you are less likely to lean on rolls and instead expected to describe what your PC was doing in order to test the player’s decisions, not the PCs stats, the encounters are not balanced as it isn’t expected that the PCs will automatically survive encounters. The almost endless stream of discussion about CR ratings and how they are to be used to help to ensure that encounters are “fair” and “balanced” is a great example of this. Balance is an artifact of catering to the PCs, not the players, as it relies on judging the PCs abilities and competencies to determine if they can survive, not the player’s skill.

The focus today seems to have shifted significantly, from player to character. 

There are probably some meta-reasons for this that are cultural and beyond the scope of what I’m discussing here. Perhaps it’s the increased focus on identity politics in our current culture, since we are very interested in how identity shapes so many aspects of our lives, it makes sense to be focused on fleshing out character identities.

Perhaps it’s the increasing influence of ‘narrative’ elements on gaming, with the shift to treating D&D as “storytelling” and the rise of more narrative mechanics in newer games, this may simply be the result of players absorbing that shift and expressing it through their characters. 

I’m sure there are other possible explanations.

The last issue related to this that comes to mind is that there is an important difference between developing your character through backstory and building elaborate character details like favorite drinks, favorite music, etc, before play, and developing your character primarily through play. 

In the first case, your character comes to a certain degree fully formed, with motivations, expectations and a history. This isn’t a bad thing, but it does mean that you are boxed in to a certain degree, and leads to a lot of “but my character would do that!” moments. 

In old school games, your actions at the table become your backstory, and you discover your character as you play. Rather than put the process of discovering your character into the phase before you start playing, that process is shifted to the table, so you discover your PC through their actions and the reaction of the game world to their actions. 

Of course, players with backstories can do this as well, but the larger the backstory, the more involved and deep it is, the more constraints it puts on how the character will be played. Yes, an elaborate backstory can also inspire role play and exciting adventures, it isn’t necessarily a straight jacket on how the character is played. But it does change the tone.

As just one example, players in my games sometimes say, “but what would my character do?” when presented with challenging situations, but most often they say, “what should I do?”, because the focus is on them, not their PCs.  I find that the more backstory focused a player is, the more likely they are to want to know what their character would do in situations like these.

I don’t think there are any big lessons to be learned here, the only real potential negative aspects of elaborate character backstories and focusing on the character rather than the player that I have seen are the anxiety associated with creating these backstories and the trauma and emotional damage done by character death.

I have witnessed the former on Twitter and first hand. There are regular threads on Twitter where new players express concern over their backstories, or even ask if they are going to be able to play properly without an elaborate backstory for their PC. We added a new player to one of my after school games recently, and when she arrived at the first session she was very awkward and uncomfortable. After the session we talked and I asked what had made the session difficult, her response was that she hadn't’ been able to come up with a good backstory for her PC so she was embarrassed and concerned that the other players wouldn’t like her PC. It’s bad enough to have performance anxiety during the game, but this player arrived expecting the worst because she hadn’t prepared a multi-page backstory beforehand. Clearly the expectations are shifting, as I didn’t ask for a backstory at all, and none of the players who were in the game had one either.

With respect to character death, yes, “old school” gamers can get invested in their PCs, that isn’t new at all. But it has been noted that the most current version of D&D softens up PC death considerably, and I don’t think this is a coincidence. If you show up to the table with an elaborate, lovingly crafted backstory that draws on the lore of the game world and creates a vibrant, fleshed out character that feels “real”, it’s not a huge surprise that you might have some problems when your PC dies in the second session. 

And I’ve seen this too. People posting on Twitter about depression associated with a character death, or people tsk tsking DMs who “needlessly kill off PCs” because of the traumatic effect that can have. I’ve even seen it bandied about that you should get permission from your players before their PCs can die in game. There could be any number of reasons for this last piece of advice, but I can’t help but think it’s connected to the fact that the game is character focused and the players have invested so much in developing the character that character death is a problem.  

At the end of the day I think it's important to be up front with your players about what is expected in the game, and to embrace player investment no matter what form it takes. So if a player wants a big backstory then they should have one! Never miss an opportunity to get players to buy-in to the game. But it is also important to be clear about what this shift in focus will mean at the table. 

Building Bhakashal - Trust the Process In a sandbox style game, the referee leaves things open and the PCs actions drive the play. This conc...