What Games Do I Want To Make?

I’ve asked myself this question over and over. Other people have asked me this question over and over.

If you’ve read any of my articles, especially those on my personal blog, you will know that I struggle a lot with two things.

• One, events from my past have made it pretty impossible for me to feel joy or be interested in anything anymore.
• Two, it’s hard to get a sustainable income from all my creative work.

I’ve tried so many things. Wrote books, recorded music, and, of course, made a lot of video games. But none of them were ever really successful. And I knew that before releasing/finishing them. If you play and make that many games, you naturally develop this “taste” or “sense”. You just know that a game isn’t that fun. That the thing you made isn’t good and appealing enough to stand out and find an audience.

Knowing this is the case, though, is not the same as knowing how to change it!

The Problem

I kept making games that did not really interest me. Games that I thought would be “marketable” or “relatively easy to finish and sell”. Games that I thought “other gamers wanted” or “fit the market”. As I played my own game, I felt no joy, and thus would be surprised if anyone else did find them enjoyable.

This is obviously bad. Everyone can smell it when a project wasn’t made with care, but instead because someone hoped it might sell.

But if I asked myself what games I would want to make, I would never find an answer. It’s this weird state of “I AM a game developer that wants to make games”, but also “nothing interests me and I don’t want to make ANY of my ideas”.

You work hard, you push through, you try things anyway … and the result is a finished game that is not bad but also not good. I think all games I ever made fall in this category. None of them are ugly, or unplayable, or terrible. Also none of them are really fun, or interesting, or made with, you know, passion and care.

And so I ask myself: “What games do I want to make?” (Or more generally: “What creative works do I want to make?”)

Then I released my first Steam game. It did exactly the same thing as all my earlier games: it’s not bad, it’s also not great. Sure, Steam forced me to become even more professional, work harder, learn a lot more. It also showed my game to a larger audience and I earned a tiny bit of money. But the end result was largely the same.

Two Important Events

Except that I realized something. After some soul searching again, trying to pick the next Steam game to make, there was a combination of feedback from players and something I read that triggered my brain.

EVENT 1: My first Steam game was this weird mixture of an idle/auto-battler game … and a very active/arcade-like game. It failed with every target audience I initially wrote down.

• People told me they play idle games to relax. But they could not relax or “be passive” at all in my game.
• People told me they play incremental games because they want full control over a simple state—everything is handled through upgrades, you can read exactly what they do and buy as needed, no surprises—but my game adds a lot of randomness and fuzziness.
• People told me they play auto-battlers so they can just make a tactical choice, then walk away while the battle happens if they want. But my game has very short battles in which at least some of your players are not automatic.

EVENT 2: I stumbled upon some essays about the meaning of fun and researchers claiming that fun is the byproduct of having a system that’s completely knowable.

• In real life, we can’t predict or understand anything fully. There’s too much chaos, randomness, unknowns, etcetera. (There’s just too much in general.)
• Games (and stories and such) are a heavily simplified, completely defined system where we can know the rules and can perfectly predict outcomes. And this is where the fun comes from. This is what makes them relaxing and nice and satisfying, even though we are actually using our brain to play the game and the game might even be hard/punishing.

These two events coincided, and I had this weird Eureka moment about games, psychology, and myself :p

I was making games to really challenge people. To not let them be “bored”. Whenever I was bored with my own game, I’d find a way to make it slightly more complex or challenging.

But most people play to relax. To zone out. To still use their brain, but not in a taxing way. To get the satisfaction of achieving things and getting stuff done, but without the pressure and harsh consequences of those same things in their real job.

They want to be “challenged” only in just the right amount/way, which is why we have genres and people religiously stick to the one that fits them the most.

People who like auto-battlers find that those games fit their brain best. More idle activities (like watching a YouTube video) are too boring, but actual battlers/shooters/action games are too challenging. Nothing I can do will change that—I should not even want to change it. Those people will never play my “auto-battler but with a challenging twist!” because that literally defeats the purpose.

The same thing is true for people who play idle games or incremental games. I was trying to add some activity and extra tactics to such a game … while the reason people play it is to relax and to be able to passively see numbers go up and achieve things. An “active” incremental game is too challenging and ruins it. It doesn’t matter how interesting I make the puzzle/tactics/activity, it doesn’t matter if it fits my theme perfectly, it’s just not what those gamers want. Now I understand why almost all these games pick very peaceful themes and almost laughably passive ways to play them.

