Negative Sum Game - Part 1

In his last post, Chris speculated on the possibility of a game in which the protagonist doesn't win and is doomed to failure from the start. This made me curious: how would a game that a player can't win really look. 'Winning' takes on very specific meanings when it comes to video games, so it may not look the way we think it would.

First, let's split games into the two basic categories so we can determine what winning means in different contexts.

• Narrative-based games: games with a plot that a player is expected to follow. In this type of game, winning can be defined as playing the game through to it's successful conclusion.


• Round-based games: similar to board games, this type of game is based around a gameplay system that is implemented across different settings. In round based games, in order to win you must defeat other players or achieve specific objectives in a set time-frame or under certain conditions.

Now, there is some crossover between these two types of games, but most eventually fall into one camp or the other. Most modern games are narrative-based games (Gears of War to name but one), and older games were more likely to be round-based (Pacman, Pong, Battle Zone). However, some more recent games are round-based, such as most racing games, the Mario party series, and all competitive multiplayer games currently on the market.

It is very simple to determine what winning is in a round-based game, and that is to beat all the other players, be they human or computer controlled. In this context, games that are un-winnable are games that would not allow players to measure who did the best in any measurable way. This would be tremendously dull. Imagine a racing game where every player has a different track and the game never let any player know if they finished before another player. In the context of a round-based game, an un-winnable game doesn't sound like much fun.

So now we've defined what winning (and not winning) is in a round-based game, how can we define a narrative-based game that cannot be won? Well, the most basic way to lose a game is to die (provided you're not playing Prey). So does that mean in a narrative-based game that you lose the game if your character dies at the end of the story?

Not necessarily. Winning and losing are more complicated than that, because winning is as much an emotional experience as it is a hard-and-fast set of conditions. Winning also means you were successful in what you were trying to do. For example, the T-800 (the Governator) in Terminator 2 'wins' because he A) destroys the T-1000 and B) saves John Connor. However, he has to destroy himself in order to avert Judgment Day (unless you accept Terminator 3 as canon, which I don't). He dies, but he doesn't 'lose,' so even if the protagonist dies at the end of the narrative, he can still win.

What if the main character were to die, and failed the objective the set out to achieve? Now that would be losing! However, games are funny in the way they structure a narrative. Unlike in a movie, where you sit and watch things happen to a character, in a game you are in control of events and you can replay the game as many times as you wish. For these circumstances to result in anything other than 'winning' at some point, then the game itself would have to be designed to result in the failure of the protagonist. But again, we're undermined by the psychological dimension of winning, because in the absence of a traditional positive ending, gamers will feel a sense of accomplishment in having completed the game, overcome any challenges and progressed through the narrative of the game. So we're back at square one again - we've created a different ending, but the player still would feel like they won, since they achieved what they set out to do. Knowing your quest is hopeless from square one makes failure meaningless, and like I said before progress through the game becomes a surrogate for actual success.

Our trouble lies in expectation. Fiction depends to some degree on a willing suspension of disbelief, putting ourselves inside the world or the game, movie or book. In non-interactive fiction, this serves to keep us emotionally invested in the characters of the work, but in a game, where we can take control over the story, suspense isn't usually as sharp because we can react to surprises and no situation, no matter how dire, is ever hopeless because we can take control. Worst of all for narrative structure, if we die, then we jump back before the dangerous event that led to our demise. We can adjust tactics and strategy and best of all, this time we won't have been victims of the element of surprise. As we play through the game, we learn what to expect and adapt our sense of winning to our experiences. So to not win, our expectations have to be subverted. The ending, and how it unfolds, must be a surprise.

So in order to lose in a narrative-based game, the protagonist has to fail, and the player has to not know that this failure is imminent as they play through the game. Otherwise, they simply adjust their expectations to the structure of the narrative. So now, we finally have a working definition of a narrative-based game that is not winnable: a game that leads players to expect they will succeed, but is designed and plotted to result in nothing other than failure. If a player suspects this twist is coming, then the game goes from an un-winnable tragedy to a dark ending to our hero's noble quest.

In the second part of this article, I'll investigate how anyone would even make a game like this, if anyone should, and what losing can teach us about how to make games better.