The Emotions of Rogue-Lites

Using technical jargon is cool, but how about discussing sensation?

Talking about emotions is hard.

As designers, artists & creators in general, it's easier to explain our medium using concrete facts: the game is that long, it uses this and that mechanic, it has this many items. However, by explaining games for what they are, we lose sight of the most important: how they feel.

Rogue-lite is one of the most fascinating genres to me precisely because of its (too-often-eluded) emotional core. The best games in the genre stand on a thin frontier between fun and frustration, luck and strategy, familiarity and variety.

If you only focus on the big factual terms, such as procedural, perma-death or metagame, you will not understand the experience underneath. You’re at a high risk of failing at recreating the feeling of rogue-lite, even if you tick every box in the checklist.

So today, we're having a look, not just at what rogue-lites are, but also at why they work and how they create strong emotions.

The Central Symmetry

Rogue-lites may use varied core gameplay, shooting, platforming, cards or action, yet they all present the same equation to the player: kill or get killed.

I discussed the question of omnipresent health systems in a past issue of The Arcade Artificer, and the main points remain valid here: in such “me or them” situations, there's an obvious tension with two clear outcomes. You may stall out a little bit, but rogue-lites always force you to face danger, even if you don't feel prepared. No time for grinding.

In such a ‘do or die’ configuration, you have two basic needs: means of offence and means of defence. The designer's role is to give the players enough options to cover both: through the permanent mechanics (action to perform/rules of the system) and additional tools to collect while playing. For instance, a rogue-lite with top-down shooting will typically let you avoid damages by moving around but may also include armours or health pick-ups.

All twists and turns are allowed to create a diverse experience, from economic loops to secret combos. The defence could mean blocking damages, healing after the fact or simply preventing the enemy's damages (using a stun mechanic, for instance). There is also an incredible variety of ways to deal damages: combo, over time, one-time, charged attack. RGD principles can help to find all sorts of meaningful variations, but at the end of the day, the player always intuitively translates it all into damages gain/loss.

To be more accurate, the player is typically focused on time: the damages per second/per turn they can inflict and sustain. Indeed, in most cases, you have a permanent option available to evade damages; it's just that the margin of error is thinner as the difficulty ramps up.

We all end up making mistakes or getting unlucky, so when the combat drags out for too long, we start feeling a sense of dread and impending doom. Mistakes, just like good decisions, compound; small fails here and there will ruin your run eventually.

Run Completeness

All of that is great, but it works for all sorts of survival and multiplayer games too. What sets rogue-lites apart is another critical component: the sense of completeness.

Indeed, a rogue-lite is structured in “runs”, a series of combat encounters & choices with an end and, therefore, a purpose. You want to finish it. The final boss is like the graduation exam in the “kill before getting killed” class. If you can't complete this challenge, you need to sharpen your strategy and come back stronger.

This sense of completeness is integral to the satisfaction a rogue-lite provides. They start simple, you're weak, but as you survive increasingly difficult challenges, you grow so that at the end, you're at your peak. Rogue-lites deliver their whole (simple) story in a single session; conveying this feeling is more important than the factual run duration.

A run-based structure gives the genre another perk: general knowledge aside, no need to remember what came before or to care about what comes after. You feel present and focused—the choices you make count now in this run.

Input Randomness

And so, we get to another key ingredient of rogue-lite that is often misunderstood: the procedural generation. The procedural isn't about randomizing stuff so it's less boring when replaying, it's about renewing the strategic feeling. Any fixed game, even a difficult die & retry, will eventually shift its mental challenge from « thinking » to «memorizing». Once you've figured out the strategy, you can just focus on the execution (and it can be a great deal of fun, like in Souls, it's just different).

Rogue-lites want to challenge the brain again and again, therefore they have to present the player with some input randomness (a unique situation) and let them figure out how to deal with it. Which rewards to specialize in, which risks to take, how to explore the dungeon, etc.

You may ask: «isn't random the enemy of strategy? If good or bad luck can destroy your strategy, how is it more fun? »

To make an analogy, random in rogue-lite is akin to poker: you get dealt a hand with a certain power level, influencing your chances, but you can still make the situation work regardless of your initial luck (through bluff or simply by minimizing your loss). Like in poker, the randomness of rogue-lites is diluted with the number of occurrences (which reduces variance): in most cases, you'll have a bit of both good & bad luck, your challenge is to succeed in all contexts.

Screenshot from the improbable crossover Poker Night in the Inventory.

The procedural generation generates a unique puzzle each time, sure, but again it's not as much about randomness as about the rules. Rules are the appealing part of the rogue-lite variety: you jump in the session quickly and learn by doing. Things such as enemies patterns or more obscures facts such as how secret rooms appear in The Binding of Isaac —this is a knowledge you acquire to help you survive and typically compliment the strategic aspect: you don't “solve” the game, you learn patterns to get better at strategizing.

Having more info about the inner working of the game helps you make more informed choices and use your resources efficiently. Humans feel content when they get better at executing a task.

Fun Again & Again

However, designers had to solve a last problem to truly enable the potential of rogue-lites: alleviating the pain of losing. Losing sucks, it always sucks. Even when it's “part of the game”, and you learned something new, it feels terrible.

The solution has a name: meta-progression. Changing just one letter from rogue-like to rogue-lite, designers leveraged modern design techniques to do two things: materialize the importance of each run and incite to try again. When they die, games let the player keep a currency: even if it's 5% of all collected items, it feels better than zero. Using this currency to then marginally improve the starting character (or unlock another one) may be symbolic, but it matters.

Games also frequently use the meta-progression to unlock new content & mechanics, which both encourages to try another run, and helps to space out the strategic aspect. Indeed, any deep system with many rules will feel chaotic and indigestible, but by introducing notions one after the other, the player can absorb a surprising amount of knowledge.

There's a tricky question to address here regarding the game’s fairness: is it genuinely possible to beat the game on the first try, or is the meta-progression a sneaky way to upgrade your raw power? Some players prefer to feel tangible progression, while others like it when their skill has to grow. The most important, however, is that the player doesn't feel like meta-progression was the key factor: you don't want them to lose the sense of presence and blame impossible starting conditions, or worse, to feel like a win is only achievable long after grinding.

Wrap-Up

Some say that asking good questions is half of the design work. I think it's even more than that. Questioning our feelings when we play is integral to learning how to design. If you only read a review that says “frustrating”, what is it about? At the system too opaque to strategize properly? Or the choices you make don't matter? Unlike technical jargon, these are the sort of things that few people discuss excplicitly, so you only have one option to figure them out: listen to your emotions.

If you liked the topic of this article, I’m sure you’d also enjoy this one: