Deep Plaid One guy trying to make some interesting decisions

All Games Are About Choices

Posted on September 24, 2011

Today I read a very thoughtful blog post by a game designer who I greatly admire, but with whom I absolutely disagree: Chris DeLeon wrote a scathing dismissal of the argument that games like Galaga are based on interesting decisions. (That argument was itself presented in response to Chris' previous blog post, titled "Many Games Are Not About Choices.")

I'd like to respond with an assertion: that Galaga really is a game based on interesting decisions; and that, in any game which includes anything that could possibly described as "challenge" (in other words, virtually all games), the gameplay is in fact entirely based around interesting decisions. My argument is that we should take Sid Meier's definition that "a good game is a series of interesting decisions" (which Chris dismisses as only applicable to certain types of games) and apply it in a deeper and more holistic way than it's typically applied; and that doing so will show how it is possibly the most important, fundamental law in the field of game design. Recognizing this may involve rethinking one's definition of the term "decision"; but I believe that thinking this way reveals certain fundamental truths about game design which seem to elude even many experienced game designers.

Mario's resume, like mine, is varied - though none of my jobs' descriptions have been "killing baby monkeys." Yet.

Learning the Ropes

I should provide a little background before I continue. My formal education isn't in game design, it's in software engineering; however, I've always had a great passion for game design, and several years ago I set about methodically self-educating myself in it that discipline. But for the most part, I was disappointed in the lack of rigorous academic material available - coming from a highly analytical and well-defined field like computer science, I kept feeling that there must be some hidden cache of "Game Design 101" educational materials that really explained what game design was about, but eluded me. To make a long story short, my education in game design has almost literally been a self-education - I was basically unable to ever find a "universal theory of good game design" which I found satisfactory... so I set about defining my own.

(Note that there are diamonds in the rough... in particular: virtually everything ever written by Marc LeBlanc; and most of the teaching coming from NYU's Game Center, especially the book Rules of Play by Eric Zimmerman and Katie Salen, which I'm currently reading and loving.)

Scientists spent years trying to decipher Einstein's coded Unified Theory documents before realizing they were actually chords for Rolling Stone songs.

Guitar Heroes and Unified Theories

My personal "unified theory of fun gameplay" didn't begin to crystallize until a couple of years ago. Until then, my definitions of "fun gameplay" and "good game design" were rather fuzzy and non-rigorous: various theories and definitions floated about in my head, but it was unclear how they related to one another. (I now recognize that this is pretty much the current state of game design theory in general.) One of these definitions was Sid Meier's "interesting decisions" quote, which I intuitively felt to be extremely important, though it was hard to explain why.

Then I read a blog post by Chris Bateman which directly challenged the Meier quote, holding up Guitar Hero as the ultimate proof against it:

"...these rhythm action games do not rely upon a series of interesting decisions, for the most part they have no decisions of any kind!"

I realized this was an important question: was the idea of Interesting Decisions fundamental to good game design, or was it optional and disposable?

I thought about it extensively and realized that it was the former: all good gameplay is comprised of interesting decisions ... but only if one expands one's definition (and understanding) of what a "decision" is. And once I expanded this definition, I finally found the "uniform theory of good game design" that I had sought all along.

Does decision-making break down somewhere between these genres? Also, what would happen if Princess Peach fought Kerrigan? That would be so sweet. Sorry, what was I talking about?

Who Turned Off the Choices?

I played Guitar Hero obsessively, and much like I played any other game: I'd go to a level that I hadn't completed yet, attempt to complete it, and fail. I would then try again and again until I succeeded, at which point I would move on to the next challenge. I noted that this was exactly the same pattern that I applied to a game like Advance Wars: Dual Strike. And though those two games clearly had huge differences, it was clear that there was some kind of fundamental similarity between them as well. Advance Wars (a turn-based strategy game) was clearly about making interesting decisions. But Guitar Hero wasn't... right?

