Attractors and Caster/Martial disparity

Systemic causes of a long-standing argument

Today I’d like to (perhaps unwisely) dip my head into one of the longest-running arguments in the TTRPG community: the caster/martial disparity. This is a very broad, though controversial, concept that affects most TTRPG systems out there. What is it? What causes it to exist? Why is it bad? How can we solve it? For answers to all of these questions, and more, read on.

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.

What are we talking about

Image 1: The caster/martial disparity argument bingo card.

Let’s begin by explaining what this whole argument is about.

Originally, I believe that the term “Caster/Martial Disparity” comes from DnD. In DnD, all players pick a “class” for their characters - a bundle of abilities which grow stronger by gaining levels. These classes can be grouped into two broad categories: “casters” and “martials”.

A “Caster” is someone that can cast spells: a wizard, cleric, druid, and so on. A “Martial” is the opposite: someone that cannot cast spells. In between these extremes there are “hybrid” cases, such as paladins, who can cast only a little bit, and are mostly similar to martials.

“Caster/martial disparity”, or C/MD for short, refers to a visible disparity in abilities between these two groups. Casters generally seem to be more broadly capable than martials, and directly outclass them in many important areas. The reverse - martials outclassing casters - does not seem to be the case. This difference in capability leads to several issues, centered around player enjoyment and difficulty of GM work.

The arguments around C/MD are very messy, and typically mix concepts from many different areas, such as:

• Does C/MD even exist?

• Does it exist only on paper or also in reality?

• Is it a bad thing if it does?

• How bad is it and on which levels of the game it is present?

• How could it be mitigated?

• Why might the proposed mitigation strategies not work?

And so on and so forth. A good C/MD argument can last for 30 forum pages without once repeating itself.

While the term may have originated in the Dungeons and Dragons community, systemic causes that lead to it are going to be present in almost any system that has magic. The reasons behind the varying levels of effectiveness of the commonly reinvented “fixes” are going to be broadly applicable too.

I think that this topic should be interesting to anyone with a passing interest in worldbuilding or game design.

Seven challenge areas

Before we get into the weeds, let’s figure out the overall shape of the argument. C/MD talks of “abilities” or “capabilities” and a “disparity” between them. These are very generic terms - can we narrow them down a little bit?

I think one of the good ways to think about it is to look at the character’s abilities in the context of different challenge areas: situations that consistently come up in TTRPG play and that present a specific type of problem to the players. I distinguish seven main challenge areas. This is, naturally, not an exhaustive list, but one that focuses on the most common patterns. If we can show that C/MD holds here, then I believe we can confidently say it holds in general.

1. Combat. Any situation where the primary challenge is killing or disabling your opponents, or helping other people kill or disable their opponents.

2. Social encounters. Main challenge is to convince someone else to do or believe something, or to get some information they know and do not want to divulge.

3. Logistics & Survival. Broad set of challenges revolving around logistics - traversing environments, moving from point A to B at speed, storing or moving large or delicate items, orienting yourself in an unfamiliar environment, finding paths to a desired location, as well as finding or storing food and water.

4. Information warfare. Any challenge revolving around finding true information or concealing it from someone else. This involves searching for something hidden or hiding it from someone else, trying to figure out the truth from a set of lies, deliberately misleading enemies, collecting information, translating things, or otherwise interacting with the information field.

5. Security & security penetration. Set of challenges related to securing locations or objects from intrusion, as well as bypassing those securities. Locks, surveillance, going through walls, bypassing or setting alarms, and so on.

6. Puzzle solving. Once information is collected, you may have to analyze it and come up with a solution that was not obvious from the start. I would call this type of challenge “puzzle solving”. This includes both explicit puzzles (e.g. magical locked doors) and implicit ones, such as deciding on how to approach a complicated situation with a lot of moving parts.

7. Uniquities. Unique challenges that avoid good classification, but are nonetheless extremely important, such as conducting or interfering with important magical rituals.

