The Nine Structural Subsystems of Any Game

By Lewis Pulsipher [03.17.09]

 A game can be thought of as a system (as in "systems analysis", for the computationally inclined). What I'm trying to achieve here is a list of the fundamental sub-systems that are necessarily a part of any game (excluding sports such as baseball or swimming). This list may help inexperienced designers, because if they think about all nine of these systems as they rough out their game, this will help them conceptualize and arrive at a playable idea.

We could discuss endlessly what is a game and what is not; let's just recognize that, within your definitions of "game", you can probably find an exception that doesn't have all nine characteristics.   I think that's a function of definition rather than a failure of the analysis, but that must remain a matter of opinion.  If one of these systems is completely missing, you might have a toy or puzzle, but not a game.

There are many examples "on the edges", such as Katamari Damacy. To me, Katamari Damacy is not a game. Solitaire (the card "game") is not a game, because there's no conflicting interest, no active opposition guided by intelligence-they are more like a puzzle or toy. But both of these activities fit the Nine Structures framework.

I want a framework that will help a designer think about games. Some people, in listing fundamentals of games, discuss "state" in considerable detail. I've tried to avoid "state" and "state-changes" as much as possible, simply because I don't think that an organization dominated by state is very useful to an inexperienced designer. "State-change", in particular, seems to lump an awful lot together in one pot. My ultimate goal is to have something that will be useful to inexperienced designers, and to be able to expand each category to exhaustively list alternatives within each structure. I want designers to be able to treat the extended list as a sort of checklist, to help them make sure they've thought about all the vital aspects of their game early in the process.

I've tried to list these subsystems in an apparently logical order, but every one is just as fundamental as every other one.

Here is the list, followed by brief explanations and some examples:

1. Theme/History/Story/Emotion/Image.

2. Player Interaction rules.

3. Objective/victory conditions. 

4. "Data storage". (Information Management)

5. Sequencing.

6. Movement/Placement.

7. Information availability.

8. Conflict resolution/interaction of game entities.

9. "Economy" (resource acquisition). 

Sometimes the system is assumed, or the choice is to have "none", but still a decision has been made about the category. For example, in Tic-Tac-Toe (Noughts and Crosses) there is no acquisition of resources, but it still has an economy of "unlimited pieces" -- it could have a way to gain resources, and there are variations where you do. Another example: a very abstract game has no theme/history/story, but the designer chose to take that approach, nonetheless.

Theme/History/Story/Emotion/Image. These are listed in order of common usage, not necessarily importance. Story can be absolutely vital to a role-playing game, but is essentially absent from many games. Historical games use history to a greater or lesser extent. Many Euro-style boardgames have a theme that may or may not have affected the construction of the game. And we can still have abstract games without anything related to theme.  Many video game designers want to design "an immersive experience" to elicit one or more emotions from players.  And even a single image in one's mind, a scene or "movie clip", can characterize a game.

Player Interaction rules (and number of players). Is it a cooperative game, or a game like Doom (the boardgame) where one player controls the "badguys" and the others cooperate against him or her, or a competitive game (typical), or is there some other relationship between and amongst the players?

How many separate interests are there in the game?  How many sides?  Some "games" have only one and so may be more properly be called puzzles or toys.  Some have several sides (many boardgames, some online RTS).  Some have just two sides but several interests because there is more than one player per side (Team Fortress, etc.). 

This subsystem determines how the players interact with one another.  For example, in a multi-sided game, are negotiations allowed? Physical intimidation? (The answer to that is almost always "No", but it is a decision, and I have seen games that involved physical intimidation...). 

Objective/victory conditions. In other words, what causes one player to win, or at least causes the game to end, or is the goal ever-pursued but perhaps never reached?  The game ending can be arbitrary ("play five rounds"), yet there will usually be a way to determine the winner at that point.  Role-playing games have no end, and usually don't have winners, but do have objectives: usually to acquire experience points and (magic) items/skills/perks.

"Data storage". (Information Management) Something has to record the current state of the game. This is often a board/map.  In Tic-Tac-Toe, it's the nine-box layout.  In card games, the layout of the cards on the table, and the cards themselves, store data. Pieces can store data, in particular the traditional cardboard pieces of wargames that contain movement, attack, and defense values. A detailed map stores LOTS of data.  A computer can store vast amounts of data, of course, though early computers were very limited in data storage, which in turn limited the games.