Now you might think this is obvious. You might be mad at me for not playing enough games or not fully understanding genre. But please remember my second point: it’s incredibly rare for me to enjoy or prefer anything. I’ve played all game genres you can imagine, just as I’ve read books in all the genres and listened to all kinds of music—but none of them are my favorite or anything. None of them made me happier than another. I’m open to try anything … but it’s very hard for me to judge whether something is actually good or enjoyable or matches what other players want of the genre.

That’s one of the massive downsides of a brain that does not have preference, or favorites, or a proper dopamine response for that matter. The downside of a brain that thinks in the opposite way compared to your usual human: I take every situation as-is and go with the flow, instead of imposing personal preferences and restrictions.

I’m permanently thinking out of the box, being flexible, working with what we happen to have instead of shaping the world to my will. And so I’m also permanently switching to a new genre in … everything. This has always made it hard for me to understand why people only buy games in one specific genre and how to stay within those confines when making a game.

Fun Games Have Completely Knowable Systems

This is where EVENT 2 comes into play. It made me realize that yes, there is one thing that I want (more) out of games. My hyperactive brain is killing me most of the time, giving me twenty ideas an hour. It’s so hard to focus and prioritize for people with such brains. Everything seems equally important—and equally unimportant five seconds later when you’re distracted by the next idea. This is also why most people like me are highly sensitive: even the smallest sounds feel equally important to our brain as the loud actually important ones :p

And so, all day every day, I’m bombarded with information from my senses, with ideas, with possible problems and also possible solutions. It is so much. It never stops. And I can’t really do anything with it, because the world is so complex, the future can’t be predicted, information is fuzzy.

So what do I want? What would I like most out of a game? Well, I agree with those researchers. What gives me some bit of fun still, and what I think applies to almost everyone, is that feeling of control over a system.

That very rare feeling of a system that is simple, clear, and knowable.

A way to shut out the chaos of the real world and be content playing with this well-defined tiny set of rules.

Completely knowable, completely self-contained, nothing else matters.

Reduce the complexity of city building to a minimalist building game with only three consistent rules. Reduce the hyperactive shooter games to just a single archer that slowly shoots arrows in predictable trajectories. Reduce complicated cameras and realistic 3D rendering with a single static camera in a simplified 2D world.

This realization explains so much about the preferences that I actually DO have.

I’ve actually mostly played action/shooter games that are 2D and less chaotic. And yes, I usually play Archer or Sniper characters, probably because they are “cleaner”. No messy in-your-face sprite overlapping melee fights. Just shoot an arrow in a straight line, once in a while, clearly hit or miss—a knowable system. If I wanted randomness and messy fighting I’d look at the real world.

For example, I’ve played a lot of Towerfall. (I’ve never stuck very long with most other shooting games.) It has extremely clear visuals and predictable system with just a handful of consistent rules. (The level wraps around, jumping on an enemies head always kills it, you must always retrieve your arrows.) Think about it. In the entire game, which has loads of different monsters and level layouts, it never destroys the rule that jumping on someone’s head kills them immediately. I can play that game and be completely certain that “land on head = kill”. I can play that game and be completely certain that my character never tires, or loses their arrows, or suddenly has bad aim, or any of the other chaotic unpredictable things that happen in real life.

I’ve played Rollercoaster Tycoon to death as a kid. But when newer versions came out (such as the 3D one, such as Planet Coaster) I did not like them at all and never really played them. They became too big, too complex, to actually really understand and know the systems. I heavily prefer the simple top-down 2D camera of RCT1 over a freeform 3D camera. Again, if I wanted to look at a 3D rendering of a caster from any angle, I’d go to a theme park.

Similarly, I played games like Age of Empires a lot as a kid. I loved the first one, even though its graphics and gameplay are extremely limited. I tried all the later ones and never stuck with them. They add too much uncertainty, too many complex systems you can’t fully understand (unless you go hardcore pro), they add too many things to track and do … when all I wanted was to relax after a day at school during the simpler Stone Age times, you know?

Looking back at my original game ideas, I can clearly see this now. My ideas were always based on these minimalist knowable systems. I always tried to treat them more like board games: “here’s 5 rules that are always true for the whole game, there’s nothing else to learn or discover, have fun!”

But I ruined all those nice ideas by thinking I had to add more stuff :/

The final video games I actually released where always more messy, more unclear, more fuzzy, and it actually sucked out all their fun. For me, yes, but also clearly for everyone who tried it, as the general response to my games was “I guess it’s okay, moving on”.

Predictable =/= Boring

I don’t want to give the idea that I only like old games (due to nostalgia or whatever), so let’s give a newer example: Rocket League. I only discovered that game once it was already 5+ years old. And it’s still probably the game I played the most.