Notably, this is a classification of challenge areas, not player abilities. Shooting out giant lasers may seem like a combat ability, but if it is used for intimidation, then the challenge should be appropriately classified as a social encounter. As a result, some challenges may fit into several of these categories, depending on how you look at them.

Given these frames, we can make a spectrum of factual C/MD arguments when analyzing a given TTRPG system:

• Hard C/MD: Casters are equal or strictly superior to martials when approaching every type of challenge: they have more and better abilities, which means they are more likely to achieve their objectives, and can do so faster and with fewer costs.

• Non-combat C/MD: Casters are equal or inferior to martials within combat encounters. Everywhere else, they are equal or strictly superior.

• Soft C/MD: While both casters and martials have their areas of expertise, casters have more of them, and their contributions outside of their areas of focus are more effective than the reverse. Overall, casters are thus superior to martials.

• No C/MD: There is no caster/martial disparity: casters and martials have similar degrees of competence across the board, perhaps with individual challenge areas leaning themselves more to one or another type of character.

These arguments are strictly factual: C/MD is either present or it isn’t. On its own, this wouldn’t make the C/MD topic so contentious. To get it there, we have to introduce a value judgement.

Why would C/MD be bad?

For the sake of the argument, let’s put aside the question of wherever C/MD exists and answer a much simpler question: if it did exist, why would it be a bad thing?

This isn’t a trivial question. TTRPGs are a team game, and, by default, assume there is no inter-party conflict. So what if some players are more effective than others, as long as everyone is working towards the same goal? After all, this isn’t a video game where we have to keep track of “balance”, lest one player kill another too easily. For example, in soccer we could say that players in the field do a lot more than the goalkeeper, but we wouldn’t say that was bad, right?

Comparisons between the effectiveness of players and the GM likewise seem quite strange. It doesn’t matter wherever the player can do more or less than an individual NPC - GM controls the entire world, and so is by definition immensely more powerful. Even if we assumed some kind of competition was in play, one of the sides loses by default.

The actual reasons behind C/MD being bad require a more detailed look at the game.

First of all, people like to feel useful. If you are playing a martial character in a party with a caster who simply outclasses you, you may feel that there is no reason for you to show up. If you aren’t contributing anything novel, why waste your time? Naturally, not everyone will feel this way, but it is a common quirk of human psychology.

This sort of feeling of uselessness falls on something of a spectrum. If you are useless for one task, but then useful for another one, that’s not so bad. But if you are useless for an entire session, that feels really bad.

Another problem is the conflict between the player concept and the player’s desire for contribution. Players do not just want to feel more useful than an empty space; if they can, they will try to minimize overall hardship for the party. This produces a social pressure: sure, you may have wanted to play a martial, but the party could really use a wizard. Similarly, you may feel compelled to pick up abilities that didn’t fit your original character concept, simply because of how useful they are.

Next, as a GM, C/MD will present challenges when designing combat encounters. Most TTRPG systems assume that all character classes present an equivalent amount of challenge as enemies at a given level. This is of course false if C/MD is true: a wizard will be a much tougher fight than a fighter, all else being equal. While an experienced GM could intuit the degree of this additional difficulty and adjust the encounter accordingly, this imposes additional mental load, and is very difficult for people new to the hobby.

Similarly, the GM may have to take the classes of his players into account. An encounter that is a significant challenge for a group of fighters can be a breeze for a group of wizards. This is especially relevant when borrowing encounters from a published adventure, as opposed to making them yourself. Adjusting an encounter to match the player class distribution can be arbitrarily difficult. Skipping this adjustment may result in encounters that are either impossible to complete or that will be bypassed entirely.

Finally, C/MD presents implicit challenges to your worldbuilding. If casters are significantly more capable than martials, then they will tend to aggregate in positions of power. Depending on how severe the disparity is, this can lead to a strong class system, with casters at the top and everyone else at the bottom. This may have significant impacts on your setting, ones you did not want to introduce.

Of course, all of this only matters if we can demonstrate that, at minimum, the Soft C/MD claim is true for whatever system we are looking at. To make that easier, I will introduce a new concept: substitutability.

