Play & Game Design

A game with too many evaluable options stalls, because the cost of deciding exceeds the value of deciding well.

Designers treat it as a structural fault rather than a player fault: a game that permits a twenty-minute turn has been designed to permit it. The remedies are hidden information, simultaneous turns and timers, all of which cut the tree the player is searching.

In practice

• A four-player game with one deliberate player becomes a game for one player and three spectators.
• A committee given every option and no constraint meets again.

Attributed to

Board game design

Also called

AP; turn stall

Further reading

Wikipedia

Page for this entry

Successive additions to a game are stronger than what preceded them, because new content must be attractive enough to adopt.

Each release must give players a reason to buy or engage, and the cheapest reason is strength. The cumulative result is that early content becomes unplayable and the design's original balance is unrecoverable, which is why long-running games resort to rotation or resets.

In practice

• A card set that sells must outclass the last one, and four sets later the first is unplayable.
• A job ladder that adds a senior title every two years has inflated the ladder rather than lengthened it.

Attributed to

Game design and collectible game economics

Also called

Escalation; content inflation

Page for this entry

In free-to-play games a very small fraction of players produces most revenue, so design optimises for that fraction.

Reported paying rates in free-to-play titles are commonly low single-digit percentages of players, with spend concentrated far more sharply still within that group. The design consequence is direct: the game is tuned for the people who spend, and the experience of everyone else is partly an input to that.

In practice

• A game whose median player pays nothing is not designed for the median player.
• A service whose revenue comes from a small minority will be shaped by that minority's preferences, whatever its user numbers say.

Attributed to

Free-to-play game economics

Also called

Whales; the paying minority

Page for this entry

Players of shared worlds divide into those who pursue achievement, exploration, socialising and dominance, in a mix a design can shift.

Bartle's 1996 paper analysed players of multi-user dungeons and argued the four types exist in a balance the design maintains, so that encouraging one suppresses another. The typology is still used in design and monetisation, and it was never intended as a personality theory.

In practice

• A world that rewards dominance loses its socialisers, and the population that remains looks like a design choice.
• A team survey that sorts staff into four types has produced four labels, not four kinds of person.

Attributed to

Richard Bartle

Also called

Achievers, explorers, socialisers, killers

Possible explanations

The explanation is that a shared world offers four kinds of satisfaction, acting on the world, exploring it, acting on players and interacting with them, and that players sort themselves by which they pursue.

In its favour

The scheme is genuinely useful in design, and the claim that the types hold each other in balance, so that encouraging one suppresses another, is specific enough to be checked against a live game.

Against it

It comes from observing one genre of text-based world rather than from psychometric work, and later attempts to validate the four types have not supported them as distinct categories.

Standing

Derived from observation of one genre of text-based world rather than from psychometric work, and later attempts to validate the four types as distinct factors have had limited success.

Page for this entry

Play divides into competition, chance, simulation and vertigo, and most games mix them in a determinable ratio.

Caillois's 1958 classification extended Huizinga: agon is contest of skill, alea submission to chance, mimicry the taking of a role, ilinx the pursuit of disorientation. The scheme is useful because the mix predicts the audience: a game that shifts from agon to alea will lose the players who came for the first.

In practice

• A strategy game that adds random rewards trades one audience for another and is surprised by the reaction.
• A festival that is contest, lottery, costume and fairground ride is four attractions sharing a field.

Attributed to

Roger Caillois

Also called

Agon, alea, mimicry, ilinx

Page for this entry

Simple rules interacting produce play the designer never authored, and often the play people remember.

A designed sequence gives every player the same experience, and a system of interacting rules gives each player a different one. Emergence is cheaper to produce and impossible to guarantee, which is the trade: the memorable episode is generated rather than written, and so is the broken one.

In practice

• A physics system meant for puzzles is used to skip the puzzles, and the story is better than the intended route.
• A workplace policy with three simple rules produces behaviour nobody drafted, in both directions.

Attributed to

Game design

Also called

Emergence over progression

Further reading

Wikipedia

Page for this entry

A participant who can no longer win retains the power to decide who does, and answers to nothing.

Common in coalition politics and multi-party negotiation. The kingmaker has no stake in the outcome's quality and every incentive to extract, which is why third-party leverage often exceeds third-party support.

In practice

• A third-placed party with 4% of seats decides which of the other two governs and extracts terms neither would have offered.
• A departmental vote where one abstaining director determines the outcome gives them leverage no argument on the merits would have won.

Attributed to

Game design literature

Standing

A recurring pathology in coalition politics and multi-party negotiation.

Further reading

Wikipedia

Page for this entry

Fun is the sensation of a brain acquiring a pattern, so a game stops being fun when there is nothing left to learn.

Koster's 2004 argument accounts for why games become boring at the point of mastery, why difficulty must keep pace with skill, and why a solved game is dead. It explains a real pattern in play and generalises further than its evidence.

In practice

• A puzzle game loses its player the moment the trick is understood, whatever content remains.
• A task that is neither learnable nor familiar is not fun in either direction, which the account handles poorly.

Attributed to

Raph Koster

Also called

Fun is learning; the pattern-mastery account

Possible explanations

The explanation offered is learning: fun is the feeling of a brain acquiring a pattern, so interest lasts while patterns remain and ends at mastery.

In its favour

It accounts for why games become boring at mastery, why difficulty must track skill, and why a solved game is dead, which designers find operationally useful.

Against it

It is a designer's account rather than a tested theory, it does not explain forms of play with little to master, such as social or narrative play, and it treats learning as the only source of enjoyment.

Standing

A designer's account rather than a tested psychological theory; it does not explain forms of play with little to master, and it treats a family of pleasures as one.

Page for this entry

A game's story and its rules can assert opposite things, and the rules are more persuasive.