But consider the following genres of game, and tell me when they stop being about "interesting decisions":

  • Turn-based strategy [Advance Wars]
  • Slow-moving real-time strategy [Kohan, Neptune's Pride]
  • Fast-moving real-time strategy [Starcraft]
  • Tactical "action" games [Defense of the Ancients]
  • Pure action games [Super Mario Bros, Galaga]
  • Rhythm action games [Guitar Hero]

At what point in this spectrum does the gameplay stop being about "interesting decisions"?

My answer: they don't stop being about interesting decisions. Each genre is fundamentally about making decisions during every moment of gameplay. There are decisions being made in every one of these games; they're just extremely different decisions, which occur in different layers of the brain.

At the bottom of the spectrum, the decisions are so minute that they're no longer what we would call "decisions" in a normal definition. In other words: the exact way you configure your fingers across the buttons to prepare for the next set of notes coming towards you in Guitar Hero is a decision that you make.

Again, this is not what we'd typically call a "decision" in day-to-day language - we might normally call it a "choice" or even just an "action." But fundamentally, they're all the same thing.

These games each use different parts of your brain. They're also both so hard that they make you want to lobotomize yourself... but each in a different part of your brain.

Fretting Over Tanks

Is there a difference between choosing what configuration my fingers are going to be in during a given millisecond-long period of Guitar Hero, and choosing what configuration my tanks are going to be in during a given turn of Advance Wars? Of course there are differences: in Advance Wars, my conscious mind is rationally considering the battlefield and making an intellectual decision; in Guitar Hero, my unconscious mind, my physical instinct, my muscle memory, and my intuition are deciding where my fingers need to be this instant, and moving them there as best they can.

But though they're happening on different levels of consciousness, they are still fundamentally the same thing. Now that we've acknowledged the differences, consider the commonalities:

  • Each are actions defined solely by my own initiative. What actions I take, and what exactly the action is comprised of, are defined entirely by myself. I never move my hand on a Guitar Hero controller without it being my decision to move it; and no one but me is deciding where my fingers are going and how they're getting there.
  • Both are always decisions which may be either "better" or "worse" than other decisions I might have made. My line of tanks could be more or less optimal for defense; the arrangement of my fingers could be more or less optimal for allowing me to hit the notes currently moving down the screen.
  • My decision-making improves as I learn. I don't just get better at Guitar Hero because I'm memorizing the level: my hand is also constantly learning better ways to move and arrange my fingers on the keys. With time, my skill increases and allows me to take on new and greater challenges.

I admit that there's a big difference between decisions that a player must make under time pressure, and decisions that the player has infinite time to make. Playing my puzzle game Connectrode (which has no time pressure) is very different from playing Dr. Mario (which does), though the games have mechanical similarities. But both types of decisions are still decisions: just because a decision has to be made within a time limit doesn't mean that it stops being a decision. They're just different flavors of decisions.

Boom, headshot. No more decisions for you.

Counter-Semantics

Essentially I'm expanding the definition of "decision" here to encompass something that happens on all levels of human consciousness. Consider a game like Counter-Strike, where within one round the player must make "strategic" decisions (what configuration he and his squad should take and what points of the level to assault with what strength); "tactical" decisions (what vectors to approach from, what hiding places to choose); and minute "action" decisions (whether to use gross-movement muscles of the arm, or fine-movement muscles of the wrist, in order to maneuver the mouse so as to place the crosshair over an enemy player's head onscreen). I think it's best to holistically view all of these as "decisions" which are made during gameplay, but which simply exist in different layers of the operations of the human brain. (For a more detailed analysis of the varied decision-making in a Counter-Strike game, read Tynan Sylvester's excellent Gamasutra feature Decision-Based Gameplay Design.)

Now, I'll admit that calling these things "decisions" does seem silly (or at least inaccurate) once we start talking about minute movements of fingers on the buttons on a plastic guitar! In regular language, no one calls what you're doing in Guitar Hero "decision-making." I would probably be better understood if I said instead: Guitar Hero tests a skill, and so does Advance Wars; and though these are very different skills, they're still both clearly skills, testing different areas of human mental (and physical) performance. But I believe that all "skills" have, fundamentally, the same "structure" - they're composed of actions.

