Unwinnable Games

From gdp3
Revision as of 08:27, 11 October 2022 by Staffan Björk (Talk | contribs) (Using the pattern)

Jump to: navigation, search

Games that do not have any winning conditions.

Some games are not possible to win regardless of how well one performs. This is because the systems they use have no state defined as winning or no way to get there. Such games can still pose interesting challenges to players either by letting players could their own goals and winning conditions or by letting players compare their performance with previous performances by themselves or others.

Carse divides human activities into Finite Games and Infinite Games in his book with the same name[1], and simplified his view of Infinite Games are Unwinnable Games that are played with the goal of continuing to play the game.


Early computer games such as Asteroids and Space Invaders were Unwinnable Games in that gamers were constantly given new obstacles as soon as they overcame their current ones. Pac-Man is designed in the same fashion but the original version has a software bug making it allegedly impossible to complete its 256th level[1], making it open for interpretation if reaching the highest possible level in a game equals winning over it. The casual games Staries and Icy Tower are later examples of games that have no winning conditions.

Roleplaying games such as Dungeons & Dragons and GURPS do not have general winning conditions. Individual campaign might be winnable but the gameplay of improving one's character can continue and may likely be in the players interest. The 4th Edition of Dungeons and Dragons have structures of retiring characters when they reach level 30 (through gaining some form of immortality) but not even this needs to be seen as having completely beating the game since gameplay can continue with a new character. While simply surviving Paranoia scenarios with at least one of one's clones alive can be considered victories, the publishers of the roleplaying game published Code 7 missions that required one more clone that players have.

Massively multiplayer online games such as World of Warcraft and Eve Online have similar structures but add challenges of fighting other players, accessing the best available equipment and gaining achievements.

September 12th and Hell Tetris are examples of games that make the realization that they are unwinnable one of the main experiences of playing them.

Using the pattern

Games may be unwinnable to specific people because of lack of time, interest, or competence. This is however more a question of the players' possibilities than the design of a game, and when it is an effect of the design (rather than say the people not being the targeted audience) it is often seen as a problem with that design. Unwinnable Games in contrast are games whose system is design so that their is no winning condition and no players regardless of skills and devotion can achieve a certain game state designated as the winning one. Both Multiplayer Games and Single-Player Games can be Unwinnable Games.

Three main - and compatible - ways of creating Unwinnable Games are Ever Increasing Difficulty, Continuous Goals, and Grinding. Ever Increasing Difficulty ensure that players at some point will lose due to not having enough competence (given that the game system can provide challenges beyond any player's competence). Continuous Goals that are necessary for continued gameplay rely on players at some time failing in competence or attention to a goal (e.g., a Survive goal for a player's Focus Loci). Grinding that lets players continuously do actions for little or no progress in the game, relying primarily on players losing interest in the gameplay and quitting for being unwinnable.

There are two typical reasons for creating Unwinnable Games. The first reason is that one may wish to reinforce the Replayability by not making it possible to ever beat the game completely. In this case there exists enforced end conditions, e.g. by strict Time Limits or Steadily Decreasing Resources and to make it possible to compare ones performance with earlier game session there typically also exists Trans-Game Information (e.g. Achievements or High Score Lists) so that Meta Games are possible. The second reason is to create Never Ending Stories, that is, games without any end condition at all. Common in Roleplaying games, these games cannot have a winning condition for the whole game but instead needs to provide incentives to play in other ways. This is typically done through offering smaller local goals, either Ephemeral Goals or Player-Defined Goals, or simple through the Social Interaction offered by the game or play activity. An issue with non-ending games is how to balance increased Gameplay Mastery and Improved Abilities with challenges so that the Challenging Gameplay is maintained. Red Queen Dilemmas is a possible solution to this by adding the difficulty as players gain more power to influence the game. However, non-ending games are incompatible with Time Limited Game Instances but this may not be a problem if Game Pauses are used since then the game and play sessions can be regulated instead.

While Unwinnable Games cannot end due to completing the game, they can have Game Termination Penalties and Game Over through dying, in fact this is common in many early examples of using the pattern.