Hocking coined the term in 2007 about a game whose narrative condemned selfishness while its mechanics rewarded it. The general point is that a system teaches what it rewards regardless of what its text says, and the player learns from the reward.

In practice

• A firm whose values statement praises collaboration and whose bonus ranks individuals has told its staff which document is real.
• A safety campaign runs beside a schedule that only works if the shortcuts are taken.

Attributed to

Clint Hocking

Also called

Story-mechanic conflict

Further reading

Wikipedia

Page for this entry

Positive feedback in a contest makes an early lead self-reinforcing, ending the contest long before it formally concludes.

Game design literature. The institutional version is cumulative advantage: an early lead in funding, attention or share compounds, so the contest is decided long before it formally ends.

In practice

• A grant scheme that funds applicants with the strongest track record produces, within five rounds, a field in which only the funded have one.
• A league where the richest club buys the best players and earns the most from winning has a contest that is decided structurally rather than annually.

Attributed to

Game design literature

Standing

The design-side statement of cumulative advantage.

Page for this entry

Play happens inside a bounded space where ordinary consequences are suspended and the game's rules hold instead.

Huizinga's Homo Ludens described play as occurring within temporary worlds with their own rules. Salen and Zimmerman named the boundary the magic circle. The concept explains why a tackle inside the match is not assault, and its limits are exactly where games meet money and identity.

In practice

• A bluff at the poker table is play and the same lie about the stakes is fraud.
• An abusive message inside a game is defended as part of the game, which is where the circle stops being a useful idea.

Attributed to

Johan Huizinga; term developed by Salen & Zimmerman

Also called

The play space

Possible explanations

The explanation is convention: players agree to treat certain acts as meaning something different inside the game, which is what allows a tackle, a bluff or a betrayal to be played without ordinary consequence.

In its favour

It explains something real about play, and it remains useful in game design for thinking about what a rule set licenses.

Against it

Game studies contests the boundary, since money, reputation and harassment cross it freely, and treating the circle as sealed has been used to excuse behaviour that harms people outside the game.

Standing

Contested in game studies: the boundary is porous, since real money, reputation and harassment cross it freely, and treating it as sealed has been used to dismiss genuine harm inside games.

Further reading

Wikipedia

Page for this entry

A designer builds mechanics, a player experiences aesthetics, and the dynamics in between are what neither controls directly.

The framework's value is the direction of travel: designers work from mechanics forward, players encounter the result backwards, and the gap between them is where design fails. It also insists that aesthetics is a plural, listing several distinct pleasures rather than treating fun as one thing.

In practice

• A rule added to encourage cooperation produces an alliance that ends the game early, and the rule was working.
• A policy is written as a mechanic and experienced as an insult, and nobody in either place sees the middle.

Attributed to

Hunicke, LeBlanc & Zubek

Also called

Mechanics, dynamics, aesthetics

Further reading

Wikipedia

Page for this entry

Play is shaped by what everyone expects everyone else to do, so the strategic environment is a moving object the rules do not describe.

The rules are fixed and the correct play is not, because the correct play is a response to the current distribution of opponents' choices. A strategy strong against the popular one becomes popular and is then countered, which produces cycles that no reading of the rulebook would predict.

In practice

• A deck that wins a tournament is beaten the following month by decks built to beat it.
• An interview technique that works becomes known, is prepared for, and stops working.

Attributed to

Competitive game communities

Also called

The meta; the game about the game

Further reading

Wikipedia

Page for this entry

A game is a series of interesting decisions, and a decision is interesting only when no option dominates.

The definition doubles as a test: if one option is always right, or all options are equivalent, the decision is not interesting and the game has a dead stretch. It gives a designer something to check for, which most definitions of games do not.

In practice

• A build order that is optimal in every game has removed the decision it appeared to offer.
• A form whose fields have one sensible answer each should have been filled in already.

Attributed to

Attributed to Sid Meier

Also called

Meier's definition of a game

Page for this entry

A good game is trivial to begin and impossible to exhaust.

Bushnell's rule for arcade machines, where a player who could not start within seconds would walk away and one who could finish would stop paying. The constraint produced a design pattern that outlived the coin slot: shallow entry, deep ceiling, and the whole difficulty range visible from the first minute.

In practice

• A product whose first five minutes require a manual has lost the users who would have stayed for years.
• A tool that is fully understood in a week is abandoned in a month.

Attributed to

Nolan Bushnell

Also called

Easy to learn, difficult to master

Further reading

Wikipedia

Page for this entry

Deliberate negative feedback that keeps a contest live, at the cost of weakening the link between performance and reward.

Rubber-banding in racing games, the NFL draft order, and progressive taxation are the same device. The trade-off is unavoidable: keeping the contest live necessarily weakens the link between merit and reward.

In practice

• A league that redistributes broadcast money keeps twelve clubs competitive and dulls the reward for building the best squad.
• A grading curve that lifts the bottom of the class keeps the room engaged and makes the top grade mean less.

Attributed to

Game design literature

Also called

Rubber banding

Standing

The design analogue of redistribution, with the same trade-off.

Page for this entry

Given the chance, players will optimise the fun out of a game, so the designer must protect them from doing so.

Players pursue the winning strategy rather than the enjoyable one, and will grind a tedious optimum for hours if it is optimal. The design response is to remove the dominant strategy rather than to ask players not to use it, since asking has never worked.

In practice

• A quest with a repeatable reward is repeated until it is unbearable, by people who chose to.
• A bonus scheme with an exploitable clause is exploited by staff who would have described themselves as conscientious.

Attributed to

Soren Johnson

Also called

Players optimise the fun out of a game

Standing

Compare Goodhart's law, which describes the same behaviour from the measurement side.

Page for this entry