Substitutability

Substitutability is an ability of something to substitute something else. For our purposes here, I am referring to abilities (e.g. spells) of characters, and how well they can substitute other abilities. For linguistic simplicity, in this section I will call the user of the ability “caster”, even though this logic applies to all types of abilities, magical or not.

Wherever ability A can substitute an ability B obviously depends on the context: what are we trying to do and under which circumstances. To clarify this somewhat, we can imagine a test environment (e.g. a blank white room with a single enemy a specified distance away from the caster) and a goal (kill the enemy). If the probability of achieving that goal is higher if the caster has ability B than if they do not have it, all else being equal, then we can say that ability B is useful for achieving the goal within the corresponding environment. If ability A is useful in the same environment in which ability B is useful, and ability A achieves the same or greater chance of achieving the goal as B, then A is substitutable for B.

Other abilities of the caster are included in the environment. For example, imagine two spells: one can shoot any enemy within 60 feet, and another can shoot any enemy between 20 and 80 feet away. If the caster cannot move, then the latter spell cannot perfectly substitute the former; but if they can move at least 20 feet, then the caster can completely negate the “blind zone” at the center of the latter spell. Therefore, the ability to move has to be part of the testing environment.

Obviously, we aren’t interested in an arbitrary combination of environment and goal. Instead, we are only interested in environments that are typical of TTRPG games: castles, dungeons, wilderness, and so on. Likewise for the goals: seven challenge areas I outlined previously can be seen as a practical example of such goals. Therefore, when I talk about substitutability, I am talking about an average measure of substitutability across all common environments and goals.

In general, substitutability is not symmetric: just because A can be substituted for B does not mean B can be substituted for A. For example, an attack that has a range of 20 000 feet can substitute many attacks with a smaller range, but the reverse is not true.

Substitutability allows us to formalize what we mean by casters being “superior” to martials: we mean that a caster is substitutable for a martial, but the reverse is not the case; furthermore, the situations where a martial could not substitute a caster are often very important.

Thaumaturgical Attractors

It is impossible to analyze every TTRPG out there for the presence of C/MD: there are too many of them. Nonetheless, I think it would be instructive to conduct this analysis for one game: Pathfinder, which I have done in a helper post. Analyzing other systems would be structurally similar, and is left as an exercise for the reader. I think it provides a very instructive look into the patterns in which C/MD manifests.

I also analyze some of the common objections to C/MD. Some of these are specific to Pathfinder; for others you can draw full or partial analogies for your system of choice. If you would be bored by Pathfinder, feel free to skip this helper post entirely.

Having done that analysis, we can glean a larger truth: that C/MD is an attractor.

An attractor is a state in a dynamic system which makes that system evolve towards it, regardless of where it initially started. For example, imagine a coin dropped into a bowl: no matter how the coin was spun, or where it entered the bowl, in the end, it will end up at the center. The center thus “attracts” the coin, and is an attractor.

Because of this attractive property, unless deliberate steps are taken, every TTRPG will tend to evolve in the direction of having C/MD, and once it has C/MD, in the direction of having more severe cases of it.

C/MD can be split into several contributing effects - properties of the publishing cycle, or consequences of simple assumptions designers will tend to make about their magic system. Individually, these effects can be harmless; but taken together they will, over time, lead to C/MD.

Contribution 1: General magic exists

Generality is a measure of how “broad” a magic system is. The more different areas a magic system can cover, the more general it is. For example, the magic system from Avatar is very specific: air benders can only manipulate air, and only in a very specific set of ways. On the other hand, the magic system of DnD - even if you limit it to only the published spells and magic items, making it very hard - is extremely general.

The more general a magic system, the higher the chances it will lead to severe C/MD, for fairly obvious reasons. Unfortunately, generality is also the default assumption people tend to make about magic. In fact, authors like Brandon Sanderson are notable because their magic systems are quite specific, and so stand out from the crowd. If the designers do not carefully think about what their magic can do, it will be capable of anything whatsoever.

