Layers of Design

or RPG design from the ten thousand foot view

What do people think of when they think of game design? Generally things like character generation, resolution mechanics, combat rules, etc. Sadly this common view often causes people to miss most of the game as it is actually played. This in turn results in either poor design that fails to achieve its goals, or in poor evaluation of existing games (which causes endless flamewars).

In the simplest terms, an rpg campaign consists of two very different influences- the Game itself and the Meta-Game. These terms are generally understood by most gamers and are a useful short hand. That split however covers too much ground for my taste and I’ve identified five layers (one in Game, and four in the Meta-Game) that differ significantly enough from each other that a solid treatment of the subject must give each their due.

Any game operates within (and is in effect designed for) these five layers. Let’s consider each of them in turn:



This layer is made up completely of the actual rules, i.e. the game’s mechanics. These rules must be objective and visible.

By objective, we mean that the mechanics would be resolved the same way and produce the same outcomes (assuming equal rolls, card picks, etc) no matter who the players are (assuming they are of course following the rules). For example, in HERO it takes a roll of 11 or less on 3d6 to succeed with an unadjusted 11- skill. Thus a die roll of 10 will be a success for any player in any group no matter the opinions and thoughts thereof.

By visible we mean that all data, resolutions and decisions in respect to the mechanics are visible to all players at all times.

It is very important to realize that this level’s primary purpose is to define LIMITS on the available actions and results. In chess rooks can move vertically and horizontality but never diagonally. A broadsword does a specified amount of damage- no more, no less in D20. Etc. It is by forcing limits that the mechanics assure objectivity, visibility and provide a framework for play.

Game play at this level is based upon known certainties. Examples: “The orcs have massed their greatest fighters in the center, we should…” or “I have a 68% chance of taking down the goblin, but only a 33% chance on the bigger hobgoblin. We’re outnumbered right now and that’s gives heavy modifiers against us so I’ll…”

In general tactical game design exists at this layer.

Near Game

This layer consists of any hidden mechanics.

This includes but is not limited to such things as pre-plotted movement (seldom used in RPGs, one example slightly modified is The Burning Wheel), various rock-scissors-paper resolutions systems (hand-to-hand combat in Top Secret, RSP itself in many LARPs, etc), hidden movement (defined by Line of Site rules for the most part, D&D 3rd edition for example defines the range limits of sight under various conditions), hidden damage, and simultaneous assignment (such as the drop of the initiative die in The Riddle of Steel)

The effect of this layer is to move decisions away from the visible certainties above to one of predicting the decisions of other players or guessing what information that they are hiding. Examples: “Tom tends to mass his power in the center but he likes the cover forests give as well. Let’s deploy here so as to be within range of both and send scouts in those two directions” or “I’ll buff with this hand, Bill can never tell when I’m buffing”.

It should be noted that this layer remains objective and as such contains many of the characteristics of the Game layer. But it has moved into the Meta-Game to a large extent as the focus switches from visible characters and their abilities (stats, locations, resources) to players and their abilities (to guess and deceive).

In general, strategic game design exists at this level.


Near Meta-Game

This layer consists of any subjective or invisible mechanics.

Subjective Mechanics are those that rely upon the interpretation of the GM (or a player) to determine the final effect. D&D alignment is the classic example of such a subjective mechanic, the GM is required to determine when and if a character violates his Alignment and the player is required to role-playing with its goals and limits in mind. Another example is fumble mechanics where the actual effect is left up to the GM.

Another way of looking at these mechanics is to view them as ‘guiding mechanics’ or ‘advisory mechanics’. They may point in a direction, but how far you go and sometimes even if you make the trip at all is squarely a heavily subjective decision of the GM/Player.

A number of systems even toss combat modifiers that they normally reserve for the Game Layer here. In HERO System for example the GM is advised to give a bonus (+1 to +3) for creative and/or unexpected types of attacks. In Age of Heroes, I leave specific terrain modifiers up to the GM after providing a few examples.


This layer consists of pure subjective resolution/methods, ‘group contracts’, and role-playing itself.

Subjective resolution is free form or mechanic-less play. The classic example is a GM or player just calling the outcome of an action: “I think your character is more than able to track down the guy by using legwork, you find him that evening at 8 PM”.

