Galaga and Making Interesting Decisions

Sep 23, 2011

I keep up with incoming links to this blog (thanks to everyone that shares links in – I notice!), and this morning I caught a discussion about one of last month’s entries, Many Games Are Not About Choices. A counter-argument came up that I suspect may be a common reaction among readers, and I would like to address it here.

As I’ve expressed dismay over before, many game studies focus on either turn-based strategy (especially board or tabletop game design) or film-like storytelling. Both of these are perfectly great endeavors and respectable game types. However the tendency to emphasize those styles in literature and workshops often leads to confusion in the design domain of the fast action tacit skill game types that I prefer.

To paraphrase the argument: one reader suggested that real-time arcade action games like Namco’s Galaga really are about making interesting decisions after all, on the grounds that the player gets different points from different targets, and can deliberately lose a ship to a tractor beam for a chance to get a double ship (shown starting at 0:40 in the video below, in case you’re less familiar with the game).

My issue with framing Galaga in this way, as a game about making interesting decisions, is that this approach could classify nearly any activity as making interesting decisions. To name a few: karate, riding a bicycle, playing hacky sack, or batting in baseball.

Yes, there is a wide range of possible outcomes, and the player has enough control over the situation to affect those outcomes. However:

1. The outcomes are largely differences of degree rather than differences in kind.

To borrow the baseball batter example: there is clearly a difference between bunting vs swinging for the fences. That represents a difference in kind, and can easily be recognized as an act of decision making. Practically speaking, though, the outcome generally has more to do with how successfully someone swings for the fences, since it’s fairly well understood that in most cases that action has far more potential for success. That – how well/far someone bats a pitch – is a difference in degree. Most people would not identify the difference between a skilled batter and an unskilled batter as one doing a better job of making an interesting decision.

In Galaga, you’re trying to shoot all the enemy ships while dodging their shots and kamikaze dives. A successful player performs those fundamental activities more successfully than an unsuccessful player.

2. The time/space-scale involved is a finely grained continuum acted in under pressure, such that conditioned reflexes and practiced skills are leaned upon as shortcuts to perform in real-time, compensating for lack of idle time to make optimal, or even well-reasoned, choices.

Someone reviewing a video of their performance in a karate tournament, given a chance to pause, rewind, and switch to slow-mo, could very likely point to each and every thing that they should have done but didn’t. That intellectual understanding of the decisions does not help the fighter in real-time, when they are under considerable pressure and their opponent’s leg is coming at them.

Again distinguishing skilled Galaga play from unskilled Galaga play: the differences may come down to dodging a few more (or fewer) pixels in one direction or the other, firing 150 ms earlier or later, or reacting by sliding left instead of sliding right when there are a half dozen projectiles and ships coming at the player. Arguing that we can classify as an interesting decision the act of moving a few pixels further, firing a reaction-time interval later, or dodging the opposite direction with a moment’s notice would border on absurdity.

3. The risk/reward probabilities involved are fuzzy and constantly in flux, such that the outcome is more contingent upon executional ability, practiced coordination, and momentary concentration.

The differences in player ‘decisions’ during play of this type are snap judgements that might be retrospectively explained as, “can I pull it off? am I confident enough to try X right now?” – not as decision making ability (“what should I do? what are the probabilities to consider for making an informed decision?”).

If someone wishes to ride a bike but has not yet learned how to, the problem is not a misunderstanding of what they’re trying to do, but simply doing it. If you can already ride a bike, think about performing a backflip off a bike ramp. Either way is the same story: it’s easy to picture, the concept is clear, it’s just hard to do and requires practice. Likewise for keeping a hacky sack going. Of course everyone knows to not let it hit the ground. Doing it consistently is the difficult part, and central to the activity.

In Galaga, the player may have a crystal clear idea of exactly what they intend to do, but panic and fail to make the move on time, or instinctively cower and dodge the wrong direction, or simply fail to time their firing action in a way that will collide with a target. As discussed in my older entry Real-Time Play: Tactical Patterns, in many of these games the player will wind up so overwhelmed by the sheer amount of simultaneous movement and danger that they’ll resort to moving in a pattern while firing blindly, iteratively refining or changing-up their patterns between lives in a search for a rhythm that works best. It would be a stretch to regard that type of adaptive search process as the player making any interesting decisions.