Sequencing. In what order do things happen? "Simultaneously" can be the answer, but taking turns is the norm in non-video games.

Movement/Placement. The most typical "piece" in a computer game is an "avatar", a figure/character representing the player. Players generally manipulate something, most often pieces on a board or cards in their hand or on the table. Chess and checkers have movement rules, the Asiatic game Go has placement rules. Movement/placement one at a time is the norm in traditional games, where in wargames a player can typically move all his pieces in one go. Even paper-rock-scissors has movement (as well as sequencing) rules.

Information availability. What information about the game is available to all players? In traditional boardgames all information is available, but in card games information is largely hidden. Five-card Draw poker has a lower level of information availability than Texas Hold 'Em, because in the latter you see some of the cards "held" by the other players.

Conflict resolution/interaction of game entities. What happens when an action of a player leads to a conflict? This can be as simple as in Tic-Tac-Toe (conflict is not allowed, you can't place your mark where the other player already has one), or it can be simple as in chess (when a conflict occurs, the moving player always wins).  In checkers you jump a man in a conflict.  In Go you surround stones to capture them.

You might prefer to say that Tic-Tac-Toe has no conflict rules, that movement rules govern where markers can be placed; but a choice has still been made, that there will be no conflict. It is quite possible to have a game without conflict, such as a race game or many card games (Solitaire) and Euro boardgames.

"Economy" (resource acquisition). How are new pieces/capabilities acquired? Some games have no way to acquire these, but that is still a decision made about the game. Even games that don't appear to have an Economy have some elements, for example, in chess you can promote ("queen") a pawn, and in checkers you can make a king. Many modern games, especially many computer games, are largely economic/resource management games.

In video games there are very often ways to obtain new capabilities, whether it involves mining resources and building factories, or just picking up medkits and weapons that sit in convenient spots.

Am I sure there are just these nine?  No, but I haven't added to the number in more than a year, though I have revised it.  I also have a list of 20 questions that designers ought to think about, but which can generally be ignored when creating the framework of a game.  This will have to wait for another time.

Very useful for learners is to take simple games and change one of the structural choices.  This is especially easy with traditional games that "everyone knows" such as Tic-Tac-Toe, Chess, Monopoly, Risk.  For example, the well-known hidden-movement chess variant "Kriegspiel" is a case of changing from perfect information to very limited information for the players (system 7). 

The Monopoly variant where someone on Free Parking collects miscellaneous fees that would normally go to the bank is an example of changing the economy of the game slightly (system 9).  Increase the Tic-Tac-Toe board to four by four, and let a player win with four in a row or four in a square, and you have a much better game: you've changed the data storage and the victory conditions (systems 4 and 3).

 Now for examples.

Traditional games are almost always turn-based in sequence, with one piece moving.  Think chess (including oriental versions), checkers, Go, Monopoly, Parcheesi.  Certain genres of video games are almost always simultaneous movement (real-time), such as most shooters (Worms Armageddon is an exception of sorts).

How do video games fit? Most "shooter" video games follow the same pattern:

1. Theme/History/Story/Emotion/Image.  Usually, the story is an excuse to get to the action, though there are shooters with deeper stories that actually affect gameplay.  Many "elicit an emotion" games are at least partly shooters.

2. Player Interaction rules (and number of players).  Generally these are one-person games, though now we're getting more cooperative/buddy versions.  Many have a multi-player (but two-sided) version as well.  There are rarely player interaction rules other than common courtesy.  Some players try to install their own rules (such as the disdain of "camping"), even though "camping" is perfectly within the rules.

3. Objective/victory conditions.  The objective is usually to kill as much as possible before you're killed, but there can be overall game victory conditions.

4. "Data storage". (Information Management) The computer/console provides the storage and management; how the software addresses the details is usually hidden from anyone not on the production team.

5. Sequencing.  Almost always, shooters are simultaneous movement (real-time).

6. Movement/Placement.  Almost always, the player has an avatar that moves in ways analogous to the real world.  The difference can come in whether the character can jump, swim, fly, etc.

7. Information availability.  Most video games involve much hidden information-one of the great virtues of electronic games as compared to non-electronic.  In a shooter, you rarely have information that your avatar cannot reasonably see or hear, though there may be scanners or other devices that detect through walls and around corners.  (Exception: many games show you, after you're killed, where your killer was when he attacked you.)

8. Conflict resolution/interaction of game entities.  Shooting.  And perhaps melee.

9. "Economy" (resource acquisition).  In most shooters you can find food, weapons, and medical kits.  In some, when you score enough you gain additional "lives", or can purchase better weapons.  You may be able to despoil the bodies or the installations of your vanquished enemies, as well.

Let's try a simple electronic game: Pac-Man.

1. Theme/History/Story/Emotion/Image.  The game is often credited as the first to have a character and there is a story of sorts, though once again the story is mostly an excuse for action.

2. Player Interaction rules (and number of players).  One player vs. the computer.

3. Objective/victory conditions.  Make it through all the levels.

4. "Data storage". (Information Management).  The game uses a square grid, more or less, as a "board". 

5. Sequencing.  Simultaneous.

6. Movement/Placement.  The player has one "piece" which can move constantly.  The opposition has up to four ghosts, though not always all of them at once.

7. Information availability.  Virtually all information is available!

8. Conflict resolution/interaction of game entities.  Pac-Man eats dots, ghosts eat Pac-man, Pac-man can eat ghosts for a limited time after consuming special dots.

9. "Economy" (resource acquisition).  Score points to gain lives.

Next, another example. The video game Civilization IV is not much different from most board war games:

1. Theme/History/Story/Emotion/Image.  Rise from barbarism to the moon.  Conquer the world or persuade it to acknowledge your nation's superiority.

2. Player Interaction rules (and number of players).  Multiple separate interests and sides.  Negotiation is possible.

3. Objective/victory conditions.  As with some boardgames, there are multiple ways to win, such as flying to the moon/stars or conquest.

4. "Data storage". (Information Management).  Civ uses a square grid, which a player can actually make visible, to regulate movement.  The computer keeps track of many details, which of course is why Civ the computer game includes far more detail than any boardgame.

5. Sequencing.  Turn-based.

6. Movement/Placement.  One side moves all of its pieces in a turn, many pieces can be in one area at a time, move into an enemy-occupied area to attack it.

7. Information availability.  Thanks to the computer, much of the information is hidden, though Civ provides various aids and warnings to give you some idea of your standing in the world.

8. Conflict resolution/interaction of game entities.  When pieces move into an enemy-occupied area, a fight occurs.  Unlike most boardgames, the combat method involves one unit at a time on each side even though many may be in the area.

9. "Economy" (resource acquisition).  Much of Civ revolves around acquisition of resources that enable technological research and construction of a great variety of pieces.

 What about a non-conflict game? Say, Tetris.

1. Theme/History/Story/Emotion/Image.  None.

2. Player Interaction rules (and number of players).  One player vs. the computer, which probably administers things purely at random-it is not a conflicting interest.

3. Objective/victory conditions.  The objective is to score points by making rows of blocks; but the game has no ending other than ultimate failure of the player's efforts.

4. "Data storage". (Information Management).  The square-grid "board" and the computer.

5. Sequencing.  Simultaneous.

6. Movement/Placement.  The computer generates pieces, you can rotate them.

7. Information availability.  You can see what's on the board, and the type of piece that will fall next.

8. Conflict resolution/interaction of game entities.  This is as close as we come to the rules for where blocks fall and when they disappear.

9. "Economy" (resource acquisition).  The pieces keep coming.

Let's try a sports video game, say Madden (or just about any other football simulation).

1. Theme/History/Story/Emotion/Image.  Simulates real-world football.

2. Player Interaction rules (and number of players).  The player vs. the computer, ordinarily.

3. Objective/victory conditions.   The same conditions as real football; even in games involving a campaign (entire season), the objective is to win a championship, just as in the real world.

4. "Data storage". (Information Management) The computer, the virtual football field.

5. Sequencing.  Simultaneous with periods of thinking in between, just as in the real thing.

6. Movement/Placement. 11 "pieces" on a side, running, passing, causing collisions.

7. Information availability.  Largely available, but similar to the real world.

8. Conflict resolution/interaction of game entities.  Complex rules for collisions including blocking and tackling, rules for possession and movement (and loss of) the ball.

