Today's column started as a "Drive to Work" podcast. (You can listen to it here.) It got a lot of positive feedback, so I thought I'd adapt it to my column as some people prefer reading things to listening to them. There will be a few concepts in this article not on the podcast and vice versa, so if this topic interests you, you might want to check out both of them.

When I first got to Wizards of the Coast back in 1995, R&D had a series of folders in which we could discuss various subjects. We had a folder for each Magic set (as well as all the other games we were making). We had a folder for playtesting. We had a folder for new product ideas. We basically had a folder for any subject people thought R&D might want to talk about. One of the folders was mysteriously titled "Kickshaw." (For those curious where it got its name, Richard Garfield, Magic's creator, likes to name his projects randomly out of a dictionary.)

Richard had made this folder because he thought it was important to have a place to talk about the concept of games. We would discuss various aspects of games, and Richard would coin terms so that we could further discuss them. (Richard, by the way, inspired me to be more proactive about creating vocabulary for Magic design.) Kickshaw was the first time I remember discussing the larger idea of what a game is.

Back in my stand-up comedy days in college, one of the common discussions the comedians would have offstage was the definition of comedy. In order to create something, it's important for you to understand what it is you're making. Now that I was making games, I shared Richard's belief that I needed to have a working idea of what exactly constituted a game.

Today's column is my attempt at creating a definition that works for me. I'm sharing it with all of you because I think it will help other game designers figure out how they want to define games for themselves. Even if you're just a game player and have no interest in designing games, I think having a working idea of what you think a game is will shape how you think of games and help you further engage in the hobby. I want to stress that I don't believe it's important that everyone have the same definition of games (just as I didn't believe every comedian needed the same definition of comedy), but I do believe it's important for each gamer to have a definition that works for them. Today's column is about my definition.

Mark's Definition of a Game

A game is a thing with a goal (or goals), restrictions, agency, and a lack of real-world relevance.

Let me walk through each part of this definition.

A goal (or goals)

There needs to be a point to a game. What exactly are the players trying to do? If there's a way to win the game, how do you win? If there's a way it ends, how does it end? Players in a game need motivation, they need something to direct their actions. That comes from having a goal or goals. Now the goals can be active (defeat the enemy) or passive (don't die), but they must give the players some idea of what they're supposed to be doing.


Games are about obstacles. The players have a goal, but something keeps them from simply accomplishing it. A game needs to have some challenge to it because the fun of a game comes from figuring out how to overcome those challenges.


A game needs to have decisions, and those decisions must matter. Having a choice where the proper way to play is always making the same choice is not really a decision, and as such is not giving the player agency. Player involvement in the game and its outcome is core to the experience of a game.

Lacks real-world relevance

A game is something that you opt into doing because you want the experience of playing it. Labeling every obstacle you run into in life, a game quickly robs the term of any meaning. We use the expression "play a game" because it's an activity we opt into for some gain (usually entertainment and/or education, but there are many reasons one can chose to play).

Missing Pieces

To help demonstrate why these four things are important, let's examine what happens if you remove one.

Restrictions, agency, and a lack of real-world relevance, but no goal

I refer to this as a toy. Let's take LEGO as an example. There are restrictions: LEGO blocks only come in certain shapes and in certain colors. There is agency: you can connect any piece to any other piece, making any shape you want. There is lack of real-world relevance: the items you're making do not serve a functional purpose in day-to-day life. But there's no goal. There's no way to win at LEGO. You can create your own goal, build a particular shape, for instance, but that's not inherent into what LEGO is as a product. (And yes, there are sets specifically designed to build something in particular, but they give you step-by-step instructions of how to do it; there's nothing to figure out.)

My favorite game example of the separation of game and toy is Minecraft. For those unfamiliar, Minecraft is a computer game that exists in a world made up of three-dimensional blocks. You start with nothing and slowly acquire substances (often by mining them) to start building items, some of which are tools to build other items. The game exists in two modes: survival and creative.

In survival mode, monsters come out, mostly at night, and try to kill you. Your goal is to stay alive while you slowly build up your resources, which in turn help you create items (things like weapons and shelter) that help you stave off the monsters and survive. In creative mode, the monsters can't harm you (and you can even just turn them off). You have access to whatever substances you need to allow you to build whatever creation you can come up with.