In the end, all games that are based on an element of challenge are by definition based on testing and challenging one or more skill. (If you think that your challenge-based game isn't based on testing any player skills, then either you're wrong and you're not looking hard enough for the skill... or else you're right and your game is neither challenging nor fun.) And all skill levels are essentially defined by what decisions you're making and the quality of those decisions. As you play the game, you learn more, thereby improving your decision-making capacity - which is the same thing as saying "improving your skills".

Letting your ship get captured: The classic risk vs. reward decision. Thing is, it represents about 1% of the decisions you make in this game.

Galaga and Garrison Keillor

While playing Galaga, I definitely make decisions, at a rate of about 60 per second: I'm either pointing my ship in a direction or not, hitting the Fire button or not... every moment of action (or inaction) is my own decision. But a large number of those minute choices are made by my "lizard brain"... or my "muscle memory", or my "instincts", whatever you prefer to call it. For some reason we don't usually call such choices "decisions"; but I believe that classifying them holistically with other types of decisions clarifies their role, and their importance, in game design, and allows us to better understand and compare game designs.

And what of the decision in Galaga to allow my ship to be taken away, so that I might recover it later as a power-up? Clearly this is a higher-level, "strategic decision", and it's actually unusual and is used to break up the constant low-level "action decisions" that the gameplay is mostly comprised of. Many great games have multiple layers of decision-making, often taking place at the same time - this is an example of that.

Ultimately, that one decision in Galaga is the one that's easier to talk about (and recognize) than the many tiny "wrist decisions", because it's the one occurring at the higher level of our consciousness. But a truly far-seeing game designer is willing to acknowledge the importance of all types of decisions, which may compose all types of mental and physical skills. Garrison Keillor said "Nothing human is beneath a writer's attention." Similarly, no human capability for decision-making should be beneath a game designer's attention... from leading a civilization, to moving a finger over the correct button - and remember, the former is never possible without the latter.

Comments (4) Trackbacks (0)
  1. What a nonsense.

    To be able to make a decision, you need to have at least two options to choose from. That’s what choice is all about.

    When you touch something hot an pull back your hand, you don’t make a decision. When you learn to drive a bike, you get a feeling how to balance your weight – there’s no decision involved. When you learn to type on a keyboard, you learn the positions of the keys and get more precise in hitting them. Where’s the choice or decision in that?

    Pavlov condition dogs to salivate whenever hearing a certain sound. Hearing that sound was so intensely associated to getting food, that the dog automatically started salivating. Do you really think the dog had any kind of choice/decision in that matter?

    It’s pure training of muscles, reflexes, hand/eye coordination, precision of movement, speed of execution. All of that does not offer any options, choices and there’s nothing for you to decide there.

    I don’t think you can create such a theory without making a distinction of how a game achieves to be challenging. Guitar Hero is a purely execution based game – like Simon (http://en.wikipedia.org/wiki/Simon_%28game%29). The challenge is based around execution, and not around making decisions.

    Now the question is, are execution based challenges real games? I mean, look at sports. Is running 100m a game? No. Is swimming a game? No. Is bicycling a game? No. Is football a game? Yes. What’s the difference?
    Football encompasses decisions, which cycling does not. It’s the decisions that make a game being a game.

    So it all comes down to how you define “game” and not to how you define “decision”. Simon is an execution challenge, just like Guitar Hero is. Actually those two are very, very similar at a technical level. So either you use a strict definition of games (that specifically includes choice/decisions) – or you make a conscious distinction of challenge based vs. choice based games.

    You can’t just bend the definition of “decision” into absurdity, just to make it fit your theory. You know, the idea of a theory is to refine an adapt the theory according to what you observe. If Guitar Hero breaks your theory, you have to revise your theory. Period. You don’t go on an break what “decision” means, just so that your theory can stay untouched. That’s the wrong way to go about things – you theory can’t grow and improve if you do that.

    Now that being said, I fully agree with the definition of Sid Meier’s: “a good game is a series of interesting decisions”. I believe that to be fully true.
    Now Guitar Hero does not feature any decision whatsoever (except for choosing the song…), does that make it a bad game? You get to overcome execution challenges in a most satisfying way (much more satisfying than Simon) and you can live your rock/pop star fantasies. That’s a great package everything considered.