This isn’t to say that specific systems cannot have C/MD, but generality is obviously a major factor in producing and exacerbating this effect.

Contribution 2: Magic is exclusive

Who can interact with magic? Not to cast spells - I mean interact with it at all. Almost universally, there is a very strict separation - people are either magical or completely normal. And for normal people, magic interactions are extremely one-sided.

Even sensing magic usually requires more magic. The only setting I can name off the top of my head where active magic can be sensed without an explicit magic-sense is Shadowrun. And of course you can forget about analyzing, modifying, reflecting, destroying or casting spells.

This pattern isn’t limited to people. The entire world is split into two parts: magical and non-magical, with very little overlap. It is almost like the magic system is grafted on top of the normal laws of physics, as opposed to being an integral part of them - because if it was, you would expect to see little interactions all over the place.

This is further exacerbated by magic access being an inherent, perhaps genetic property of a person. If you don’t roll right on the birth lottery, then you are completely locked out. You can pick up some skill with lockpicks, or dip your toes into diplomacy; but when it comes to magic, it’s an exclusive club.

There are obvious reasons why designers may decide to make their magic system exclusive. Keeping it separate from physics is much easier than tightly integrating the two. There is also a long fictional tradition of magic being “inheritable”, so it is natural for designers to rely on this trope. But this tendency makes it very hard for martials to even arrive at the same playing field.

Contribution 3: Few trade-offs

First two assumptions are not sufficient to cause C/MD. We can easily imagine a magic system which is extremely general and exclusive, but because of significant trade-offs in learning or using magic, avoids C/MD. To give a hyperbolic example, if casters could only learn a single spell throughout their entire career, then the generality of the system as a whole is irrelevant.

Unfortunately, designing major trade-offs is hard.

First of all, consider the business constraints under which the system designers operate. They have to make their game system interesting to consumers in order to hook them into purchasing the rulebooks. In broad strokes, this means that their magic system has to present interesting possibilities from a cursory read.

This puts a “floor” on the strength of the magic system: it can’t be weaker than what would be interesting at a glance. On its own, this isn’t a death sentence. You can design a system of tradeoffs that leaves a magic system both obviously interesting and not overpowering. But where would such trade-offs come from?

First option is that system designers could make a conscious decision to block magic abilities from some area of the game. For example, they could decide that magic outright cannot affect the social sphere: you have to use non-magical skills to do diplomacy. This would, inherently, cut out an area of the game which requires skills; if you also make sure that investing into both non-magical skills and magic is prohibitively expensive, then you will counter C/MD.

But this has to be a conscious decision by the designer. As I explained above, the natural, unspoken assumption is and always will be that magic can do anything, and thus that magic can substitute non-magical skills. If additional effort isn’t put into working against it, your system will simply slide right into the trap.

Another option is that magic is inherently incapable of some things, due to what we can call a “theory of magic”. This is a set of “physical laws” determining what your magic can and can’t do. For example, in Avatar, the magic system is built on the ideas of controlling the four elements: it is thus almost exclusively limited to a sort of telekinesis, and nothing more.

Obviously, this is only possible if the magic system is built on such a strong theoretical basis in the first place. Establishing this foundation is hard work, requiring an investment of time and resources. For TTRPG designers, who often operate under time and monetary pressures, committing to designing a system of magical physics in order to avoid a niche problem with C/MD can be an unjustifiable expense.

Furthermore, such a system of magical rules also puts strict constraints on the setting of your game, cutting into your market size. It is not a coincidence that DnD is much more generic than the other, less popular TTRPGs.

Another option is for the system designers to split magic into several areas, responsible for distinct types of spells. In DnD, there is something similar to this: schools of magic, such as Divination, Necromancy, Illusion, and so on. Advancements in different areas of magic could trade off against one another, due to limitations of time or skill investment.

