Wednesday, December 9, 2009

Abstract Game Criteria, Revisited

All games worth discussing for their ludic merits are strategy games. All such games involve planned set of actions that players attempt to execute in order to achieve some kind of goal set by the game through the use of a set of rules. The genre distinctions used in the media are primarily differences in the interface the player uses to participate in the game. First-person shooters, RPGs, puzzle games, and Poker are all games of strategy—they just use different ways of expressing the aspects of strategies and the effects of strategies. Because all games are strategy games, one end of the spectrum should be those games considered more “purely” games of strategy: abstract strategy games.

Why are abstract strategy games considered the purest? Because they most emphasize

  • Player agency. The players should have a direct say in what happens within the game. Game results should be the results of the players’ actions alone.
  • Mechanic-based Strategizing. The only strategies should be those that operate within the context of the game mechanics. In this way, the strategy-space is uncluttered by external factors that change independent of the game mechanics. The game mechanics themselves are all that need to be considered in order to invent a successful strategy.

There are three factors that can detract from these emphases:

  • Imperfect Information. If information knowledge is asymmetric within the game, players do not have sufficient evidence to strategize successfully while maintaining full agency.
  • Chance. Elements of chance reduce player agency by removing the guarantee that a simple strategic action (like moving a piece in chess) will complete. If there is a chance that an action will fail independent of player choice and stategy, player agency is damaged.
  • Non-dyadic play. If more than two people are playing a game, politics seeps into any successful strategy. Politics is outside of the mechanics of the game, and therefore violates the mechanic-dependence of the strategy-space. (The word “dyad” is awesome, too.)

Sid Meier’s Civilization as a Concrete Strategy Game

In the beginning of a game of Civilization, I have no knowledge of the world beyond the borders of my one city. How do I chose my strategy going forward? I chose a strategy here that will have profound effects on the rest of the game, yet I’m severely lacking in information about the world and the other players. Over the next 30 turns, I will make decisions that could entirely doom me without my knowledge, even if I have a perfect understanding of how the game mechanics work. If I choose to tech rapidly and forgo upgrading my military and three squares into the fog of war is Tokugawa with 4 archers that are twice as strong as my strongest military unit (of which I only have 2, one that is acting as a scout), he will declare war on me and destroy me as soon as he finds me. I cannot make informed strategic decisions because of imperfect information in this example and it has a significant effect on the strategic outlook of the game.

Later in that game, I build up a military that consists primarily of knights, a quite strong unit of the medieval era. I go to war with a Catherine, who has been expanding too fast and hasn’t defended her frontier cities well. I have a stack of three knights invading her frontier. My strategy is to grab the high-value cities on her frontier where she has not had time to muster a defense. At the first city, I pit a knight against her measly defenses: a warrior. A knight has triple the attack of the unit she has defending this city (even modified), so I gladly send my knight to crush the city.

But my knight dies and the warrior is unscratched.

What should be a surefire victory has turned into a miserable and costly defeat due to nothing within my power as a player. Chance has stripped my agency.

I’m not suggesting that Civ games should have deterministic combat, I’m simply showing how chance takes away from the strategic agency of the player. Chance can be used appropriately and to good effect in games, but it’s important to understand that it has a deleterious effect on player agency.

Back to the game. Let’s say that I’m playing Civ over a network with a few friends. As the game progresses and we start to encounter one another in the game world, we begin to talk in private conversations about what we should do. Players begin lying to one another about what they have and what they plan to do. Strategies are formed in conjunction and with reference to the relationships players have outside of the game (though the relationships are in reference to the game). The weaker players form an alliance against the strongest player and crush him. Politics, not game mechanics, lead to the defeat of she who was formerly the strongest player. The player with the most skill and knowledge of the game, due to no failing of her manipulation of game mechanics, has been defeated. The strategic space that effects the game expanded beyond the mechanics into the realm of politics—the game mechanics do not mediate the whole scope of gameplay anymore.

Civilization is in two ways a concrete game: Its mechanics are given meaning through their representation of real world phenomena, and its mechanics contradict the “pure” mechanics outlined here and in my previous post on abstract games. Civ is not, though, a “perfectly” or “purely” concrete game because it’s not a particularly good simulation, but it is certainly on the concrete side of the spectrum, well away from games like Chess and Go.

1 comment:

Brian 'Psychochild' Green said...

A bit late on this comment; I was waiting for anyone else to comment before derailing the discussion a bit here.

What should be a surefire victory has turned into a miserable and costly defeat due to nothing within my power as a player. Chance has stripped my agency.

Yes, but this also stops the positive feedback loop of someone getting ahead in development and being unable to be stopped. This is the reason why you wouldn't want deterministic combat, as you said.

I also think it's a bit strong to say it has completely stripped you of your agency. You still have some agency in deciding to attack or not, and evaluating the odds to see if a specific choice fits with y our goals. Sometimes there is an upset, but the result depends on choices you made. If you funneled everything into making a single super-unit to conquer the world and lose it in an early battle, you're going to be worse off than if you established a strong economic base first and then cranked out better units than your enemy.

Without the random element the game mechanics would actually reduce the player's choices in the games as the plaeyr would have to get an advantage faster than the opponent. Again, this would be a positive feedback loop, since a superior unit could wipe out enemies and conquer their cities easier.

There is also the issue of game limitations and designer intent. Adding a persistent hit point system might have been beyond the technical ability of computers at the time the original Civilization was released. The designers might also have wanted aggressive combat to feel somewhat random, making it a more risky choice than alternatives such as building up a strong economy. The risk can add to the enjoyment of some players who find pleasure in elements of gambling as well.

Some thoughts to consider. I do agree that too much random can reduce the fun of a player if their choices seem to have little or no effect on the outcome.