I'm going to argue that every game can be described with the following four elements: action, strategy, chance, and interaction.
Some games are nearly pure action games. These games have strategy, but the biggest draw is what is going on instead of why things are going on. The foremost examples in this case are sporting events and some video games, especially shooters. I have some ideas on these types of games, but they're not my passion so we won't be covering them in detail (at least not for a while).
Strategy games are focused on conceiving and executing a plan. Some sports (notably football) certainly have elements of this, but the purest strategy games are board games and board game like computer games. Chess would be the classic example, though I prefer games like Puerto Rico, Civilization, and X-COM.
These games are my favorite kinds of games, and the kind of thing that I would try to design. John and I have worked on a few of these games, though he's done more towards actually making any of these a reality. One of his I particularly enjoy and I'll ask him for permission and post the rules at some point. Strategy games are the games that are almost entirely about choices, which is why I'm most interested in them.
Games of chance are just that. Nearly all games have some element of chance, whether it be the die roll in Monopoly, the tiles drawn in Puerto Rico, the cards that come in poker, the way a ball rolls in football, etc. Chess is the obvious exception to the rule here. Meanwhile at the other end of the spectrum some games are purely chance with the only choice being how to bet (think Roulette or the slots). The key to these games from a design standpoint is to make winning appear more likely than it actually is. This requires a healthy amount of probability calculation and is not very interesting from a design standpoint. Instead what we want to draw from chance is a limited mechanic: one which makes outcomes uncertain, but also one which generally rewards good choices.
Then there are games of interactivity. I'm going to call these games "social" from now on because it's easier to understand. These games focus their attention on how one interacts with their fellow players. The current most prominent example of this is probably World of Warcraft (though it obviously has elements of chance, strategy, and action it is predominantly a social game). But there are other games that have prominent interactive or social aspects. A good example of this is poker, where knowing probability is hugely helpful, but having some social skill can get you a long way. Nearly all "party games" fall into this category as well for obvious reasons. Games like Apples to Apples, Loaded Questions, or Balderdash are all about who you're playing with to be fun.
One note: simulations (Sims, Sim City, Flight Simulator, Spore) are ignored here because they are more "toy" than "game" though some of them fall reasonably close to strategy games.
The question of choices comes up in all four game types to varying degrees. Choice in pure games of chance is about WHAT to do. Choice in action games is both about WHAT to do and HOW to do it. Choices in strategy games are about WHAT to do, HOW to do it, WHY you should, and WHERE these choices are taking you (Basically, is there a plan? There should be). And finally social games all of the above take place but you have to add the question of WHO are you playing with?
Social games are the toughest to play and the hardest to design effectively because the number of choices you're asking the player to make are large.
Making things a little more complicated is that almost no game is a pure game type. So things aren't easy to categorize. For example, knowing an opposing football coach and how he thinks in the NFL is a crucial element to winning games. Or in Risk if you become known as a backstabber, it's hard to make alliances and you become more likely to be targeted earlier in future games. But fundamentally Risk is a game of chance with strategic and social aspects. The NFL is an action game with strategic and social aspects. (Texas Hold 'Em) Poker is a strategy game heavily influenced by social and chance elements. Other variants of poker have more chance elements (think Draw) or social elements (Blind Man's Bluff).
But all games are fundamentally predicated on having those four elements. Obviously I'm most interested in the games that are fundamentally about strategy. It's the math geek in me. But that's for next time when we ask the seemingly simple question of "why are choices the key to a good game?" and maybe (we'll see how I feel) the more interesting question of "what does this tell us about other topics like the inevitability of history?"
