Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 31, 2014
arrowPress Releases
October 31, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

Games That Can't Be Won
by Altug Isigan on 11/06/11 03:36:00 pm   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.



There are a lot of games that can't be won. All we can earn ourselves in those games is a honorable spot in the high scores list.  Examples are plenty, but if we must name a few, there are Tetris, Centipede and Space İnvaders.

In discussions on whether games are stories or not, such games have often been given as examples in order to argue that games can't be stories. However, the argument is flawed, and this article tries to explain why.

The Protagonist Takes It All?

One assumption that leads to this flawed argument is that stories are always solved in favor of the protagonist. In other words, stories are pictured by game researchers as if they'd always be "won". Inescapably leading into defeat, non-winnable games draw a completely different picture. This makes it easier to claim that games must be very different from stories. 

However, there are a lot of stories that haven't been "won" by their protagonists. Examples that come in mind are movies like Braveheart and Seven. So, the assumption that stories are always "won" by the protagonist proves to be wrong.

What Does Losing Really Mean?

But how come that a game or a story still makes sense despite a defeat of the protagonist? Or despite our prior knowlegde that we can never solve the problem in our own favor? After all, the Titanic will eventually sink...

Interestingly, neither games nor movies of that type seem to feel incomplete. In fact, they often make a great experience.

The answer to this lies in the relation between plot and climbing tension: A plot is build on conflict, that is, clash of interest between two opposing forces. The tension will keep climbing until one of the opposing forces is eliminated. The elimination of one of the forces brings a resolution to the conflict. The climbing tension comes to a halt, and the 'drama' is over. Even if the protagonist has been defeated, the story itself is being 'complete'.

Resolution Trumps Protagonist

What confuses people is that they perceive non-winnability as the game having no end or resolution because it can't be won. They tend to interpret this as some sort of open-endedness, which is wrong. There is a fine line to this, and we can't afford to overlook it: defeat *is* a valid solution to a conflict, hence there is nothing wrong with a non-winnable game. It is still a completely valid story-structure: After all, a resolution may or may not be in favor of the protagonist. From a plot perspective it doesn't matter, because what counts is that the conflict has been solved: Resolution trumps the protagonist.

In other words: The defeat that we as protagonists eventually face in a non-winnable game still brings an end to conflict and resolves the plot, hence it is completely valid as a resolution.


Non-winnability is not necessarily an indicator for absence of story. Games that can't be won are still stories, but stories that never solve their conflict in favor of the protagonist.

The reason why we still consider these games as a complete experience is the fact that our defeat meant that the conflict has been solved, and that the story has been rounded up.

So, please enter your initials and remember this quote from Samuel Beckett:" So you've failed? Fail again. Fail Better."

Related Jobs

Magic Leap, Inc.
Magic Leap, Inc. — Wellington, New Zealand

Level Designer
Grover Gaming
Grover Gaming — Greenville, North Carolina, United States

3D Generalist / Artist
Demiurge Studios, Inc.
Demiurge Studios, Inc. — Cambridge, Massachusetts, United States

Lead System Designer
DeNA Studios Canada
DeNA Studios Canada — Vancouver, British Columbia, Canada

Analytical Game Designer


Aleksander Adamkiewicz
profile image
I'd be interested in hearing your opinion about E.Y.E. Divine Cybermancy, a recent representative, on this matter. Its relevant in the sense that the developers created a game with no ending (which upset many people).

They created a paradox where the gameplay becomes meta in the sense that quitting the game, i.e. not playing, is the only winning move in the context of its story.

Its the first time I've actually seen quitting a game to be treated as a game-mechanic.

I found this genius. A lot of people disagreed...

Luis Guimaraes
profile image
Metal Gear Solid 3 too. No spoilers though.

Altug Isigan
profile image
Hi Aleksander,

I haven't played the game you've mentioned, but from what you tell about it, it seems to take a very interesting approach to (non-)winnability.