But this runs into largely the same problem as before: where would this categorization come from? Coming up with distinct areas of magic is a lot of effort. If the areas are not cleanly separated from one another, then over time, their boundaries will flake, as usage areas will bleed from one category to another. DnD is replete with examples where Illusion spells are indistinguishable from Enchantment or Conjuration spells, and Evocation is famous for being so generic you could rename it into “some magic”.

Taken together, there are multiple systemic reasons that would lead to the magic system being left with fewer trade offs than necessary.

Contribution 4: The Publishing Cycle

There is one final notable game design constraint: capitalism.

Unless game designers abandon the system entirely, they will have to sell new books. This means coming up with new abilities, plots and locations: things that tend to lead to power creep. Power creep is a simple observation that as games develop, newer abilities tend to be somewhat stronger than the older ones. This can have many causes.

First of all, over time, game designer skill and understanding of a system increases. They also acquire more game testing data, as well as player feedback. As a result, their newer designs will tend to fit better into the overall system philosophy, and will fill known holes in the ability space.

Another cause of power creep is purely statistical. You can imagine a bell curve of abilities: some are stronger, some are weaker, and most are somewhere in the middle. During character design, player may need to pick a certain amount of abilities out of all the ones available. If there are 100 options and they have to pick 20, they will pick the best 20%; but if there are 1000 options, they will pick out the best 2%. By increasing the overall number of options, game designers therefore increase the power of clever ability selections.

Finally, let’s not discount the fact that a designer may be inclined to “punch things up” with a new release, for whatever reason.

Power creep is not the cause of C/MD. But by its nature, it will exacerbate the tendencies already present in the game: strong things will become stronger, while weak things remain weak.

Why C/MD can be invisible

Fine, fine, - I hear you say - Sable, your arguments may be convincing, but I just don’t see this so-called C/MD at my table. How can you explain that?

Very simply, my friend. Anecdata isn’t data, and there are many reasons why C/MD can remain invisible for a long time.

1. You don’t run that many games

The simplest explanation for why you may not see C/MD is that you just don’t run that many games.

Imagine that you have been running a campaign for a year, with a session every two weeks. That may seem like a very long time, but actually, it is only 24 sessions. It is in fact quite likely that your players have only used a single set of characters. As a result, your experience with the topic of C/MD may be fairly limited. The chances are quite good that none of your players happened to pick up a caster build that would really showcase the problems.

2. Your players don’t play casters, or play weak casters

Of course, some people run a lot of sessions, and with very varied parties. But how often do they change their gaming group?

Individual people often play similar types of characters. Some people simply don’t like the vibes of casters, and never want to play them. Others play casters, but prefer to focus on areas of magic that aren’t that problematic - for example, specializing in direct damage. Still others may play casters, but are trying to do something so unusual that they avoid C/MD by sheer chance. Finally, some players know about C/MD, and consciously avoid doing things that exacerbate it.

Of course, if your gaming group is like this, you will not see C/MD in action. This reason can combine with the previous one - if your players play casters, but only rarely, then that will further reduce your effective experience.

3. Your players don’t play martials, or optimize well

Of course, the reverse may be true too! If your players don’t play many martials, then you will not see C/MD in action. Alternatively, if your martial players optimize sufficiently, they could make up for the inherent deficiencies in their classes with pure system mastery.

4. You, consciously or subconsciously, play into the hands of martials

In the previous sections my analysis was based on an assumption that casters and martials are equal before the cold gaze of the rules. But should we make this assumption?

GMs routinely deviate from the rules. In fact, this is completely unavoidable: nobody can remember everything that is written in the rulebooks. As various issues come up during play, both players and the GM rely on what they can recall of the rules - with their faulty, biased, human memories. It is completely natural that some GMs may rule more in favor of martials and against casters on key subjects - thus mitigating C/MD.

Furthermore, even if we assume that GM rulings are equitable, the vast majority of the game isn’t based on rules. NPC reactions, skill check difficulties, number, type and tactics of opponents, all of those facts of life are completely up to GM’s discretion. If the GM habitually tilts things in favor of martials, then this too could mitigate C/MD. These sorts of tilts may even be done subconsciously, without the GM realizing what they are doing.