9. "Economy" (resource acquisition).   Trades, drafts, and other ways of acquiring new "pieces"; injuries.

Finally, let's try a game that may not fit, because it uses the human body only -- Rock, Paper, Scissors:

1. Theme/History/Story/Emotion/Image.  None.

2. Player Interaction rules (and number of players).  One player vs. another.

3. Objective/victory conditions.  The circular superiorities rule determines a winner.

4. "Data storage". (Information Management).  If there is any, it's the human brain, and only insofar as, if you play best two out of three, something must keep track of the score.

5. Sequencing.  Simultaneous.

6. Movement/Placement.  No pieces, nothing, really, other than your hands.

7. Information availability.  Only what you can glean from your reading of your opponent.

8. Conflict resolution/interaction of game entities.  Here we have the paper beats rock, rock beats scissors, scissors beats paper.

9. "Economy" (resource acquisition).  No new resources, but anyone with a hand can play.

Let's use this framework to quickly make big changes in a game.  Examples here are for chess (none have been playtested...):

1.  Theme.  Supposedly chess once represented real (Indian subcontinent) warfare.  But today it is an abstract game, and adding a story that actually makes a difference in the gamepkay is more than we have room for here.

2.  Players.  There are commercial chess versions for three or four players.  The board is larger and not quite square; for three players the overall shape is triangular.  It would be quite difficult to change the player parameters without changing the board . . .

3.  Victory/Objective.  First player to take at least X pieces and have two more than opponent wins the game.  Or simply, first to take X pieces.   (X to be determined by playtesting.)  Or even more unusual and less likely to degenerate into stalemate, first to take all opposing pawns wins.  In either case, checkmate of the king is still a way to win.

4.  Data storage.  3D chess exists commercially.  Or make some squares safe havens, where pieces cannot be captured (king cannot go there).  Or add one "hyperspace" connected to all of the middle 16 squares of the board.   You can move to it from any of the 16, then must stop.  You can move out to any of the 16.   Perhaps the most practical change is to treat the board as a cylinder, that is, the left side and right side are connected to one another. 

5.  Sequencing.  What would chess be like if you could move two pieces at once?  Probably white would move one, then movement would be two at a time thenceforth.

6.  Movement/placement.  There are vast numbers of "fantasy chess" variants with new pieces (and even unusual captures).  What if you could move through your own pieces (the knight can do this already)?  Or through your own pieces of lesser power only?  Bobby Fischer advocated a variant of chess in which the back-row pieces are distributed randomly at the start of the game (and mirrored for the two players, I believe). This could be regarded as a board (data storage) change as much as a movement change.

7.  Information.  The 19th century game "Kriegspiel" uses three chess sets, two players, and a referee.  Only the referee can see all the pieces, each player has a board showing only his own pieces.  The referee let a player know when one of his pieces disappears (is captured).  You can add rules for "sight distance", of course.  This is a natural for computerization (e.g.

8.  Conflict resolution.  When there's a conflict, each player rolls a die, high number wins, attacker wins ties.  Attacker also rolls one die type higher (or adds one point).  Pawns roll d4, bishop/knight d6, rook d8, queen d10 (or even d12).  Even the king has a d4, and there is no checkmate, you must actually capture, but still warn the opponent of check.

Or make it one die per level, so a pawn rolls one d6, bishop and knight two, rook three, queen four, king one.  And the attacker gets an extra die, or one extra pip per die. This variation is more practical because unusual dice are not needed.

9.  Economy.  Specify some squares on the board to be "supply centers".  If a player occupies such a square, he gets a "supply point" at intervals (every 5 moves?).  The points can be used to buy back dead pieces, using the standard point values for pieces (Queen 10 down to pawn 1).  Pieces return to play as a move, showing up in a vacant square that they would have started in.

This can be done with other traditional video and non-electronic games as an interesting exercise in game transformation.  (We need more video games that let the user actually change some of these parameters to try out their own versions.)  Use this framework to help you see things in a different light, to notice things you might not otherwise notice in your games, whether you're in conception or playtesting or modding an existing game.

Return to the web version of this article
Copyright © UBM TechWeb