I think it differs from the non-winnability I talk here in the sense that refusal of dealing with the conflict is a valid resolution to it. And a plot that in a *meaningful* way forces you to give up to deal with it, is always an interesting situation, because it is not the type of resolution that you come accross every day. I think it's very difficult to set up a plot in which people may quit a game as a resolution and not as an expression of frustration. That seems to be the genius that you talk about.

The game seems to employ withdrawal as a resolution, to use the exact term. Now that may have many shades, and since I don't know the game, I don't know how the quitting felt to you. It could have felt tragic, for instance. Or it could hav been fun too, once you realize that this is the trick to "beat" the game. It also reminds me of some similar mechanics in games, where you can refuse to carry out a move when you realize that it would lead to defeat. Or where you can declare that you will no longer follow up your interest when you realize you can't reach your goal. Strange endings, but not non-valid ones.

Such endings risk to feel weak. Because in terms of story design these type of endings seem to violate a few basic rules. First of all, they resolve the story not really in a way that feels like a climax. I mean with that, that a resolution is usually expected to come after a final big battle that decides the winner, and withdrawal is often not a desired solution, since it would feel like we have been kept there for nothing. That is like we wait for the final duel between the sherif and the bad guy, and the bad guy suddenly says he doesn't want to be a bad guy anymore. Bummer! Second, it violates the rule that the resolution must be fought out and shown. But here the player may chose to not fight it out but just accept defeat. Again, something that would often be considered to weaken the story. Third, it violates the rule that under normal circumstances a character always choses to do what is good for the story, i.e. does things that keeps it moving forward to the climax. But here, the choice to not do that is apparently one way to solve the problem.

While all this rule violation thing may sounds confusing, I think what really counts is this: Did it feel like an end to you? Did it feel like it resolved the problem you were dealing with? If quitting felt like an answer to the problem posed by the game, then it is still a resolution and a valid end to the story in this game. Just like the rule in graphic design that a line remains a line as long as perceived as such. In my opinion, in the game you mention, the quitting does not come *after* the end, nor is it a refusal to play out of frustration. It *is* the end, born out of the causality of the game structure, hence, it is part of the game. Hence I would not hesitate to call it a valid resolution to the game in question. And I think I understand why you call it genius, because it is... :)

Aleksander Adamkiewicz
profile image
I see what you mean in the distinction between non-winnability and withdrawal as resolution.

In a sense this kind of approach brings with it a lot of risk. People are too used to having an obvious ending spelled out for them (for both situations).

E.Y.E. has a lot of problems in itself, especially the delivery of the narrative, if you want to know more, I wrote a complete review on it here:

Eric Schwarz
profile image
Not to mention that many games tend to outright ignore inaction as a valid approach to a conflict. At best, it means that you'll walk away without a reward, and there might be some sort of later follow-up (more likely in RPGs, but still uncommon). At worst, it means that you fail, and can't move forward in the game at all. Players have been trained for a long time that all solutions have problems and that all outcomes are either positive or negative; when a game subverts those expectations, of course they're going to be taken aback by it.

I suppose all I can suggest is that, at a broad level, more developers take a closer look at what it means to "win" and "lose", and how those end states can be more effectively framed in a way which doesn't imply a binary opposition. Planescape: Torment is one of the only games I've ever seen take this approach, with a wide spectrum of outcomes and moral relativity running through all of them - people can suffer or succeed by your hand, but the game never passes judgement on you as a player, so long as you make headway in the story. Why it's still not standard practice to study it for understanding interactive narrative, I'll never know.

Altug Isigan
profile image
Thanks for the question that gave me something to ponder about, and also thanks for the link. I'll check it out!

edit: Btw, remembered the movie Devil's Advocate as I re-read through our conversation. Didn't Keanu beat the devil by just refusing to play the game out? Also when you think of the resolution in Seven, all that Brad Pitt had to do to "win" was to resist the temptation to play out the game that the serial killer had set up (which was very difficult to be honest, considering that he knew that his pregnant wife had been beheaded).

Daneel Filimonov
profile image
I'd like to chime in and add an example of a game that can't be won.

Dwarf Fortress.

Anyone who's played Dwarf Fortress will tell you that winning is not part of the game. In fact, it's quite the opposite. It's rare to come across a game (even more so, an indie game) that focuses on losing rather than winning. In fact, the word "lose" is synonymous with the word "fun" for many people who play Dwarf Fortress because that is essentially what you do. Lose.

I'd probably categorize (as much as I hate categorizing games) these types of games as simulations (or even "Timebomb" simulations) because you never really win (or lose, in some cases).

Altug Isigan
profile image
Thanks Daneel!

What I like to add here is that most non-winnable games also provide a denouement (or epilogue) that follows the defeat. The typical example here are high score lists. Despite the impossibility of winning, you still find a chance to assess your performance and attach a meaning to it, which is often crucial in rounding up a story, because it allows you to assess and express what it meant to you to go through this adventure. The conflict is solved, and the protagonist has a last chance to make a final comment on what the experience has changed in his personality: this clearly follows rules in storytelling in regard to character growth and change.

It's funny to realize that these kind of endings have a very old tradition. The myth about Jason and the Argonauts had as a conclusion that what mattered in the end was not to reach the final destination, but that the goal was the voyage itself. I think this notion applies to non-winnable games.

Timo Naskali
profile image
My thoughts on the subject:

A game is not something tangible. A game is a subjective experience that is only born out of the activity of playing. Every player experiences a different game when they interact inside the virtual space created by a video game sofware.

From this follows that every game "ends" when and only when the player stops playing. Whether or not the text "The End" or "Achievement Unlocked: Completionist" is printed on your monitor has nothing to do with it. You can never actually win any game in a definite sense, but you can feel like you did.

A story is similarly a subjective experience, but not one equal to the game, it's a separate one. Every story experienced through gameplay comes to an end when and only when a player feels like it has come to an end, when they get a sense of closure and decide they're satisfied. For some people who watched Reservoir Dogs or read Batman: The Killing Joke the story never came to an end - others found a sense of closure.

So, in my opinion games are not and cannot be stories, but stories certainly can be experienced through playing; just like the words on a book don't make a story but reading them might invoke the feeling of one.

Bart Stewart
profile image
There are variations on this question, some of which are definitional.

There are story games (or, at least, games with stories in them) that end in what would be considered non-traditional ways in Western culture. STALKER: Shadow of Chernobyl comes to mind here; the ending makes perfect dramatic/literary sense considering its source, but Western players may conclude that the game is unwinnable. But even if a game's story is a Kobayashi Maru, it's still a story, and possibly one that we may find gripping or satisfying even if we don't win.

(Side note on this: Braveheart may not be the best example of a fail. Although William Wallace was indeed captured and executed by the English, his example and that of Robert Bruce -- especially after Bannockburn -- were the inspirations that preserved Scottish independence. Of course, the further story was one of frittering away that independence through squabbling and scheming, so in that sense you might say that Wallace "lost." And yet it was the Bruce's descendant, James VI of Scotland, who became King James I of England after Elizabeth I. So in the end, maybe William Wallace won the game after all.)

The other way of looking at non-winnability is what might be better called unwinnability... except that this isn't necessarily always a Bad Thing. James P. Carse wrote an interesting book called _Finite and Infinite Games_, the point of which was this: a finite game is played to be ended, and necessarily has both winners and losers, while an infinite game is won when all players cooperate to keep it from ending.

For a lot of core gamers (and probably some designers) that latter notion will seem completely nuts. How can something even qualify as a "game" if it doesn't end? But I think the existence of MMORPGs as persistent worlds is an answer to that objection. In a way, by coming back to a gameworld, players are participating in an infinite game: as long as they keep playing (and paying), there's no reason why the game would ever end.

Extending that concept to a single-player game, where the journey -- not reaching a destination -- is the goal, is something I'd really like to do. In the meantime, I'd like to hear more thoughts on the value of unwinnability.