This may, in fact, be beneficial to your playing group as a whole; but if we are to analyze the overall game and come up with solutions applicable to everyone, then we cannot rely on haphazard unstructured approaches invented by individual GMs.

5. Your players just don’t care about the disparity

C/MD is a problem due to many reasons, but a core one of them is that players with martial characters will feel powerless to affect the narrative. Of course, this is just a statistical tendency, and some players will simply not mind that much.

If your martial players have this personality type, then it is entirely possible that even though C/MD is present in your games, none of the players will complain to you about it, and therefore, you may not even realize it is happening.

Faulty solutions

So, here we are. We established what C/MD is. We figured out why it’s bad. We agreed it is present in your TTRPG. We even listed some common reasons why individual people may not observe it. Now, what can we do about it?

In truth, there is no one knock-down solution to C/MD. In order to create one, you would have to address C/MD at its core - at the systemic pressures that slowly drag it into any given TTRPG system, in a way that is applicable to your TTRPG. But there are some bad solutions, which come up a lot when discussing C/MD. Let’s review them here.

1. No More Spellbook

One approach addresses C/MD by attacking casters at their core: by cutting their access to magic.

The title of this section comes from one such approach. In DnD, Wizard class records their spells in a spellbook, and they require access to this spellbook to prepare new spells every day. So, the logic goes, if the Wizard is being “a problem” then simply take away their spellbook. Have it be stolen, or destroyed in combat, or whatever else. Problem solved!

This is something that is very specific to DnD, but I think we can expand the idea across many different games and systems. The central idea behind this approach is a hostile stance taken towards the caster. Many games have implements required for casting, which can be similarly taken away. Direct danger from magic can also be ramped up - for example, spells may occasionally explode, or the surrounding society can be bigoted against mages to the point of lynchings.

Of course, this is an absurd approach. While it may, perhaps, work to “eliminate C/MD”, it fails on a much more fundamental level: it directly targets player fun. In much the same way, you could “eliminate C/MD” by preventing all casters from existing, or by punching the player in the face any time they try to cast a spell. The whole point of even coming together to play for some players is to have fun with magic. This is a pyrrhic victory.

Now, I am not saying that you could never put your players in a situation that limits their access to key resources, or one where visible magic would be undesirable. These are legitimate challenges, and I am personally quite fond of an occasional use of this trope. But by being introduced in the context of a C/MD argument, this proposal assumes habitual use of these techniques, common enough to actually impact C/MD.

The point of TTRPGs is to maximize player utility. If you have players who want to play casters, their utility is obviously not going to be maximized by constant hostility towards them, unless they happen to be a rare specimen that revels in actively punishing situations.

If you think that the loss of utility from making a caster-preferring player play a martial is outweighed by the gain in utility from other players not having to deal with C/MD, then discuss this with your players from the start. Sneaky hostility down the line is never the solution.

2. Remove Items

Occasionally, people propose to limit player access to items - mostly by limiting what they can purchase on the market. I never understood this proposal, but it deserves a cursory mention.

I think it stems from a misconception regarding the causes of C/MD. Magic Items are not and have never been the real cause - it was always the inherent and inseparable abilities of the casters. In fact, while items have their faults, they are also one of the best ways to mitigate C/MD, by giving martials some access to the same toolset.

If you remove easy access to items, then in all likelihood you will worsen C/MD. Do not do this.

3. Add more items

An inverse of the previous suggestion - to give players more items making up for the fundamental weaknesses of the martials - is more workable. Unfortunately, as I have discussed above, items cannot eliminate C/MD. At best, they can mitigate it in relative terms.

This may, in fact, be sufficient for your gaming group. If the disparity goes down from being present in 80% of challenges to only 10%, players may not even notice that it is not completely solved. But when we talk about systemic differences, this approach seems insufficient.

For some players, gaining strength from items might go contrary to their preferences. It can feel quite lame to know that if not for your gear, you would be quite useless. This kind of strength does not feel “inherent”, unlike waving your hand to shoot out a fireball.