By my definition, survival mode is a game and creative mode is a toy. Survival mode gives you a goal—don't die—which then drives almost every decision you make as you play. Creative mode is essentially just virtual LEGO. You have the freedom to build whatever you want, but that motivation is not driven by Minecraft; it's driven by your own imagination.

I want to stress that I'm not knocking toys. I think toys are an essential human need, even for adults (although they aren't always labeled "toys"). I'm just trying to point out that when you take away the goal, instead of creating a game, you are instead making a toy.

Goal (or goals), agency, and a lack of real-world relevance, but no restrictions

I refer to this as an activity. Let's take jogging as an example. Imagine each day you want to jog for five miles. There is a goal: run five miles. There is agency: you can choose where to run and how to run. You can decide what to wear and what items to bring along, such as a water bottle, to aid you on your run. There's no real-world relevance: you're not running to get somewhere. Usually you run in a circle so you're back where you started. But there are no restrictions. There's no particular obstacle other than willpower and the physical ability to run. It's something to do, but it's not something to solve. It's not something to work around.

My favorite example of the separation of game and activity is something I used to do as a kid: hitting a tennis ball against my garage. Growing up, I played tennis. To practice my hitting, I would put the garage door down and then take my racket and hit the ball against the garage. It would always bounce back, and I would hit again. Now, I was doing the same action of hitting the ball in a tennis game as I was hitting it against the garage door, yet the first is a game and the second is an activity. Why is that?

Well, in a game of tennis, I had a bunch of rules I had to follow. I had to keep the ball in the court. I had to get the ball over the net. Sometimes I was serving, sometimes I was receiving. I had an opponent whose position on the court might change my strategy of how I hit the ball. There was a score, and someone could win. In short, I was managing a whole bunch of restrictions while also trying to hit the ball.

When I was hitting the ball against the garage, I didn't have any restrictions. The only goal really was to just keep hitting it, and if I missed it, it didn't matter. I just retrieved the ball and started over. Both things made use of similar skills, but the context was completely different. As soon as I wasn't guided by restrictions, it became more about the form of doing the action. It was an activity.

Goal (or goals), restrictions, and a lack of real-world relevance, but no agency

I refer to this as an event. Let's take a movie as an example. It has a goal: I want to know what happens in the story. It has restrictions: I have to go to a certain place at a certain time to see it, I have to pay money, and I have to find a seat. It has no real-world relevance: it's entertainment, it's not a required thing needed to complete some day-to-day task. But there's no agency. No decision I make is going to affect how the movie is going to play. It will be the same regardless of whether I'm even there. It's something that's happening that I am witnessing, but I'm not affecting it.

My favorite example of the separation of a game and an event is the difference between playing Tic-tac-toe as an adult and playing it as a kid. For those unaware, Tic-tac-toe is a game where there's a three-by-three grid wherein two players take turns placing their pieces (one is Xs and the other is Os) attempting to get three pieces in a row, either vertically, horizontally, or diagonally. Upon playing Tic-tac-toe enough times, you start to recognize that there's a strategy that makes it impossible for the opponent to win. If both players are aware of this strategy, the game can't be won.

When kids play, before they've learned the limitations of Tic-tac-toe, it's definitely a game. They're making decisions that affect the outcome. They have total agency. Interestingly, when two adults who both know the limitations play, it turns into an event. You don't know exactly how the game will start because there are nine opening moves, but very quickly the responses become preordained. There's no surprise, and no decision you make is going to change the fact that it's going to end in a "cat's game" (Tic-tac-toe lingo for a tie). This means that something can be a game for one person while not being a game for another person. It comes down to the agency of the person playing.

This brings us to a popular topic, a favorite when the definition of a game is discussed—Candy Land. Candy Land, for those unfamiliar, is a children's "game" where the players are trying to be the first one to get to the end of a track. The board represents places in Candy Land, each referencing a different kind of sweet, that you are moving through to get to the castle of King Kandy. You advance along the track by choosing a card from a deck. The cards all have one or two squares on them of various colors (if two squares, they are always of the same color). The player then advances to the next spot on the track of that color, or the second spot if you pull a card with two squares.