However subjective methods covers more ground first comes to mind including the ‘world’ and adventure design essential to RPGs. If the GM has decided (without resort to mechanics) that the mob boss has three bodyguards, each highly skilled- he has engaged in the pure subjective resolution of a question.

Huge sections of most games are turned over to Subjective Resolution/Methods. One of the interesting aspects of that is that by nature you can’t design rules for it, each person does that himself according to his own needs and desires. At best one can provide advice. Thus in game design terms, this method is defined not by rules enforcing limits, but by the absence of rules preventing subjective decision-making. It is worth remembering that not including rules is as a result game design as well.

Group Contracts are those agreements made by members of a group to either engage (“we want to do a lot of hack and slash”, “Let’s try to keep the game focused on court intrigue”, “players should create characters that work with others”) or refuse (“we don’t want evil characters”, “no rape will occur in this campaign) events, ideas or storylines. Such group contracts are often informal although there are exceptions, and they tend to be added on top of the game outside the control or influence of the designer.

Role-playing itself is typically done at this layer with the players determining the personality, actions and reactions completely on their own without input from mechanical rules.

The main characteristic of this layer is that it’s subjective. It is also limited only the constraints accepted by the players themselves. Since it’s contained mostly within the minds of the players, it is also in large part hidden with uncertain influences or outcomes.

Far Meta-Game

These are influences unrelated to the game itself, but even so they still carry great impact. There is almost no way to define all the possible examples for the extent of their reach. Common ones are “GM’s girlfriend syndrome’ or ‘Sherri worked late and is off her game, let’s be easy on her’.

There is however one point in this layer where game design does matter: the decision to play the game at all. Here we get game design that attempts to make entry easy and attractive either by way of light and simple rules, by seductive settings, or ‘new’ concepts. These days we see lots of effort to design at this layer in an attempt to expand or open up new markets, often at the cost of other Layers.


Designing by Layers

One of the first things a designer should do is decide which parts of the game are to be handled by which Layer and for what reasons.

This will in large measure determine the character of the game and the campaigns that result from it. Those things contained at the Game layer will be highly defined and limited. Those at the Near Game will defined and limited, but unknown to some in the short term. Those at the Near Meta-Game will be guided, but not specifically controlled. The Meta-Game level itself is its own lord and master. The Far Meta-Game may cause the design to forgone certain mechanics as ‘too complex for his market’ or otherwise unacceptable.

For example, I designed Age of Heroes to handle character creation, advancement, combat, etc at the Game Layer. A few elements (like the Personal Appearance Stat) are covered in the Near Game. I assigned a large number of areas (all the world design, storyline, and role-playing) to the Near Meta-Game. I gave no attention to the Far Meta-Game not really caring why people chose to play.

An important concept to keep in mind is the fact that it’s quite possible to move elements I lumped into specific layers above to another.

A number of games for example move some parts of role-playing from its normal Meta-Game subjective method to a subjective mechanic in the Near Meta-Game. (D&D’s alignment) or even to the Game Layer (CoC’s Sanity at certain points). Another example of crossing Layers is those systems that determine at the Game Layer which player has control (and sometimes for what ends) in the Meta-Game.

Since such ‘crossing’ of layers is typically defined in one and resolved in another, I’ve coined the time “Calling to the X Layer” (such as Calling to the Meta-Game) for this type of design. It’s an increasing common method that many are finding exciting.

Judging by Layers

Anyone attempting to judge the usefulness of a game system can benefit by considering five layers if they are interesting in not greatly misrepresenting other people’s campaigns and tastes.

For example, it’s easy to say that D20 or Age of Heroes are purely hack and slash designs given that most of the rules cover combat and near combat events. They have forgotten the importance of the Meta-Game level and the fact that both games are specifically designed to use it for certain parts of the campaign. It’s entirely possible for a group to spend five gaming nights in pure role-playing without a single die being toss, and then engage in a single evening’s combat. To characterized such a campaign as hack and slash would be a grave error- and a defining statement of the limits of looking at a single Layer.

Knowing the Layers and your own tastes in them can be helpful anytime you’re thinking of trying out a new game system. It will quickly point you towards things not to your taste and allow you to house rule it away from the start or to turn your attention elsewhere.

Theory and Topics Main Page