Why? Because, as expected, it’s a simple, predictable, knowable system!

You drive a car in the way cars usually work (with some … rocket-y upgrades of course). You hit the ball into the goal. Buttons always do the same thing. The ball bounces as you’d expect.

There are no surprise mechanics. No hidden numbers. No exceptions in which the car suddenly behaves like a boat. No fifty pages of lore to read before you understand a part of the system. After winning ten matches, you don’t suddenly progress to a new area with new vague rules to figure out.

This may sound silly now, but lots of games do these kinds of things!

They do not communicate their internal rules to the player (or they’re too complicated to ever communicate), which has the practical effect of someone playing with a system they CAN’T predict/know/understand. They don’t give enough feedback, or they have too many exceptions to systemic rules, or they think they need to be realistic and therefore ruin the ONE THING that makes games FUN.

I’ve watched my brother play some competitive games from time to time. No matter how much I look and focus and ask him questions, the stuff happening on the screen is just not what I’d call simple, predictable, knowable system. There are way too many things left to random chance or not communicated at all. There are too many things happening in a tiny space for human eyes to properly parse and react to. There are too many exceptions to the system, making it ill-defined. (“But no, if you play THIS character, and it’s Phase 2, and you’re low health, THEN the rules are actually X and Y!”)

For some people, the action, the visuals, the tension, it’s still enough to have fun and be motivated to play. Just not for me. And not for a large group of gamers, the one I should be targeting. That’s the realization.

Again: if people wanted realism, if they wanted the uncertainty and chaos and unpredictability, they’d just go do something in real life.

Though Rocket League can be summarized as “drive ball into goal” with no surprises—at least not until you get to a high level or see some pro play!—it is not easy.

Oh no no no, your first 50 hours are just missing the ball. Your next 50 hours are hitting the ball but into your own goal. But I kept playing because I kept getting better at this one well-defined system I could fully understand.

Being simple is not bad. Being predictable is not bad. Having one simple system that just repeats forever is not boring or limited.

No, the simplicity of it, combined with knowing exactly how it works, is what allows much greater depth and variety in gameplay.

This is where fun comes from for most people. It’s also what has the greatest chance of making me feel something.

Being able to do something, to achieve something, without the chaos and messiness and unpredictability that you’d always get in the real world.

What do we take away from this?

Whenever I dream up new ideas, or think about the next thing to try and make, it’s basically always an idea that matches what I just explained.

Sure, I have many ideas that don’t really match, but they come and go and I never think about them again (or even write them down).

The ones that stick with me, the ones that give me some brief excitement to try and make them, are those where I can …

• Take something from the real world, which is inherintly drowning in uncertainty and complexity
• And turn that into the simplest, most predictable, most controlled system to play around with.

A toy, really. A tinified version of something from the real world, something I biologically know or find important. A way to cut out the chaos and the never ending worries, and be able to focus on just this toy system for a bit.

I’d dream of toy kingdoms to manage. Toy banks to rob. Toy towns where I can save kittens from trees or build homes for families. Toy mountains to climb in a way that’s waaaay simpler and less dangerous than real climbing. Toy theme parks to run. Toy pet shops to run, as well, without all the complexity and sad stuff from handling living animals in a real store.

I think this is true for most gamers. When they say they play (passive) games to “relax”, they really just mean “we want games with simple systems we can completely predict”. When gamers like shooters, for example, they really just mean that they have a great understanding for the systems in such games and can thus play around and predict them to their satisfaction. When you make a shooter that does not satisfy that, it doesn’t matter that it’s the right genre for the target audience, they still won’t like it and won’t play it.

And so I’m really trying to transform my mind-set and process with this in mind.

Don’t think “Well people are buying THIS kind of game, guess I’ll try and make one of those too”. Certainly not if I’m just going to eyeball superficial details about this game/genre and merely copy those. Instead, identify the simple predictable system underpinning the genre, and ensure that exact thing is in own game too.

• Randomness is still fine of course … if it’s simple and knowable! If players know “this thing is random chance” and, if possible, they know “the probability is X”.
• A handful of consistent rules that always hold and are always clearly communicated. (One of the biggest sins I’ve encountered when playtesting my games is the moment someone goes “why did THAT happen?” several times before they shrug and stop taking the game seriously.)
• Exceptions, mysteries, things to figure out as you play, it’s all fine as long as it’s used sparingly. As long as it’s used as seasoning on top of this well-defined system instead of being this black hole at the heart of a vague system.