Additionally, there are worldbuilding effects to consider. If items are strong enough to compete for narrative weight with magic, and could be used by anybody, their effect should be felt in the setting. Imposing worldbuilding limitations on casters is fairly natural by simply tinkering with population statistics; limiting items, on the other hand, is a different type of challenge.

4. Ban Specific Spells

This proposal often comes up as a part of a larger mitigating strategy. There is absolutely nothing wrong with selecting specific spells emblematic of the C/MD issues and removing them from your game. Yes, this might not eliminate C/MD; but together with other approaches, it can get your group to a satisfactory point in a cost-effective manner.

This approach has two obvious limitations. First of all, you have to know which spells to ban - this requires either following the advice of someone knowledgeable about the game, or being knowledgeable enough yourself that you can intuit it. This isn’t trivial; I have seen a lot of misguided banning advice online. Secondly, whenever you introduce new published material into the game, you will have to play catch-up and expand your ban list.

It is important to note that fairly often this topic comes up not as an actual proposal to mitigate C/MD, but as a deflection, a way to deny the existence of a problem. A C/MD scholar will say “In social situations Charm Person is obviously rather strong”, and their opponent will claim ”Well, ban it if you have such an issue with it, I don’t see a problem”. Obviously, this should not be taken seriously.

5. Rule Zero

Final balancing approach that deserves mention is “rule zero”. This term refers to the idea that regardless of what the game rules say, the word of the GM takes priority; therefore, if game rules are written in such a way as to make casters superior to martials, GM can always change them.

While this is literally true, it is also a deflection. If I buy takeout, I can add whatever spices I want to it once it arrives; but I also expect it to be tasty without any modifications. Vast, vast majority of the GMs have neither the experience, nor the time, nor the inclination to do complex rule refactorings to deal with C/MD; that they aren’t prohibited from doing so doesn’t mean they will actually do it. If this problem is to be addressed, it has to be done by competent rule crafters, in a way that could be implemented right away by other people. Ideally, in fact, this should be done by the system designers themselves.

History of C/MD in DnD editions

Now that we have gone through all the various facets of the C/MD argument, let’s look at the changes that were made between the various editions of DnD with a focus on C/MD.

DnD 3.5 -> Pathfinder -> Pathfinder 2e

3.5 is the edition that, as far as I know, has the worst case of C/MD among all TTRPGs. This is partly due to its long timespan: power creep had a long time to set in, ramping the initial problem up into the stratosphere.

Pathfinder was designed by several third-party 3.5 writers that wanted to correct the perceived mistakes that were made when Wizards of the Coast released DnD 4e. Large chunks of the 3.5 system are copied wholesale, and many printed books are fully compatible between the two.

In transition to Pathfinder, many famous spells got eliminated or reduced in power. I do not think that Pathfinder designers were deliberately trying to address C/MD; instead, this was done due to the changes in the mechanics, and to make GM’s job easier. Nonetheless, it also obviously had beneficial effects on C/MD.

Unfortunately, this kind of accidental improvement could not radically change things. Even in the core rulebook of Pathfinder, C/MD is out in force, and it only gets worse from there.

Pathfinder 2e did have C/MD in mind during the design process; nonetheless, they relied on the types of strategies that could only mitigate C/MD, but not resolve the core problem. Over time, the situation will inevitably worsen, and we will be back at square one.

DnD 3.5 -> DnD 4e

Chronologically, Pathfinder was designed as a response to DnD 4e taking a very novel approach to DnD design. Changes 4e made were highly controversial, but they did, in fact, solve C/MD.

First of all, combat was radically altered, turning it into a tight strategy game similar to the likes of XCOM or Divinity Original Sin. This necessitated a much stricter separation between combat and non-combat interactions than what was present in the previous editions. Each class has their own distinct combat “purpose” or a niche, making combat decisions more interactive.

