We talk about 'games' a lot.. but at the core what is a 'game' really? It turns out that defining 'games' is not an easy thing to do. People have been trying to define the term for (probably) as long as games have been in existence, and the emergence of 'video games' has greatly increased the interest in defining games in general.

Experts from many different disciplines have over the years attempted to define what it takes for something to be a 'game' and to describe the most basic requirements of a 'game', but invariably each definition is different. For example, some definitions try to isolate 'games' from 'puzzles', while others try to isolate 'games' from 'sports', other definitions try to isolate 'single-player' activities from 'multi-player' activities, and still others try to isolate 'cooperative' activities from 'competitive' activities.

Why should we care about a definition?! Well, I believe that how we define games has a significant impact on how we design games, and further that this definition has interesting implications for video games in general and MMOs in particular.

While I don't mean to dismiss or discount the definitions put forward by many notable game designers, or claim that my own list of requirements for an activity to be considered a game is the 'correct' one, I'm going to suggest that there are five basic requirements for an activity to be considered a 'game', and rely on a French Sociologist by the name of Roger Caillois for some of these requirements. His 1958 book, Man, Play, and Games (sometimes translated as Games and Men) is still considered one of the definitive works on games, despite being written long before video games even existed.

Just to preface these requirements.. many of these basic requirements have just as much of a cognitive component (they apply to the mindset of the 'player') as a physical component. I'll explain more in-depth for each requirement, but just keep in mind that some of the requirements to something being a 'game' might have to do with your own outlook on the activity.

Requirement #1: Governed by Rules - the activity has rules that are different from everyday life.

This is the one requirement that essentially everyone agrees on. All games have rules, and operating within and around these rules forms the core of the game experience. Game rules are so critical to the game experience that they can be the source of a lot of conflict and animosity. My own memories of elementary school recess are filled with days when the pick-up soccer, football, wall-ball, four-square, etc. game devolved into a heated argument (sometimes even a physical one) over the rules, and I'm sure I'm not alone in this.

Requirement #2 - Uncertain - the outcome of the activity is unforeseeable.

In any game, there is a fair amount of randomness and unpredictability. In a game of soccer, you don't know what the final score will be before starting the activity, or even if your team will score at all. In a game of Chess, you don't know who will Check Mate the opposing King or how it will happen, or even if it will happen at all (Chess can also end in a draw, either by insufficient material to accomplish a Check Mate or by triple repetition of position).

This requirement eliminates many activities that we would call 'puzzles'. Examples of puzzles includes activities like jigsaw puzzles, crosswords, and Sudoku. In each of these activities, you know up-front precisely how the activity will turn out.

Why is this important?! Well, without getting too much into what motivates a person to play a 'game' (a topic for a future article, hopefully), I'll just mention that what motivates a person to play a game is somewhat different than what motivates a person to do a puzzle. Take the player-vs-environment 'raiding' of many MMOs.. many of the most-vocal critics of PvE activities make the claim that raiding isn't a game at all, but a big puzzle, and whether you agree with that statement or not may help determine if you prefer player-vs-player activities over player-vs-environment ones. This requirement also has interesting implications for video game replay. As the player becomes more familiar with a particular video game (e.g. they've 'beaten' the game), the 'outcome' of that activity becomes more and more certain, and in the mind of the player that 'game' gradually shifts toward being a 'puzzle' instead.

Requirement #3 - Separate - the activity is circumscribed in time and place.

This requirement is about the when, the where, and the how long.. the constraints of the game environment itself. Many games have specific time or score limits, or objectives or circumstances which end the game, the point being that players need to know 'at the end' who wins, and when that 'end' is going to happen.

Why is this important?! This requirement has interesting implications for online and other long-term games, where an official 'end' is effectively nonexistent. At a basic level, this requirement is about giving the player 'benchmarks', moments where they can know that they were successful. MMOs often accomplish this through boss fights, achievements, online character tracking and comparison sites, etc.

Requirement #4 - Controllable - the outcome of the activity can be influenced.

As my favorite video game design commentator Daniel Floyd would say, game design is all about designing 'decisions'. Decisions, whether expressed as 'problems' or 'choices', give the player control over the outcome of the game. How much control a player has typically comes down to the mental, physical, or mechanical rules that govern that game. How well can you hit a baseball with a wooden bat? How well can you count cards or anticipate your opponent's next move? How long have you spent leveling up your character and acquiring better gear?

'Games of Chance' (i.e. gambling) are an interesting example of this requirement, and a good illustration of the cognitive component of games. I'm going to suggest that people who enjoy games of 'pure' (i.e. true random) chance treat these activities as games because they do believe they have some measure of control over the outcome: luck. For the rest of us, we'll either engage in games of chance that are not 'pure' (e.g. Blackjack or Poker) or we won't engage the activity at all because it's not a game to us.

Why is this important?! There are two aspects of this requirement that are especially pertinent to MMOs. First, video games often use a variety of methods for allowing players to influence the outcome of events, for example gear and levels, but one of the more difficult-to-manage methods for allowing players to influence the outcome of events is the 'skill factor'. Game players often want to see a direct correlation between their personal skill level and their level of control over the game events, and some games and activities emphasize or de-emphasize this 'skill factor'. Second, MMOs often require a fair amount of teamwork and working with a group of players. Players interested in personal control over the outcome of game events often dislike having to give up some of their control by being forced to rely on a group.

Requirement #5 - Goal-Directed - the activity has some desired outcome.

Whether it's to defeat the opponent, get the highest score possible, or conquer the world, all games have desired outcomes. For some games, the desired outcome is the same as the 'end' condition (e.g. Check Mate the opposing King), but in a lot of games there are many possible goals a player might work toward. For example, take a game like Baseball. While the overall goal is to score more runs than the opposing team, your personal goal might be to hit a home-run.. or to just get a single.. or to catch that pop fly coming your way.

Why is this important?! Video games in particular are often very good at giving the player lots of potential goals, and this goal-forming process is very important in attracting the player to the game and getting them to keep playing. Helping the player select and pursue goals is a critical aspect of good game design. A player who gets involved in a game without a goal in mind is very unlikely to enjoy the experience, or to stick with it for very long. That activity just isn't fun for that person.. it's not a game to them.

Why does any of this matter?! The really important point to all of this is that if the above requirements are accurate, we can directly alter an activity to become a game (or not become a game anymore) by adjusting a few parameters and helping the player cognitively accept the activity as a game. For example, take the activity of flipping a quarter (i.e. heads or tails). Clearly, this activity is not a game. Let's look at why not:

  • The activity is governed by rules, of sorts. We'll be flipping the quarter by hand, and when the coin comes to rest we must accept the results of the flip (we can't decide we don't like the results and turn the coin over, for example).
  • The activity is semi-uncertain. While we don't know the exact result of each flip, we do know that in a true random environment the coin will land heads 50% of the time and tails 50% of the time.
  • The activity is not circumscribed. We haven't yet established an 'end' point or win condition to the activity.
  • The activity is not usually considered controllable. Barring slipping in a fake two-headed coin, most people would say this activity is pretty close to true random.. after all it's used to 'fairly' decide on who goes first in a lot of games.
  • The player has no goal in mind for this activity. Most people would probably consider this activity to be mind-numbingly boring.
With that in mind, let's look at how we might adjust the activity to meet all of the five requirements:
  • To make the activity circumscribed, we need to give it an 'end' condition. As examples, this could be to only flip the quarter x times, or it could be to flip the quarter until you got heads y times in a row.
  • The player actually has more control over this activity than they might realize. For example, you could suggest that the player adjust the position of the coin on their hand before flipping it, or how high in the air or how fast they flip the coin, or how soon they catch the coin. While it's not a lot of control, the key to this activity is convincing the player that they do have some control over the outcome.
  • Let's say we want the goal of the activity to be to get the highest ratio of heads to tails. As mentioned above, the important issue here is not that the activity has a goal, but getting the player to want to achieve that goal. How would I get the player to take on this goal? I'd probably use some kind of reward for 'winning'. How does $50.00 for getting the most heads in 50 quarter flips sound? Ready.. set.. go!
Certainly video game design is a lot more complex than the above activity, but examining how effectively video games meet these basic 'game' requirements could be a valuable tool for increasing a game's replay value and enticing a broader spectrum of game players.