So, we’ve talked about what stats make for interesting choices, and we’ve talked about how to provide information to contextualise those choices. Now it’s finally time to get to the main event: writing the choice itself.
But what is a choice, actually?
Ultimately, a choice consists of three things: a setup to contextualise the decision being made, a decision point presenting multiple options, and consequences which occur as a result of taking one of those options. We’ve already talked about how to handle the first and third parts of a choice, now we just have the decision point itself. So, how do we present these decisions in an interesting and compelling way, which not only feels fair and intuitive to the player, but also present enough of an ethical or mechanical challenge to keep them engaged in the process?
A decision, at its simplest form, is a point where someone is presented with two or more outcomes and has to choose one: do I want pizza or cheeseburgers. Do I want to take the bus, or drive? Do I want to work on my Patreon articles, or work on that project I can’t tell anyone about yet? The tension between these two mutually exclusive outcomes is what I’ve heard referred to as an Axis of Choice, and creating balanced and varied Axes of Choice is how you create interesting choices.
So let’s drop down another level. What’s a balanced Axis of Choice? Well, think about most of the decisions you make every day: the ones which you probably make without even really thinking about them: the decision to get out of bed in the morning. The decision to eat, to shower, brush your teeth, flush the toilet. These are all decisions which you probably don’t put much thought into because the Axis of Choice for each of those decisions is so heavily weighted towards one end: of course you’re going to brush your teeth, because the alternative is having them rot in your gums and fall out. Of course you’re going to get out of bed because you can’t do much of anything else in bed. The choices you really have to think about aren’t these heavily weighted ones, but the more balanced ones: pizza or hamburgers, the bus or the car, paying work I like or somewhat different paying work I like.
Of course, perfectly balanced Axes of Choice are impossible from a game design point of view. Despite our best efforts, none of us can actually read the minds of the people who play our games, which means choices which might seem balanced to us might still be obvious ones to a given individual player: someone who really likes burgers, someone who lives in a city with terrible bus service, someone whose Patreon is a lot more successful than mine. That’s why those Axes of Choice also have to be varied – and there’s two ways of doing this.
The first is to change what those choices pertain to in subsequent decisions: give the options of pizza or ice cream the next time instead of pizza or burgers, biking instead of bussing and so on. This means the player isn’t simply choosing the same thing over and over again because it means the player’s optimal option isn’t always available, so they’re forced to choose an alternative. This approach has a bunch of pitfalls, the first being that every one of those alternatives still has to provide a meaningful consequence – which generally means that each of those alternatives has to have a stat supporting it. That’s something which leads to the stat bloat I talked about two months ago. The other problem is that while the choices are varied, a given player is likely still to see the choice as a binary one between something they like more than something they don’t. Ultimately, the result is something more complex that choices which offer the same two options over and over again, but remains somewhat less than ideal.
This brings us to the second way, which can – and should – work alongside the first. By providing combinations of outcomes instead of single outcomes per option – and by providing more potential outcomes than just two – a single decision can have multiple Axes of Choices. Now, the player no longer has to just think about whether they prefer one of two things more or less, but whether they prefer multiple sets of mutually exclusive outcomes which are bundled in with one another: maybe I have to take the bus to get pizza and drive to get burgers. Maybe I could choose to bike and get ice cream too. Now someone who prefers to drive but likes pizza, or someone who hates ice cream but prefers to bike everywhere is presented with at least two potential outcomes they might want to take – or at least, would like to avoid least – and they have to do some serious thinking to figure out which one to pick.
And in case you haven’t noticed, we haven’t really increased the number of total variables at all.
Combining these different principles should allow you to create interesting choices for any kind of narrative-driven game, ones where the player will be forced to encounter multiple options which force them to choose between different outcomes they find equally appealing or unpleasant. While not every choice will land the same way for every player, by varying the types of potential outcomes on offer – and by varying the combinations of outcomes in the options of each choice – you should be able to consistently give players an engaging experience. Even if one choice doesn’t engage them, the next one might, and so on.
But a good narrative-based game isn’t just a question of well-designed individual decisions. These decisions also have to be arranged in a way which supports the narrative effectively, and makes the player feel like the game is judging them “fairly” for their choices. It has to ensure that players don’t dig themselves into a hole and render the game unwinnable (unless you’re making a game themed around that sort of thing). Finally, it should create a mechanically interesting challenge for the player, so that ‘winning’ the game still feels as satisfying as concluding the narrative does.
Paul Wang 