Out of combat, the skill system was changed from a “simulationist” to a “narrativist” approach. Instead of skill checks being used to achieve concrete simple effects (e.g. climb a wall), they are used as part of so-called “skill challenges”: group roleplaying events where all players come up with ways that their skills can contribute to overall problem resolution. This approach radically expands the usage of all skills; in fact, I would say that with it, you could, arguably, justify any skill as applying to any problem. I am not a fan of this sort of radical equity, but it does largely solve C/MD out of combat.

Finally, utility spells were changed into “rituals”, which can be cast by anyone with a ritual caster feat. Wizards do get that feat for free; but in principle, out-of-combat spellcasting ability is now available to everyone more or less equally.

The 4th edition way of doing things is just one approach to solving the C/MD problem. It isn’t for everyone, and many valid criticisms can be levied against it. Personally, this approach clashes with my reasons for playing DnD, but I still believe that it is a highly instructive learning case.

DnD 3.5 -> DnD 5e

DnD 5e largely went back to the 3.5 drawing board after the commercial failure of 4e. Nonetheless, significant changes were made, which lead 5e to become the largest commercial success among DnD editions to date.

Overall, changes can be summarized as “simplification”. Number of choices available to players in combat and during the character design stage were massively reduced; this brought the skill floor required to build competent martials way down. Various combat statistics - such as armor class - were deliberately designed to be harder to increase, bringing everyone into a fairly tight bracket of power. The skill list was reduced, eliminating many skill traps, and the versatile skill point system was removed, increasing skill trade-offs. On the magical side, all spells got a significant drop in power, most famously limiting continuous spells to only a single one acting in parallel. The total number of magic items used by a single character, as well as their availability, were likewise limited.

Together, all these changes significantly reduced C/MD. With combat no longer relying on easy availability of various magical buffs, monster and adventure designers are forced to keep things within the power limits of martials; this has a feedback effect. Out of combat, fewer spells can be truly considered “universal problem solvers”, letting martials better compete.

Nonetheless, same as with Pathfinder, these changes do not strike at the core of C/MD. 5e has not abandoned the historic roots of DnD, seeped in C/MD as they are, and as it continues to expand the problem will rear its ugly head once again.

Pathfinder -> Path of War

An alternative approach to handling C/MD comes from a third-party modification to Pathfinder called Path of War (PoW).

The basic idea behind PoW is fairly simple. There are two key problems with martials in combat: lack of interesting decision making, and a lack of ways to address unusual combat challenges. Casters, in comparison, can draw from a very diverse toolkit. To address this, PoW introduces a system of “maneuvers” - essentially, martial spells - and a set of classes built around them.

It is important to point out the differences in design direction between maneuvers and spells. While some maneuvers have utility uses, the overwhelming majority of them have exclusively combat uses: this means dealing with enemies, preventing damage, increasing speed or movement types of allies, and so on. Unlike spells, maneuvers can be recovered and immediately used again. This is a quick process, generally taking only a single round; as a result, PoW martials have an immense attrition capacity.

PoW does not address the out of combat issues of C/MD, missing the mark in regards to why this is a major problem in the first place. Nonetheless, it essentially completely closes off the gap in effectiveness between casters and martials in combat. PoW martials take a solid place in the combat triangle of DnD - excelling at putting down threats, and capable of doing some control and buffing if necessary.

Given the prevalence of combat in Pathfinder play, I think PoW is a very interesting case when it comes to addressing C/MD. Expanding the maneuver lists to include more out-of-combat options could pave a path towards a full solution to C/MD.

Nonetheless, some people dislike what PoW brings to the table. Concerns that it turns martials into “anime fighters” are common. While I do not share the preferences of people who express these concerns, I can definitely see where they are coming from. If giving martials a direct line to using semi-magical abilities is not your style, then you will need to search for a different solution to C/MD: perhaps even one that nobody has thought of up until this point.

---

I hope that this overview of the issues of C/MD was instructive, and you could take something out of it - either when it comes to running your own games, or designing TTRPG systems.

If you are looking for other posts on my blog, check out this list of all other posts, and if you enjoy what I write, you can subscribe to receive updates by email: