RPG Systems, and two Schools of thought

If you're familiar with pen and paper RPGs, you may recall some heated internet discussion several years ago when Dungeons and Dragons 4th edition was released. 'Edition wars' are not a new thing and I won’t repeat the debate here. I do however want to highlight one of the more common complaints I noticed, which was the notion that D&D 4th was ‘too much like World of Warcraft’.

I found it an intriguing notion. I did end up liking 4th edition, but still kind of understood what these people were getting at. There was a definite change in how the game felt in play, but back then I didn’t have the vocabulary to properly express these concepts. And I suspect, neither did the majority of people involved that discussion, which resulted in folks shouting past enough other in the internet trenches.

Now that I have a little experience, I do.

Two Philosophies

Most designers should be familiar with the term ‘Aesthetic’. Broadly speaking, it refers to how a game or an aspect of a game ‘feels’.

Aesthetic is unrelated to technical competency. A game with photorealistic 3D graphics may be impressive technically but the lack of a creative spark may leave it ‘feeling’ hollow or derivative.

It can also refer to non-visual elements of a work such as sound, writing, presentation, user interface, or game mechanics. It was the mechanics of D&D 4th in particular that were aesthetically different from it’s past iterations, hence the different aesthetic.

To elaborate, there are two distinct methods of approach to game systems in an RPG, for video games or pen and paper alike. Think of them as two opposing philosophies or on opposite ends of a sliding scale.

The Abstract approach is when numbers and effects are arbitrary, and a means to an end for creating interesting gameplay first and foremost. Abstract RPG systems have less regard for context: an enemy is a group of numbers balanced first and given form later.

A good example that I’ll be using often is World of Warcraft, where players went from a maximum of 6000 health during the game’s launch to literally over 500,000 health during the recent Mists of Pandaria expansion.

The number is inherently meaningless by itself. Does this mean player characters are thousands of times ‘healthier’ than they were a year ago at launch? Or does the enchanted armor giving us that health increase in magical power thousands-fold?

No, the number is what it is to create good gameplay. Barring gimmicks, a boss fight will always deal enough damage to challenge - not completely destroy - a player’s group. Ancient Troll gods will deal damage in difference of hundreds of thousands, with the only difference being that one appeared in an earlier expansion.

Another hallmark of an Abstract system is unity, in the sense that the many varied effects in said system all fall under well-defined classifications.

For example: A strong degenerative curse, a vicious poison, or stepping in lava may all count as ‘damage over time’. A wizard’s freeze spell and a thief’s tripwire both stop the player, and thus both count as a ‘root’ effect. Despite the wildly different contexts, they equate to the same net effect.

So what are the benefits of this approach? Your mechanics and systems usually end up more sound and coherent to everyone. Since numbers and effects are codified, the overall the game can be better balanced and players can understand the rules with clarity.

It also leads to a more versatile system for creating gameplay. For example, you know that a player will have X resources and stats at a certain level or point in a game. You then know you need an enemy that deals Y damage per second to challenge them. You crunch out the math, then give it context later.

The downside is that the game-ness of your system is exposed front and center, which might hamper immersion if people keep thinking of your game as a game. There would also be more pressure on your mechanics and systems to be well designed, since they’re on display so much.


The Deterministic approach is when mechanics and numbers are a part of the fiction of the game. They have inherent meaning.

In D&D 3rd, elder dragons had an Armor Class statistic wildly out of proportion to their level, giving martial heroes an almost unfair challenge. Yet in the fiction of the world, dragons have incredibly tough scales and the Armor Class reflected that. That mechanical imbalance was itself a form of storytelling, reinforcing how dragons were just that badass and special.

Earlier editions of D&D took this further by having a set of statistics for characters called ‘saving throws’ that were reserved primarily for defending against spells. The ‘save’ system was clunky and difficult to grasp, but because its’ mechanics felt odd and arcane to the player, it lent that personality to the spells themselves, and as a result dealing with magic felt unique and special.

A Deterministic approach is great for giving a sense of place to the setting, by having your mechanics be a part of the actual fiction.

The downside to Deterministic systems is that they’re less flexible, especially for video games. It would be quite difficult to create a well-balanced team vs team arena style game if you couldn’t shift numbers as necessary.

You are also expected to have rock-solid internal consistency. New content cannot break the rules followed by previous content. If trolls in a part of the world had a certain amount of health, trolls in another part should be the same unless the fiction reflected that; perhaps harsh climes or a breed with harder skin, not merely they are a higher level.

Back to the Dungeon

So how does this relate to Dungeons and Dragons? D&D 4th underwent a drastic change in it’s mechanical aesthetic, swinging hard from Deterministic to Abstract. As an example, let’s take a look at how the player classes changed.

In previous editions of D&D, martial heroes (Fighters, Thieves, etc) were very restricted in their options in combat. There were a few - such as tripping or grabbing a person - but the rules were complicated and unsatisfying, and it was generally more practical to simply deplete an enemy's HP as fast as possible. This led to the martial player eventually declaring 'I hit him with my sword' every round of combat.

By contrast, magic heroes (Clerics, Wizards, etc) had a lot of tactical and strategic choice by being able to cast practical and diverse spells with many different effects. Every time a magic hero used a spell it was a meaningful choice, because that resource was extremely limited and effectiveness changed with positioning, enemy type, and lots of other circumstances.

In D&D 4th, martial heroes have the same amount of practical options as magic heroes. Instead of ‘spells’ they have access to ‘maneuvers’ or ‘tricks’. The context may be different, but the way the abilities operate are the same under the hood.

You can read this and think 'Great, they balanced the game! Isn't this a purely positive thing?'

It's true that in previous editions, the options and power levels of each class varied wildly. However, that very unevenness contributed to the overall character of D&D and is almost iconic of the franchise. By giving every class roughly the same effectiveness and amount of meaningful choice in combat, some of the game's uniqueness was ironically lost.

While I cannot say for sure, I can see good reasons for World of Warcraft's dev team choosing to have their systems lean towards the Abstract. It gave them greater freedom in planning their content and allowed the loot-based progression they have at the game's level cap to exist.

It was also more important for them to have a well balanced game. We need to remember that In pen and paper, a weak character or class was not necessarily a problem. The imbalance could result in interesting roleplay, or the game master coming up with some other way for them to be interesting or otherwise contribute to the game/story. In World of Warcraft, if you didn't bring enough numbers to beat the content's numbers, a player was dead weight.

Without a game master present to create a living story, what you had was a purely mechanical combat challenge with a light layer of context. And without a game master to tailor that challenge to the group, class balance became paramount.

I can also see why the designers for D&D 4th adopted the Abstract philosophy. The mechanics of earlier editions were messy, often hard to grasp, and yes, unbalanced. There was an entire grid/positioning tile based combat system that groups used or didn’t at their whim, and judged as a game system it was far from perfect.

Their overhaul worked - D&D 4th is quite mechanically sound - at the cost of changing how the game felt in play.

Further Inflection

I’m not here to pass judgment on which approach is better. Both are valid for different purposes and can be used to evoke different aesthetics of play. Most games also don’t fall entirely at either extreme end of the scale. A particular game will usually lean towards one approach, but throw in some of the other philosophy to add spice to the game.

Take Diablo 2 - an almost purely Abstract game. Boiled down to it’s core, Diablo 2 was a gear treadmill with numbers on loot climbing ever higher and higher. Yet the green-colored set pieces carried properties and abilities not reflected by merely higher numbers. These abilities could also be unique to only that weapon. This, combined with a flavorful name, lent a lot of personality to those set pieces and made them a truly memorable part of the experience.

No matter the approach you choose for your game, what is absolutely critical is consistency.

Once you have chosen where on the sliding scale your game system lands, stick to it. You still have to design your gameplay around the approach you've chosen, and having internal consistency will help you avoid pitfalls of all kinds. Besides being rule 1 of worldbuilding, a player masters game systems ultimately by experimentation and trial and error. If the rules of that system are not consistent, mastery is impeded and causes much frustration.

A favorite example of mine to use here is the 10 player dungeon of Karazhan in World of Warcraft. The first boss is an undead horse. Slightly challenging, but there were no outstanding gimmicks that required specific classes to beat. Any defender or 'tank' class could fight him with no problem.

And one day, the horse was no longer affected by bleeding attacks.

This makes no sense from either an Abstract or Deterministic point of view. Looking at Warcraft as an Abstract system, the immunity to bleeding hurt Druid tanks who relied heavily on said attacks to be effective. Even worse, other ‘damage over time’ sources such as fire spells or curses were not affected. By excluding an entire class/character build from contributing, this change created bad gameplay - inexcusable under an Abstract system.

Looking at it Deterministically didn't help either, as there were other undead horses in the game and they all bled perfectly fine. If all undead things were immune to bleeding across the game, this change would make sense, and groups could plan around this restriction with certainty.

In Closing

While I’ve used RPG systems as a context, the Abstract / Deterministic philosophies also apply very much to mechanics from other genres and even entire games.

Look at how Left 4 Dead implements their weapons, compared to a tactical sim like Rainbow Six. In Left 4 Dead, the guns have exactly enough ammo to support the action-zombie gameplay - 50 rounds in an M16’s magazine, unlimited pistol ammunition, and so on. Rainbow Six models guns on their realistic specifications - an M16 rifle had 31 rounds (30 in the magazine and 1 in the chamber), all the better to support the realistic tactical simulation-fiction they were trying to build.

Mechanics propagate throughout the industry very rapidly, be it a pen and paper RPG in days past copying D&D, or a modern shooter adopting the Left Trigger - Right Trigger dynamic of Call of Duty. Something that’s popular is always quickly picked up by others.

Yet at this point, I feel the industry’s matured enough that we as designers can start to recognize exactly how a mechanic affects the overall gestalt feeling of playing our game. We no longer should be copying or iterating blindly. An ‘imperfect’ mechanic may be iconic of your franchise, for better or for worse.

As the industry slowly transitions from creating ‘fun’ to creating engaging experiences,  recognizing and being able to describe aesthetic will only benefit all of us in the long run. At the very least, we’ll be able to better tell others why something is great/sucks, and that by itself is a pretty worthy thing.