The presence of a witty one-liner in an otherwise serious film does not mean we ought to consider that film a comedy. How we categorize media artifacts may seem like a matter only of interest to archivists, but categorization really affects everything from development targets to user expectation and critical reception. The hint of a subconscious tactical decision in Galaga – or, for example, the freedom to switch between guns in Doom – does not mean it’s a game about making interesting decisions, any more than the rescuing of a princess makes Super Mario Bros a story game.

Intellectualizing real-time gameplay activities as interesting decision-making misses what’s at the forefront for a player during gameplay, and can lead to nonsensical or misdirected design discourse. It’s sometimes a result of thinking about one’s memories of a game, instead of playing the game and paying close attention to the experience of really playing it. It partly arises from overconfidence in recall, trusting the brain to emulate the experience (it can’t).

To reiterate, then: many games are not about making interesting choices. Some games are. Games are not better or worse based on whether they are about making interesting choices, though in many cases these elements appeal to two different types of players – or at least two different desires/moods, sometimes both residing within the same player(s). Attempting to redefine terms to maintain compatibility with a popular sound bite is not productive, as doing so can only lead to confusion about game design, rather than improved clarity.

I think that one reason people sometimes feel compelled to force their favorite classic (or own) games into a mangled redefinition of “decision” is from confused thinking that to be good, we must somehow explain how the game is about making interesting decisions. And who could blame them, since the oft-circulated quote by Sid Meier that spawned the previous article in question is, “A [good] game is a series of interesting choices” (Rollings & Morris 2000, p. 38).

Declaring that a particular game is about making interesting decisions is not an evaluative statement asserting its quality. Declaring that some other game is not about making interesting decisions is not a slam on the game’s quality or worth. It’s similar to saying that some movies are serious, some are funny, and some movies succeed in blending both (which is not necessarily superior to either in isolation!) – it’s only identifying a difference in kind, not a difference of degree.

Learn and practice team game development with Gamkedo Club.
Membership worldwide. Professional support. Proven process.

Subscribe by e-mail to receive weekly updates with Gamkedo.Community interviews and YouTube training videos for game developers!


  1. […] Follow-up: Galaga and Making Interesting Decisions […]

  2. mccrimmonmd says:

    Well said!

    I wonder if there might be some sort of relationship between whether or not a game is rules-based and how much importance it places on making interesting decisions. The converse would be games that emphasize more law-like systems (as in the laws of physics), where the importance is instead on learning interesting skills/behaviors. What do you think?

    • Chris DeLeon says:

      I certainly think that there is a connection between these two ideas (turn-oriented/decision vs real-time/dexterity, connecting to human rule enforcement and automatic/continual action structure respectively). Thank you for calling attention to this!

      I perhaps have not been very explicit about this, but I have similar motivations for writing about both, that being to better recognize and make sense of these game types that I find under recognized or generally misrepresented in much of the existing game design writing. To better address this topic in a suitable way would (will) require another entry though, which in response to your reply here I’ll be putting on my todo list for an upcoming post. Thanks again!

  3. […] 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 interest…. (That argument was itself presented in response to Chris' previous blog post, titled "Many Games […]

  4. IQpierce says:

    Interesting take – I’m afraid I wholly disagree however!

    A couple of years ago I decided that taking a broader approach to the “interesting decisions” definition led me to the best overall understanding of game design (or at least challenge-based games’ design).

    This inspired me to finally write up my own theory in more detail, my reply is on my own blog:

    Thanks for inspiring me to put my own thoughts in writing; I’d definitely be interested in your reply to my own position!

    • Chris DeLeon says:

      Thank you for taking the time to reply! I respect you as a developer as well.

      From what you’ve written, it’s not actually clear that you disagree with my point. You are acknowledging very clearly that much of what happens during gameplay is unlike what anyone refers to as “making decisions.” However you then insist that we refer to these different things as making decisions, too, for reasons which I did not find stated clearly, except that they happen in different parts of the brain (but this is true for everything in the realm of human experience and action, right?). It seems at least partly to be to preserve the perceived sanctity of Meier’s statement, which I acknowledge as a natural reflex but nonetheless erroneous justification for overgeneralizing the meaning of decision making to include everything someone does (from my own list: “karate, riding a bicycle, playing hacky sack, or batting in baseball”). You alternate between admitting that these real-time reaction activities are very different, then insisting that they are the same thing. If they were, you would not need to make such a distinction between why they’re different, besides of which you and I would not be able to be so obviously clear in defining exactly what it is we’re referring to that are unlike what people casually refer to as decisions.