    But if you talk about game design, about designing game-play on a technical level… and if you don’t consider the fantasy and the execution challenge – it’s about pressing buttons in the right (preset) order and timing. Is that good technical game design? Not at all in my opinion.

    So Sid Meier is completely right – if you assume he is talking about the technical game-play design!

    And many good games combine execution skill and choice. Shooters like Counter Strike require those much-lauded mouse-twitch player skills, but they also require decisions like which weapon to pick, where to run to next, which point on the map to storm, which point to defend, whether to stay together or to split up, and many many other things. So does Galaga.

  2. This is a very well-thought-out reply.

    It seems that you’re mainly rejecting my attempt to apply the term “decision” to things that are extremely minuscule actions. But I openly admitted from the beginning that I was going to be abusing the term “decision” into referring to things that we would never usually define that way in normal language! Perhaps I should have coined a new term to avoid this confusion instead. But as I talked about in the post, I do believe that there are commonalities between “high level” strategic decisions and “low level” unconscious/instinctive decisions. And I certainly see how it can seem disingenuous to expand a theory’s definition to encompass something that its creator never thought about it encompassing (although that happens all the time, such as with interpretations of the Constitution).

    Let’s try talking about it this way. When you started learning to drive a car for the first time, you were unfamiliar with the controls. When you came up on a turn, your driving instructor probably asked you, “You’re coming up on a turn, what do you do?” And you would run through a mental checklist and say “I’m going to turn on my turn signal and then depress the brake to slow down.” Then you would do those things. Later when you had no instructor you probably still had a time where you said to yourself step-by-step: “okay, activate the turn signal… now, depress the brake…” In other words you’re making decisions about how you were driving in a very conscious and intellectual way (and also in a very slow and deliberate way).

    After thousands of hours of driving, in a familiar vehicle, do you still run through an itemized mental checklist when you come to a turn that includes “activate the turn signal” and “depress the brake”? No; at some point you stopped making a conscious decision to flip the turn signal, you do it without any conscious thought, while your conscious mind is probably thinking about something completely different. Most people would say it became “instinctive” or that it’s now handled by your “intuition” (which means you take the actions much more quickly).

    All I’m essentially saying is this: games that are about conscious decisions are not fundamentally different from games that are about “instinctive” decisions. The first time I played Super Mario Bros. (at 5 years old), my decisions of which buttons to press on the controller were very much conscious, deliberated decisions that I had to think about. (Sometimes the deliberation took long enough that the first goomba would kill me!) After hundreds of hours of playing platformers, my physical interaction with the D-pad and the jump button are no longer conscious decisions anymore – they’ve become “instinctive decisions.” But the game is still, on some level, about these “decisions” of which button I should press when – whether or not “decisions” is what you would choose to call what your brain is actually doing at that point. Playing Super Meat Boy last year was still very much about me learning more and more about making exactly those “micro-decisions”, and learning how to make those “micro-decisions” after each failure.

    Guitar Hero and Simon are certainly “acid tests” of this theory because they’re games that are about little more than putting your hand in the right place at the right time. But they’re still games, they’re still “challenge-based games” that test a player skill… and what I was looking for was a holistic theory that I could use to define and understand all challenge-based gameplay. I think that classifying all of these things as “decisions” is useful in yielding such a theory (though I haven’t taken the time to describe all the insights that this theory has yielded for me). This theory is so “unified” that it describes games like a 100m-dash race (which most certainly IS a game) as well as games like Civilization.

    You could simply think of this “Expanded Interesting Decision Theory” as another “lens” (in Jesse Schell’s terms) through which to view games: when looking at a game design, ask “where do the ‘decisions’ live?” and “are the ‘decisions’ being made in my conscious brain, or is it ‘all in the wrist’?” and “Which types of these ‘decisions’ are in my game, and which are being made where?” I believe this is important because finding where these “decisions” are in your game design reveals where the “Challenge Fun” of the game lives.

  3. But are those “wrist decisions” (the decisions that your wrist makes) really what makes games good and fun?

    This very much leads into another argument I had some time ago – which was about “player skill” and the cult around that. Have you ever noticed what amount of pride and ego some shooter gamers draw from their mouse twitch skills – commonly called “player skills”? This is pure execution – and as I said before, all better shooters encompass a mixture of execution and decision. In Counter Strike it is very important to cover the right spots on the map, and to play together as a team efficiently. Yes, you need a certain level of those twitch skills first – as a prerequisite – before you are able to get into the higher – more cerebral – aspects of play. And even the most cerebral aspects can never reach the same levels that chess or round based games can – because of their (mostly) unlimited thinking time.

    Now when shooter player change their game – go on to play a new one – they want to take their twitch skills with them. It is most important for any new game, to be able to accommodate pre-trained skills people take along from playing other games. Players actually do not want to relearn those physical execution skills. They want the game to feature new stuff in the cerebral department. New features of map-design, new decisions to make, new strategies to implement. They want new decisions. But they don’t want to retrain their twitch skills.

    Most players claim that “player skill” is all about those twitch-skills and mouse-aim. And they fully claim that round based games don’t require any skill at all. But we all know that’s just genre-fanboyism. Chess is fully round based, fully cerebral, completely free of twitch skills, and it takes a damn lot of “player skill” to be good at it.

    I even go so far to claim, that what people really love most about shooters is the cerebral stuff too. The team-work, the map-mechanics, all that stuff. They are not so keen about training the execution. They did it for the first shooter they ever played (like I did myself) – but they don’t want to repeat that process. Ever. Adding new stuff to learn on top of that is okay, but a shooter game so distinctive that players would have to relearn their executions skills from scratch would inevitably fail.

    And the same thing is true for music rhythm games. Any new rhythm game that would depart to far from the tried-and-true four-colored button formula with buttons/colors running along the screen – would fail.

    So I really don’t believe that those execution skills – those “wrist decisions” as you term them – have a comparable impact on the quality and fun of playing a game, than good choices/real decisions have.

    When a conscious decision I make in a game leads to noticeable consequences in the game, I feel empowered, I feel like it is my own actions that actually move something in the game. I feel that the decisions I make have a meaning. I feel that I have to think about my choices and pick cleverly, to be rewarded with the outcome I desire.

    Getting better in a game because training my reflexes and timing made me a more skilled player is fun and satisfying on it’s own right – but something completely different to decisions.

    Execution skill and decision making are not the same.

  4. I’m not saying they’re exactly the same thing – I’m saying that they ARE both similar in that they’re both skills.

    To me, the skill of playing a great strategy game is a player skill that can be challenged and honed; and the skill of hitting the right notes in Guitar Hero is a player skill that can be challenged and honed. Civilization is about testing one type of skill, Guitar Hero is about testing another. (Both games have other “types of fun” going on in them too of course; but they both contain “challenging fun” and I think that this type of fun has fundamental similarities across genres.)

    My skill in playing a strategy game is improved through a loop of “play, fail, learn, change my decision-making process, play again.” I feel that the same fundamental loop is taking place in improving my skill in Guitar Hero – and the only difference is that the learning and the “decision-making process” are all happening at levels that are sub-conscious. I couldn’t for the life of me explain WHAT I’m learning and precisely HOW I’m changing my technique to get better as I improve at Guitar Hero – but the fact remains that I do. (Note that I’m ignoring the “memorization” side of GH and focusing on the technique/skill aspect – getting better at the game overall, rather than getting better at one particular song.)

    The “shape” of both of these “skill/challenge loops” seems to have so many similarities that I can’t help but try to find a “unified theory” for both. I’m not a neurologist so I can’t tell you about what parts of the brain are doing “decision-making” as I play Guitar Hero notes, but it seems clear to me that “decisions” about what I’m doing are being made somewhere between my conscious mind and my fingers.

    Mostly I think you’re expressing a preference for games that test “higher reasoning skills”, over games that test moment-to-moment physical skills. That preference is fine, but I find it interesting that when looked at through the “lens” of this theory, they seem to have much more in common than a casual observer might think.


Leave a comment


No trackbacks yet.