Altug Isigan
profile image
Thanks a lot for this very mind-opening comment, Dear Sire :)

I can really add nothing to it, except that I need to say that it gives me a lot of stuff for consideration. Unwinnability is something that got me curious and I will have a look into Mr. Carse's book as soon as I can find it.

Also the MMPORG example is a good one. Interestingly enough, it makes me think of Shehrazad's storytelling trickery in 1001 Nights...

Oh, one last thing: Technically speaking, I would say that the plot of Braveheart is being solved. While we know that historically things didn't stay there, I believe that as a piece of drama, it is complete in itself. That's why I think it is not bad as an example for resolution that was reached through the defeat of the protagonist.

Wylie Garvin
profile image
Extra Credits got this one right... MISSILE COMMAND !

Regional defense during a nuclear war, is the ultimate game that can't be won.

Altug Isigan
profile image
We have a saying in our country that goes like "A beaten wrestler is always hungry for more wrestling". :)

Bring on more nukes, dude, you'll see I can beat 'em with my regional defense! ;)

Altug Isigan
profile image
Hi Timo,

I think I need to clarify something in regard to the "games aren't stories" thing. It was back in the early ludology days when some game researchers gave non-winnable games as examples of a lack of narrativity in games. For them, it meant that games shouldn't be studied from a story or narrative perspective, because obviously, games did things differently: "See, how some can't be won? So there isn't anything storyish in them."

The problem for me here was, and still is, that their argument is flawed because they did not have a proper understanding of drama theory and narratology. If they would have had, they wouldn't have come up with such weak and outrightly wrong arguments. I still am of the believe that game studies will fail to produce the right concepts if they do not improve their understanding in regard to drama theory, narratolology, and story in general.

Now, of course, it is a problematic thing to say straight away that games *are* stories. At some point we must accept that there is a categorical difference between the two. However, for me that doesn't mean I can't approach games from a drama theory perspective. In fact, I've learned a lot about games by trying to cover them with terms borrowed from drama theory and narratology.

Coming back to your comment: I do not agree with you that there is just one "end" to games: namely the end that happens when the player stops playing. I'd say it is important to understand how the end was conditioned: Is it the end of the game as it was designed and foreseen by the author? Was it the end we were asked to reach? Or one of the possible ends within the horizon of the design? Or did the player put an end to it because of frustration?

I think we must distinguish between two realms: the world of the game, and the real world. Quitting out of frustration doesn't count as an ending in dramaturgic terms. I think it would be wise to maintain the distinction, otherwise it can get really confusing.

Timo Naskali
profile image

Ah, okay, thanks for elaborating on your point. I'm completely with you there that games can be "storyish" (narrative-driven) and benefit a lot from drama theory etc. I've never understood the reasoning of people who disagree with this either.

I still stand by my statement that a game ends when a player stops playing it, however it is based on certain semantical premises. It seems our views are quite similar on a practical level, we just approach the subject in different ways.

My preference to think of these things in this way derives from how I define games, and from wanting to think of player agency as something that is an integral part of a game's narrative. I find this way of thinking better encompasses games of all different types, including games with branching narratives, MMOs, tag or bouncing a ball against a wall. I just find it a better fit for the interactive medium of games. I don't think a path a story takes during a gaming session is any less valid if the author never realized it was possible, thus I don't see why it's important that the definition of game excludes these. And if a player can reach a satisfying end without it being one of those the authors had thought of, I don't see why that should be excluded either.

"I think we must distinguish between two realms: the world of the game, and the real world. Quitting out of frustration doesn't count as an ending in dramaturgic terms. I think it would be wise to maintain the distinction, otherwise it can get really confusing."

My definition is not in conflict with that because I hold there to be a distinction between game and story. Quitting a game out of frustration can be the end of that game if the player never resumes playing, but then the story will probably never feel like it reached its end.

I don't see why it's important to distinguish between reality and how a game was meant to be experienced; the reality is all that matters in my book.

Altug Isigan
profile image
Hello again Timo,

I think the point where we agree lies in my reference to the graphic design principle that says a line is a line as long as it perceived as such. You say something similar about endings, and I think it shares the same ground of reasoning. The perception and interpretation of the player is of course important, but I would add to this that we might be wrong in assuming that such different perceptions and interpretation were not calculated by the designer. I think we cannot completely ignore authorial intention, even though if the player is enrolled to make large parts of the game/story happen with his actions.

Where we seem to fall apart is our approach to the relationship between game and story. I think your distinction between them draws a bolder line, but to me the line is blurred at best. This makes us see things from different perspectives (which isn't bad at all).

When a player quits out of frustration, I believe that this is a non-ludic response, it is a withdrawal from the game contract as a whole, back into the real world, and out of the magic circle. Whereas a withdrawal from pursuing the conflict as a way to solve it is still part of the story. When you withdraw in that way, you are still honoring the contract, that is you are still at play, and your decision is one that still counts within the boundaries of the magic circle. That is why I think it is important to not confuse an exit to the real world with an exit from the conflict. They are not the same, if you ask me.

Timo Naskali
profile image
You are right in that in colloquial language it's natural to say stuff like "I only got halfway through the game before I stopped playing" - I just find that kind of a paradoxical thing to say on a philosophical level and considering that strictly speaking games are usually considered an activity. But perhaps I'm just splitting hairs and should embrace the informal/customary usage so others will understand what the hell I'm talking about :P.

Do you draw a distinction between plot and story in game? I'd be more comfortable with equating plot with game experience and I think of story as a slightly higher level concept.

By the way, even though I see a game's plot as something inherently subjective, I don't mean to suggest I don't respect the designers that make that experience a possibility for the player. I may not give *definitional* value to could-have-beens (designers' intentions), but that doesn't mean I don't appreciate finely crafted game dynamics and such.

Altug Isigan
profile image
Yes I do draw a distinction and I think it is close to yours. To me, plot is a number of functions and operations carried out by the author to set up a discourse that the audience is been taken through to be informed about a story. In other words it is the way the content is presented to us through narration. Hence I agree with you that this is largely the game experience, what we go through in play-time, so to say. Story on the other hand means to me the mental picture of the story as a whole, a picture that we can only build in retrospect, based on how the discourse has chosen to inform us about the story.

But to go into more detail, to me a game experience is a constant updating process in which we reconstruct the story-so-far, based on the informations and confirmations of those informations that the discourse-so-far has revealed to us. So this is a living process in which we switch back and forth between story and discourse; a process containing all sort of activity, be that mental or physical. Without the divide, no player could derive joy from the playing, and no author could derive joy from setting up a plot. It's the divide that makes it all possible.

To come back to our question in regard to game/story endings: there are many ways in which that can happen: An end can be simply just the moment in which the discourse is brought to a halt. That can for instance happen because the player decides the discourse is too boring and the story too senseless and he pulls out at any point of the run-time of the discourse: he doesn't follow any longer his interest in the mental picture of the story, and also not into the functions and operations that the author decided to employ. Or it could simply happen because mama says you have to get to bed. But the kind of end that we are really talking about here is not that one: We speak of an end here in which the player is interested in completing the mental picture of the full-story and is also satisfied about the way the author employed functions and operations. It is the kind of end in which we feel that the plot that entertained us somehow to come until that point has been resolved. We talk about an ending in dramaturgic terms here, which we arrive at just short before the discourse as a whole is finished, and that allows us to draw a final mental picture about the story, and that also allows us to assess what both, the final mental picture of the story, and the experience we went through to capture it eventually, meant to us. I consider non-winnability, unwinnability or various methods of withdrawal as endings that are belonging under this category. They are dramaturgic endings that are carried out when we are still within the magic circle.

(Oh boy, that was a difficult passage to write :P)

Regarding your last paragraph: It is difficult to capture the relationship between author, text and reader, and I don't think we are the first to have a debate about it, and maybe this isn't anyway about which aspect is the decisive or central one, but rather about how we can put this relationship in a way that satisfies reader, writer and text-oriented approaches. I don't believe in too player-centric views, but also not into too author-centric or text-centric ones. As the gulf of games and narratives keeps streaming, theory will always fall victim to exceptions, and at the end, it is probably again theory that benefits from it.

Timo Naskali
profile image
Well put! Now that all the confusion over terminology has been cleared, I find what you write very agreeable.

Good job with the definitions for plot and story, they resonate well with my (more poorly defined) ideas of them.

Interestingly, while the necessary data for a sensation of dramaturgic closure is exchanged within the magic circle, conversely the actual feeling of closure can only be interpreted outside of that circle. The player can't realize that the game's story has concluded until he takes a step back from it and sees the story's connection to something larger. If the story doesn't resonate with reality outside the magic circle in any way, no sense of closure can be achieved - one will just be left wondering "what was the point?"

That's probably a healthy philosophy you describe at the end of your post. While laying some solid theoretical groundwork to build upon can be helpful, one should never become so protective of those foundations as to blind himself to exceptions and hamper progression.

Things should be made as simple as possible, but no simpler.

Altug Isigan
profile image
Yes, I think I agree with you that the ultimate picture can only be drawn once the discourse is over and we're sent out of the magic circle with that picture in our hand (+ our impressions about how the discourse was constructed in order to give us that picture). Probably that's what we take with us when we go to friends and tell them the story, but say "you must play to understand, it's done wonderfully, I can't really explain".

It's kinda interesting to realize that we can distinguish between many closures within and outside the magic circle, which makes it more and more intriguing to think about it, and to try to remember games, movies and novels that have this richness of endings and closures... It might me well worth to make a diagram or something that would capture the possibilies and map them (as accurate as that may be done).

You say a wise sentence at the end of your comment, and I take that to heart :)

Thanks a lot for the very constructive conversation, Timo, I hope we can do that anytime soon!

Timo Naskali
profile image
Thank *you*! It was a very thought-provoking discussion indeed.

(P.S. I have to give a little bit of the credit to Albert Einstein for coining that wise phrase)

Pablo Mera
profile image
Hi, Altug.

Finally made some time to read your article. It was very interesting, as I too have been thinking about this for a long time.

You probably have seen this, but please take a look at You Only Live Once:

I use this game to talk about "death" in videogames, but it is also related to "games that can't be won" and how the player faces this scenario. The short answer is: he has no option, he literally only lives once.

I agree that a game that can't be won complies to narrative structures, but you have to be careful how you manage that. Some time ago we were being pitched an idea for a game (a kind of infinite runner) where a character was rushing to a machine to put a final piece to make it work. You could not win, ever. When the player lost, whenever that was, the character would trip just next to the machine. This person thought it would be funny, so it took a lot of effort to convince him it would be frustrating. The story would be resolved, but always in such a way as to be infuriating to the player.

Anyway, nice reading. Thanks. :-)

Altug Isigan
profile image
Hey Pablo! :)

You're welcome my friend, thanks for taking the timeto read.

Uhm yeah, I was thinking about the game you describe and I think it's not a good idea to infuriate the player, it's very difficult to be able to make use of such infuriation to the favor of the game and increase attachment through it. That's probably what only real pro's can do :) Design-wise I think it is a problem related to not being able to put yourself into the player's shoes I guess. Maybe it was a cloven designer thing, and it looks like the evil side of the designer was at work, I don't know.

Pablo Mera
profile image
Well, I think it was more about inexperience in design, as well as, like you said, don't getting into the player's shoes. Explaining in your game's narrative how is it that you can't win (or how quitting is a valid narrative option) it's also very difficult, and most examples are very incomplete... but from time to time there is a game that makes it work somehow. The rest of the time, we must trust players won't care that much if they never get to "win", as long as they are having fun. :-)

Altug Isigan
profile image
For the sinful game designer, fun is like god; merciful and forgiving ;))