      As for simply redefining how a term is used: surely it’s understandable why, when my target is an oft-repeated isolated quote and not an extended written work, simply attaching an unconventional definition to that quote will not suffice?

      > As you play the game, you learn more, thereby improving your decision-making capacity – which is the same thing as saying “improving your skills”.

      This is, to borrow from my reply to zoviri’s comment, the distinction between a coach’s explicit knowledge and an athlete’s tacit knowledge. The coach may be excellent at knowing what decisions to make, but that doesn’t mean they’re top-level athletes – this is true even for the coaches and trainers of world class athletes. By comparison, in many sports activities and real-time videogames, if the player has to make decisions about what to do, when, and how, at an executional level (besides a few spread out strategic choices), they will almost universally lose to someone that has drilled the correct reactions until they became automatic.

      Would you argue that the difference between the swing of various professional golfers is that they are making interesting decisions differently, in terms of when to contract various muscles in particular sequence?

      > I definitely make decisions, at a rate of about 60 per second

      Any decision that you claim can be made in 17 ms (roughly 1/10 typical reaction time), I maintain is an arbitrary overextension of the term to call it a decision, and thus a mistake to design it as though it’s a decision, or study it as though it’s a decision.

      Your post brings up plenty of worthwhile topics for discussion, and valid points that warrant further investigation. It’s not so much an argument with what I’ve written here, as it is for the most part a fine example of studying and discussing those types of videogames and real-time mechanics that I’m making an effort to recognize are different than making interesting decisions. You just insist on referring to these things as interesting decision making, which I believe muddles an opportunity to have a more clear discussion in which we wouldn’t need to qualify each step forward with a specialized redefinition of decision.

  5. zovirl says:

    While I agree with your overall point that interesting decisions aren’t the only enjoyable game mechanic, I think you give snap judgements less credit than they deserve. They are decisions, just fast ones. The difference between good players and bad in these types of games is often the ability to make decisions faster. Often new players will be so overwhelmed they won’t even be able to make any decisions at all and will have to resort to button-mashing, but that doesn’t mean that good players aren’t making decisions…

    A few examples:

    1. I recently took 4 fencing classes. In one of the bouts, I was up against someone much better. I did very poorly, and it was obvious that it was because the other fellow was simply making better choices than I was. He would bump my sword, I would push back way too hard, leaving myself open, and he would quickly jump in to touch me. Once I looked up at the buzzer (bad decision) and he took advantage of my distraction to score a point. At the end of the bout, the instructor pointed out that he had been holding his sword slightly wrong, exposing a few inches of glove which I could have used to score points…a better fencer than I would have then been able to make a decision about whether or not to go for that target. In general this all happened too fast for me to follow but after he scored each point it was obvious that he had been controlling encounter & making decisions the whole time.

    2. I’ve seen the same thing in kung fu (at which I’m also a beginner): much of the beginner work involves learning which hand to use in which situation. When I choose wrong and end up blocking with the wrong hand, it leaves me open. To get better I will need to get better at making the right decision every time.

    3. Are you familiar with John Boyd’s OODA loop ( It is useful for all kinds of things, including fighter pilots dogfighting. The basic idea is that in a dogfight, each pilot is in a decision cycle where they are observing information, making decisions based on that, then acting those decisions out. The pilot who is faster at this has a huge advantage: by getting to “act” first, they get to change the conditions. They means the opponent is now making a decision based on outdated information. By getting ahead of the opponent, the pilot is able to make better decisions because they are working with more correct information.

    4. In the videos of the famous “Daigo vs. Justin” street fighter match (google it…) I can’t follow what’s going on so I don’t see what’s Daigo is doing until after the fact, but the crowd starts cheering *immediately*. They are all good enough at the game to be able to process the incoming information fast enough that they know Diago has decided to parry as soon as he does.

    Like I said at the start, I’m not arguing with your overall premise that not all games have interesting decisions, but I do think that in many of the examples you gave there are, in fact, interesting decisions. It is just that they have extreme time pressure and you may need to be an expert to see them quickly enough.

    • Chris DeLeon says:

      I agree with the gist of everything you’ve written about here. Except that I think, based on what I believe I outlined rather clearly above, that it is a mistake in thinking to insist on calling tacit skills and their reflexive/conditioned application “interesting decisions”.

      > I think you give snap judgements less credit than they deserve.

      Where did I say that?

      What I am intending to accomplish here is to suggest that these ‘quick decisions’ (not my ideal choice of terms, but trying to relate) are of such a fundamentally different character in terms of gameplay development, user experience, and scholarly discourse that they need to be understood and thought about in their own regard, not as some inferior or simple subcategory of decision-making. Or rather, consider entries like this one my expressing concern that thinking about reflex-level performance in the same way that we think about decision making will lead to polluted results, often detached from reality of the experience, and that we need to consider different practices and concepts when thinking about these aspects.

      I’m also a (former) fencer, and submission grappler (3 years wrestling, 3 years mixed martial arts), so I hear you out regarding the types of distinctions that you are referring to here.

      If the analogy may help, the athlete’s trainer or coach (batting coach, boxing trainer, wrestling corner coach) possesses the sort of explicit knowledge useful to identify what’s wrong in terms of decisions made, but the athlete’s failures and successes (providing their coach has led them to drill into habits actions and reactions consistent with what the coach considers sound decision making) are more often not on the level of decision making but on attention, precision, timing, reaction, dexterity, and (these last three are somewhat less applicable to videogames than sports, though not completely inapplicable) complex matters of muscular coordination, physical fitness, or enduring various types of fatigue.

      Re: Daigo vs. Justin, I believe that a distinction needs to be made here between what is going on for the audience as opposed to what is going on for Daigo and Justin during the flight. Clearly, the audience is able to interpret what they are watching as decisions – to explain why one player did this, or why the other did that. However what the players in the match are experiencing, under constant pressure and existing in the moment of uncertainty, cannot trivially be assumed to be the same as the audience’s experience, which takes place under no pressure and enjoys the luxury of time after/during each move to reflect on its probable wisdom. (As I have stated elsewhere, I believe that a common source of confusion in writing and thinking about gameplay has been to assume that studying the experience as a spectator is the same as studying the player’s experience.)

      I am unfamiliar with the OODA loop, but it indeed looks very relevant, and I’ll dig into it soon. Thank you for the lead! (Upon first glance it looks like a more straightforward and general case of what I was attempting to make sense out of in Diagram: Player Actions and Memory.)

      • IQpierce says:

        I’ll move my own reply down here – it seems that zoviri and I similar points, i.e. that we believe that “micro-decisions” are important in their own right…

        I won’t deny that these different “classes of decisions” (as I’ll continue to call all of these things, though I’m starting to agree that this tack of definition may be more confusing and distracting than it’s worth) have massive differences. Ultimately I press my point – that they DO have important similarities – because it was this realization that clarified a great deal about game design for me. Essentially it led me to believe that Tetris, Flight Control, Guitar Hero, Super Mario Brothers, Canabalt, Solitaire, Advance Wars, and many many other games share a structure that’s virtually identical: they each are based around testing a skill, and they each involve continuing to learn and improve that skill. In this model of gameplay, I further break down each skill into its component parts, which is what I’ve been calling “decisions”: essentially my argument is that every human skill is composed of “decisions.” Improving this skill is the process of realizing which decisions are right and wrong in given situations and variables. Improving a skill through learning is therefore synonymous, for me, with honing a “decision-making algorithm”, i.e. a “strategy.”

        Again, to me, all of this is true of all these genres of games, but it only becomes clear if you realize that some things that we never would think of as the same thing (e.g. making a thoughtful and considered high-level strategic decision in a game of Civilization, versus making a flick of the wrist in a game of Galaga) ARE actually the same thing within this model of challenge-based gameplay – they each fill the same role in this model. Normally we would say that players who play Galaga have a “technique”, not a “strategy” – but I believe that the role of each in their respective game designs are the same thing.

        Perhaps not everyone else is as excited by this for me, but to me it creates a definitive model of what fun gameplay IS – in other words, a formula for creating fun gameplay. (Note that his model also encompasses Koster’s “Theory of Fun”, one reason I call it a “Unified Theory.”) To me this feels like as important and useful for game design as Joseph Cambell’s “metamyth” theory is for story writing.

        • Chris DeLeon says:

          Does it not suffice, then, to say then that a game that is fun due to challenge (the many other types of fun, such as fascination with spectacle, notwithstanding) involves skill, and that we ought to acknowledge that skill, though different in Go than in Quake 3, is an important factor in the design and play of each?

          In short, I still do not see the need to pull ‘interesting decision making’ into this otherwise valid point you are bringing up 🙂

          > the role of each in their respective game designs are the same thing.

          This does not make them the same thing, though.