One design possibility that is more suitable to Unwinnable Games than others is that of Entitled Players; it is less problematic to trust players with powers affecting how the game is played if they have less reason to misuse that power.


Every game state in an Unwinnable Game is a Unwinnable Game State. At first glance it can seem strange that one can have or would want games which cannot be won since it removes what is typically the reason for activities to become meaningful in them. While other reasons for playing games exists, many games can however be the basis for Meta Games through Trans-Game Information (such as Achievements or High Score Lists) or Progress Indicators and can then still provide meaningful gameplay concerned with measuring how well one has performed. When Unwinnable Games allow this kind of gameplay, they somewhat paradoxically promote Replayability. Sometimes realizing the unwinnable nature of a game is an intention in the design; the Serious Game September 12th uses the impossibility of winning it to argue the futility in the struggle thematized by the game and the (earlier) movie War Games appropriates Tic-Tac-Toe to make a similar argument. Providing Illusionary Rewards can also make parts of Unwinnable Games meaningful since players may appreciate these challenges even if they are not affecting the game state.

When players notice that games are unwinnable they may see this as revealing an Exaggerated Perception of Influence in the game design. This may cause them to stop playing (which in a sense was the intention with the serious games mentioned in the previous paragraph since this would make the point that one should stop with the activity portrayed) but may also cause players to set up their own Player-Defined Goals that they consider win (and end) conditions. Regardless of when and why players stop playing Unwinnable Games, they do so by Surrendering (to the game design).

Although Narration Structures can support players the gameplay structure in Unwinnable Games, the narrative aspects are likely to not attract interest unless they change in significant ways between game sessions. More specifically, Predetermined Story Structures are difficult to have in Unwinnable Games since their content is limited and this may be exhausted by sufficiently long gameplay and Main Quests are in direct conflict with Unwinnable Games since completing the Main Quests are ways of winning the game even if gameplay continues afterwards.

Unwinnable Games may be seen as a rather easy way to avoid having a Predictable Winner at any point in a game, but this is not strictly true. First, the Predictable Winner may be nobody as the Tic-Tac-Toe example shows and second players may set up their own winning conditions between themselves, e.g. who can get the highest Score in a game of Pac-Man (so Scores can be used to modulate Unwinnable Games). Even so, Unwinnable Games can be used as part of a design strategy to make players less likely to identify any specific player as being able to reach the winning conditions set up by a game.

Multiplayer Games which are also unwinnable, e.g. most Tabletop Roleplaying Games, will need to deal with Early Leaving Players since this will occur in all game instances except the ones where all players simultaneously stop playing.

For rather obvious reasons, i.e. having no winner, Unwinnable Games are incompatible with Winner determined after Gameplay Ends and Winning by Ending Gameplay. It is also not possible for Unwinnable Games to have Endgame phases. Unwinnable Games can only appear to have Higher-Level Closures as Gameplay Progresses to those that do not know that the games are Unwinnable Games, so the patterns are generally incompatible.


Can Instantiate

Never Ending Stories, Replayability, Surrendering, Unwinnable Game States

with Multiplayer Games

Early Leaving Players

Can Modulate

Multiplayer Games, Single-Player Games

Can Be Instantiated By

Continuous Goals, Ever Increasing Difficulty, Grinding, Time Limits, Steadily Decreasing Resources

Can Be Modulated By

Achievements, Entitled Players, Ephemeral Goals, Game Pauses, Game Over, Game Termination Penalties, High Score Lists, Illusionary Rewards, Meta Games, Player-Defined Goals, Progress Indicators, Red Queen Dilemmas, Scores, Trans-Game Information

Possible Closure Effects


Potentially Conflicting With

Challenging Gameplay, Endgame, Exaggerated Perception of Influence, Higher-Level Closures as Gameplay Progresses, Main Quests, Predetermined Story Structures, Predictable Winner, Time Limited Game Instances, Winner determined after Gameplay Ends, Winning by Ending Gameplay


New pattern created in this wiki.


  1. . Carse, J.P. 1986. Finite and Infinite Games. Penguin Books.