There are no decisions to make in Candy Land save a shortcut that is 100% always the right call strategically. The players don't know the outcome yet because the information on the cards is hidden, but if you looked at the deck before the game began with knowledge of the number of players and their order, the outcome would be a known quantity. So, is Candy Land a game? My answer is no, but with a little caveat. If you are young enough to believe that by picking the card you are influencing the outcome, I would argue that the game has agency for you and thus, for you, it is a game.

Goal (or goals), restrictions, and agency, but no lack of real-world of relevance

I refer to this as life. Let's take packing suitcases for a plane trip. Most airlines will charge you per suitcase and will charge you extra if the bag weighs more than 50 pounds (a little under 23 kilograms). There is a goal: pack everything you need for the trip. There are restrictions: use the fewest pieces of luggage while making sure no one piece weighs more than 50 pounds. There is agency: you have total control of what you do and don't pack and what piece of luggage each item goes into. But you don't lack real-world relevance. This is not being done for entertainment or education, it's being done because you have to do it.

My favorite example of the separation of a game and a real-world-relevant item is a flight simulator game and actually flying a plane. Imagine the best possible flight simulator that realistically captures the control panel of an actual plane. It's still not the same as flying a plane. The biggest reason is there is no real-world consequence of failure. Crash in the flight simulator? Oh well, maybe next time. Crash a real plane? You're dead. The weight of consequence makes the two experiences radically different even if the actions you'd perform might be similar.

The reason this category got added to my definition was how often during conversation about this topic, we'd realized that half the things we did in life met the requirements. Life has a lot of goals and restrictions and agency, but it isn't a game, so I added this final requirement as an acknowledgement that part of playing a game is taking on a responsibility that you weren't required to take.

Overlapping Categories

When defining a game, I found that a couple other categories tended to intermix with games, so I wanted to take a moment to talk about those.


Sports are games or activities that involve a strong physical component. Some sports are games because they have all the necessary components defined above, such as football, baseball, or soccer. Other sports are activities because they lack restrictions to work around. The most common group of sport activities are things that test a particular skill but without restrictions, adding extra complications to account for. The 100-yard dash, for example, is about running as fast as you can. It's an activity, not a game.


Puzzles are a tricky category. Not all puzzles are games because they lack agency; not in the sense that you, the puzzle solver, don't make decisions, but rather that your decisions can't influence the outcome of the puzzle. A crossword puzzle, for example, doesn't change its answers based on decisions you, the puzzle solver, make. This makes them more events than games, but unlike a movie or an art show, you have a more active participation. That doesn't mean that puzzles can't also be games, but they require a dynamic solution. What that means is that you, the puzzle solver, have the ability to find different solutions based on your actions.

A good example of a game that has a puzzle component while still being a game is the video game Scribblenauts. In the game, you're given objectives to meet and then given the ability to create any item you can come up with. The first challenge, for instance, is to get a starite (basically a star that's the item you're always trying to get) out of a tree. You can make a ladder and climb up the tree. You can make wings and fly up to the starite. You can make a chainsaw and chop down the tree. It's a puzzle, but it's dynamic, meaning that there's no one singular solution but rather a series of solutions based upon all the various possible actions. This makes it a puzzle and a game.

Also, puzzles are often used as components in games, so it's possible that the singular puzzle within the game isn't itself a game, but is part of a larger structure that is a game. The Legend of Zelda games are a good example of this.

Your Turn

Now that I've spent 3,000 words defining what I think a game is, it's time for all of you to share what you think a game is. Feel free to use any piece of this column as a jumping-off point, or come up with different criteria all your own. The point of the exercise is to examine what a game means to you.

I normally ask for feedback, but as this is a topic near and dear to my heart, I'd like to hear what you think of my definition as well as hear you tell me your own. You can email me or contact me through any of my social media accounts (Twitter, Tumblr, Google+, and Instagram).

Join me next week when it's time for another Storm Scale article.

Until then, may you play many games, and toys, and activities, and events, and life.

#541: Dominaria, Part 2
#541: Dominaria, Part 2


This is the second part of a three-part series on the vision design of Dominaria.

#542: Dominaria, Part 3
#542: Dominaria, Part 3


This is the third and final part of a three-part series on the vision design of Dominaria.