          An old notion of keeping play interesting is to provide the player with more responsibility than they can perfectly handle – in chess, Risk, and other strategic games of discrete movements this is managed by permutational complexity of the domain, whereas in games like Centipede or Robotron 2084 it is by the intense speed and number of simultaneous movements to be dealt with, i.e. attentional limitations and demands on dexterity. Is this the type of idea you are referring to as shared? (It is related to fun in the form of Csikszentmihalyi’s Flow, popularized within the videogame industry by Jenova Chen’s MFA work and thatgamecompany’s subsequent game.)

          > Improving this skill is the process of realizing which decisions are right and wrong in given situations and variables.

          But this again is disregarding the performative factor, which as a characteristic of the cases to which I am referring, is nontrivial to the point that failure or success is at least as often from lack of (or from sufficient) practice to execute properly what one could reason as a right decision, and that there is a great gap in such games between knowing what to do and being able to consistently do it properly under pressure. This is so much the case that in many real-time videogames (as in many sports situations), the “right”-ness and “wrong”-ness for an action in a given circumstance is as much a function of a player’s ever-changing confidence, mastery, and momentary focus as it is of the game’s structure.

          • IQpierce says:

            This reply is delayed, sometimes actually developing games really cuts into one’s time for game-development theorizing. 🙂

            I do believe that Flow theory is very important to the ideas I’m expressing here – that Flow itself might be represented as entering a loop which could be characterized as “make decisions, take action, see results of action, learn from results of action, alter decision-making pattern.” Flow can be identified in both “high-level”, cerebral human activities involving clear conscious decision-making (e.g. programming) AND “low-level” twitch-based activities (e.g. playing a musical instrument).

            As different as they are in every other way, these activities do clearly have something important in common in that they can both allow the participants to enter a state of Flow. I believe that “fun” in challenge-based games is, in fact, Flow. (Puzzle games such as Braid or Grim Fandango wouldn’t fit into this “challenge-based” category by the way, though there are some applications even to those games.)

            I guess that, fundamentally, l’ve come to believe that the distinction between how our conscious mind works and how our unconscious minds work is not as important as we make it out to be. There’s more and more research than many of the motives we would use to explain our decisions are, in fact, fictions – that even for big decisions, much of our decision-making actually takes place on a subconscious level that we can’t explain. Marc LeBlanc says that in the creative process, using your intellect is using your brain’s “low gear”, and using your intuition represents doing the same thing but in “high gear” – in other words that you should only be consciously thinking through your decisions when you’re on unfamiliar ground and need to take your time… it’s a necessary evil which will and should go away with time.

            I think this is true of all human performance: you start out making decisions very consciously and eventually you find you’re making them unconsciously. But just because you stop being aware of the process of how you’re making them doesn’t mean they stop being decisions. (Though you could argue that definition semantically, I don’t think there’s a true fundamental difference.)

            As a final aside: I love that I’m now talking about a concept that’s applicable to the activities of both making games AND playing them (LeBlanc’s intellect/intuition advice was for how to design a game and I’m applying it to the player experience of making decisions within a game). Game design is pleasantly recursive: it’s an activity in which you design activity… and I often engage in creative problem-solving to figure out how to let players engage in creative problem-solving. As my blog subtitle says, I consider myself to be “one guy trying to make some interesting decisions” – in both of the senses in which that sentence can be interpreted.

  6. Paul says:

    What you call “mangled definition,” I call “basic definition.” Decisions are decisions, even if they aren’t nuanced, labored, etc. Your bicycle backflip analogy is not remotely apt because you’re describing a purely procedural act. You perform the same steps in the same order to achieve the same results. Absolutely, there is no decision there. However, when the order of the steps change, a decision must be made.

    I don’t have a particular interest in the soundbite (especially not applying or denying the “game” title based on it), but it’s worth noting that the qualification of “interesting” and “series” are perhaps more revealing because they’re relative and context-sensitive. Series are not just groups of consecutive, individual events, they’re related. An incredibly simple, quick decision that is uninteresting to you in a vacuum could become interesting in the context of a series. Figuring out why that experience arises or how to learn the skill of predictably creating interest due to context is what I think is most worth considering because it’s where we leave semantics behind.

Leave a comment

Comments Form
  • Your email will not be used for any purpose other than to update you in replies to your comments. Your website address will be shown as a link from your name with your comment. Your profile photo is auto-magically provided by Gravatar.

All contents Copyright ©2017 Chris DeLeon.

Site production by Ryan Burrell.