“A game is a series of interesting choices” – Sid Meier
The above quote is one of the first foundational ideas one encounters when studying game design. Sid Meier, most well known as the designer of the Civilization series of strategy games, has gone on since saying this in 1980 to expand the idea, to explain what in his view makes choices interesting – or not. While this is a worthwhile exploration, and I am likely about to retread some of the same ground, I’d like to abstract this out one level higher and ask: What is a decision in the context of a game? What are its traits, how can it be compared to other decisions in other games?
Before getting to that I’d like to talk about my personal taste in game design for a moment. I dislike having a huge number of options at my disposal, a bag of tricks each suited to different situations. When I have such tools, I find that the experience of playing the game most often comes down to either deploying the one tool a particular obstacle is designed for, which feels like glorified administrative work, or frantically running around trying to find an appropriate tool in a big sack of similar tools. Neither is very fun for me. I usually prefer a few very flexible options, where success comes not from having the perfect tool for the job but from learning how to use a tool the perfect way for a job. This is one reason why I am often drawn to 2D platformers – choices of which special gear or technique to use are seldom emphasized, but very subtle distinctions in position and momentum can become very important.
When considered from that angle, it’s clear that interesting choice is not merely the large discrete decision of what unit to build, item to use, position to take, spell to cast, but also the thousands, even millions of tiny decisions of momentum, orientation, terrain and sight-line. One of my frustrations with cover shooters as a genre from their inception was that the entire concept was of taking a fine-grained organic decision – what is the best position to take in this 3d space in terms of offense and defense, judged based on subtle properties of sight-line and attack trajectory – and made it discrete and formalized, a binary decision, either in cover or out of cover. In games, as in life, the decisions that shape our outcomes are not merely checked boxes, discrete actions, but also tiny micro-decisions, details of route and word choice, that in the end might put us at the head of a household or in front of a train. If not all decisions are large and discrete, what else might we have overlooked in this conception of “choice”?
In considering this, I have developed a taxonomy of sorts for evaluating “interesting decisions.” I spent an embarrassingly long time trying to figure out a snappy acronym, and so far the best one I have is ORCID: Options, Resolution, Consequence, Information, and Determinism. While I developed this framework to describe the overall decision-space offered by a game, could also be used to evaluate any individual decision. I will explain each trait in turn.
Options: The number of candidate decisions that must be considered before a choice can be made. Generally this will be high for strategy games and low for action games. In order for options to be meaningful, there cannot be one clearly correct or incorrect choice – sometimes adding options for options’ sake just results in obscurity, in decreasing the information the player is able to glean by overwhelming them with extraneous clutter. The difference between low and high options is in having two meaningfully distinct buttons and having two-hundred.
Resolution: The degree of flexibility within a decision after committing to it. For instance, a jump in classic Castlevania wholly commits the player to an arc of travel, always rising the same amount before falling, with no ability to change direction once the jump is initiated, so this has very low resolution – the choice is merely to jump or not to jump. Conversely, in Super Mario Bros the player can control the height of the jump and retains control over the character’s momentum, so many different types of jump and jump techniques are available. The difference between low and high resolution is in having an elevator call button that summons the elevator to your location and having one that gradually moves the elevator towards you and can be released at any moment to stop its advance.
Consequence: How important each individual decision is to the overall outcome of the game. This can vary a lot from decision to decision, and at times much of the challenge of the game can come from evaluating this – more on that in a moment. Most games have a philosophy on which decisions should be consequential – either that it’s attention to fine detail that ought spell the difference between success and disaster, as in Souls games and in XCOM, or big important decisions ought be clearly presented as such, as in most story-focused games, or no decision need be too important since having an undesirable outcome would interrupt the story or the player’s power fantasy, as in many AAA games. The difference between low and high consequence is in having a button that delivers a pizza and having a button that delivers a nuclear warhead
Information: How much information is provided about the potential benefits and drawbacks of each decision. A game with perfect information is one where the player is completely aware of the current state of the game – though even in a game like that some information is obscured, such as what their opponent is currently thinking. A game with minimum information is one where you have very little idea what the outcome of a choice might be – which might sound frustrating, but can also be charming and whimsical, as with many classic adventure and narrative-based games. Some obscured information is static and can be looked up on a wiki somewhere, whereas some is dynamic and hidden for every player outside of cheating, but this distinction can be deferred for the moment. The difference between low and high information is that between a correctly labeled button and a mystery button.
Determinism – The objective certainty of a specific outcome resulting from a decision. In Chess, every decision is completely deterministic and always has the same outcome – at least when it comes to individual moves. In Dark Souls or Dungeons and Dragons, many choices could fail to execute as planned either due to player error or unlucky dice rolls respectively. In many cases low determinism is only expressed through a chance to fail, but a choice could have any number of outcomes beyond merely success/failure, and this measures how much these must be taken into account. While games with explicitly randomized outcomes are overtly non-deterministic, this also includes any game where it is difficult to assess the specific outcome of a choice due to other factors such as challenging execution or complex and unforeseeable interactions. The difference between low determinism and high determinism is between a button that randomly (or based on some circumstances not entirely under your control) delivers a pizza or a nuclear warhead and a button that always delivers one or the other – regardless of whether you know which it is.
These properties are not entirely separable from one another: Extraneous options can obscure information with noise, high resolution reduces determinism through micro-failures of execution, and if consequences of or information about decisions are reduced enough they cease to be regarded as meaningful decisions. Clearly, a mere numeric evaluation of a game from this perspective is clearly lacking in a great number of ways but I’m about to do it anyway, from 0-5.
Civilization Series:
Options – 4 / Resolution – 1 / Consequence – 3 / Information – 3 / Determinism – 4
There are an overwhelming number of options, particularly as the game progresses, but since few of them have any granularity they are still manageable in aggregate. While few choices are unimportant it’s uncommon for a single decision to be the difference between victory and defeat. Though it can be difficult to keep track of, the game generally tries to keep the player well-informed of the potential outcomes of their decisions – though, of course, only within the parts of the map they have vision of. Elements of random chance affect outcomes in combat and diplomacy and must be accounted for, but most major decisions have no chance for failure and are entirely deterministic.
Dark Souls:
Options – 3 / Resolution – 4 / Consequence – 3 / Information – 2 / Determinism – 3
The player has many options at their disposal, both for which challenges to attempt and how to attempt them, and the movement and combat reward very granular choices of positioning and timing. The rewards and punishments of judicious decisions and failed actions can be significant, though not quite as brutal as many seem to believe. The context and consequences of your decisions can be extremely cryptic, but generally can be inferred from careful observation. The player can expect many of their choices to fail due to poor execution on their part, such as missed timing or positioning, or being interrupted by an enemy, but such situations can also unexpectedly benefit the player.
Visual Novels
Options – 2 / Resolution – 0 / Consequence – 5 / Information – 1 / Determinism – 5
This is a broad generalization, but the genre tends to focus on giving the player a few options which can significantly impact the outcome of the story. There is no granularity to these options and seldom any divergence in outcome, but any information given about what may come of a decision is obscure at best.
Sometimes, examining a game this way can prompt you to see it in a whole different dimension. Chess is widely regarded as a game where there’s each choice has a defined outcome and the player has perfect information – and this is quite true, at the scope of individual decisions. However, if you expand your view and consider the decision of what strategy to employ, all of a sudden each choice of strategy has many outcomes, and information, expanded to include insight into how an opponent might respond, becomes imperfect. So we can really view the decision-making of chess in two separate layers:
Chess (Tactical)
Options – 3 / Resolution – 1 / Consequence – 4 / Information – 5 / Determinism – 5
Chess (Strategic)
Options – 5 / Resolution – 4 / Consequence – 5 / Information – 2 / Determinism – 2
This duality is unusual, even among other classic strategic games – in a game such as, for instance, Go the individual moves are so simple and unimportant as to mean nearly nothing compared to the overall strategy, but for novice chess players many only understand the game in terms of individual moves at first, only coming to see the strategic component later.
While in Chess’s case this strategic/tactical game emerges naturally from the gameplay mechanics, other games have an explicit division of strategic and tactical layers. One example of this is XCOM, where you have a light strategic game, where you can make basic decisions about what missions to take and research to invest in, and an extremely harrowing tactical game where you can lose the whole campaign to one careless move. Again, we can regard this as almost being two separate games for the purposes of evaluation:
XCOM (Strategic)
Options – 2 / Resolution – 0 / Consequence – 3 / Information – 4 / Determinism – 4
XCOM (Tactical)
Options – 4 / Resolution – 1 / Consequence – 5 / Information – 4 / Determinism – 1
Here’s a few more example evaluations, just for the hell of it:
Deus Ex
Options – 3 / Resolution – 3 / Consequence – 3 / Information – 3 / Determinism – 4
Spelunky
Options – 2 / Resolution – 5 / Consequence – 5 / Information – 1 / Determinism – 2
Super Mario Bros
Options – 1 / Resolution – 5 / Consequence – 3 / Information – 4 / Determinism – 3
The Legend of Zelda
Options – 3 / Resolution – 3 / Consequence – 3 / Information – 2 / Determinism – 3
None of these high or low scores are meant to imply that anything is lacking, merely to give an idea of how these different game designs approach problems conceptually. We observe that action games tend to have high resolution and low determinism, tactics games tend to have high options, RPGs tend to have low information, and so forth. These things make sense intuitively once you examine them: Action games create granular choice through smooth and reactive controls and create a context where failure of player skill is possible, RPGs are interesting because of exploration of unknown territories and thus tend to provide relatively little information, strategic games are geared specifically to provide “interesting” options to deliberate on so have low resolution and high impact. You’ve probably noticed you have preferences in what sorts of games you prefer to play. Perhaps like me you get stressed out by having too many options to sift through – or maybe you like collecting devices and tricks to use in every situation! Maybe you like the continuous micro-decisions of positioning and sight-lines, maybe you prefer big discrete choices; maybe you want each decision to have a huge impact on the outcome, maybe you like lower-pressure options where you know things will generally go where they need to even if you make a mistake or two. It can be helpful to think it through, to understand what making a choice means in each game, and why those choices are interesting – or aren’t!
While nearly any approach can probably work and create an enjoyable and interesting experience, most players prefer to know what sort of experience they’re in for, and this is one area where ORCID evaluation is useful – If there are several viable options, if the resolution of each choice is high, if the consequences are severe, if the choices are informed, if your choices are non-deterministic, these are things you should understand about your game’s approach as you craft these choices – and which you should be conveying to the player as you go. It’s no use providing the information if the player doesn’t understand how it relates to the choices, or of having consequential outcomes if the player doesn’t understand the consequences, or of having many possible outcomes if it’s never clear how the outcome connects back to the choice that was made. Even if your design is weird and obscure, it needs to communicate that weirdness and obscurity to the player.