Don’t think “Players will be bored! We must add tougher enemies! More ways to lose!” Instead, realize that people actually play games to get away from the challenging real world. They should provide similar systems, but simplified, safer, more controlled. That’s why people play. To still get the satisfaction in an easier and cleaner way. The “challenge” of games is in optimizing what you do within the system, but that does not require a complex or punishing system at all.

Don’t think “I need to add BIG TWISTS on the formula to make the game worthwhile and innovative!” Instead, realize that people play games in this genre precisely because they understand and know the core system. Only one big twist is allowed, and it should probably be some secondary thing that’s easy to understand. (Change in systems happens one small step at a time. Big innovative games are the lucky once-in-a-while result of many slightly innovative games that paved the way!)

Don’t think “The game idea isn’t really working, but once I add nice graphics and animations it will be fine!” Instead, ditch ideas where the system ends up being too ill-defined or too unpredictable. Nice polishing or adding more stuff won’t save that. At the very least, those ideas will never lead to games that I would find fun to make or play.

Instead, I try to ask myself: What part of life do I wish existed in a toy-like version?

What thing would be satisfying to turn into a well-defined controlled system? What can I simplify/minimize, so that only the satisfaction from playing with the system remains, and none of the noise/unpredictability/annoying stuff?

(I guess it’s a variant on the idea of the player fantasy they’ve always wanted to be able to do: What would I like to do but can’t in the real world for various reasons?)

I’ve found that the answers to these questions allow me to suddenly cut half my ideas—it’s a loooong list, believe me—and find more fun in what remains.

What do I want? I want to be able to do good, like healing or saving people. Can I reduce that to a simple system that you can completely understand and removes all the real-life reasons that would stand in the way? It would be satisfying to be able to do good this way.

What do I want? I always thought it would be cool to run a newspaper and investigate important stories. Can I turn that into a system that simulate it with only a few well-defined rules? It would be satisfying to create your own customized newspaper and see it grow.

What do I want? When I was young, I used to dream of made-up sports. Quidditch was probably the culprit here. The idea of a sport that’s just far too dangerous or impractical to be real. Just like most real-life sports are too intense, expensive, hard to play yourself. It would be satisfying to be able to experience the joy of sports or helping a team win the cup in a much simpler way that you can completely predict and manipulate.

What do I want? I want people to stop being stupid and ruining the climate. Can I reduce pollution and climate to a system that people can understand and play with without consequence? (As opposed to the real world, of course, where climate/weather is far too hard to fully understand, and there are all sorts of incentives to not reduce pollution or make sensible choices for the future.) It would be satisfying to be able to understand things better and achieve meaningful change (more quickly in-game than in the real world).

Conclusion

This is kind of two articles in one, but I had to get my thoughts out and I think the two topics are highly related.

TAKEAWAY 1: putting a first game on Steam taught me a lot about what gamers actually want from (specific) games. It made me realize that I mistakenly thought games had to be “not boring” by being “challenging”. When in reality games have to be “not boring” by being “a challenge this specific player knows and can understand”. This means the system has the right “mental load” for the target audience—some people play games after coming home from a long day at work, whilst teenagers can often play for hours with no other responsibilities. And you can reduce this mental load tremendously by having the game fit all the genre constraints that the gamers already understand/know.

TAKEAWAY 2: Games are fun because they do something the real world can never do. They have systems that are completely predictable, knowable, well-defined. Any game that makes their system too “realistic”, is likely not fun to play. Any game that does not communicate its system or does not define it well, is likely not fun to play. Games have the advantage of special effects (sounds, flashy animations, etcetera) to still make them appealing and fun to play for some people—but these elements have simply never worked on me at all. All my life, I’ve only really stuck with games with a simple knowable system.

Now I will try to keep this in mind. Stop making my games the wrong difficulty level for the genre/target audience. Stop adding things to make my system more “complex” or “realistic”, as this nearly always comes at the cost of predictability and understanding. At the same time, stop being mysterious/vague or adding no feedback about the system and all its interactions.

In a way, I guess, games are just a problem solving loop, a loop of puzzles. And our brain looooves a problem to solve—if we can actually feasibly solve it. If the scale of the problem is too big, we zone out. If the problem is unclear, then we don’t know what to solve. If the rules of the system are unclear, then we don’t know how to solve it.

But if you present a system just right, with simple consistent rules about how things interact, then our brains automatically start working overtime to think about how to play around with it and get a specific result. And then when we do get it, which should be somewhat regularly in a good game, we get that satisfaction of “having predicted X would lead to Y” -> “we were right, X actually led to Y, yay!”

Or just remember the deadly sin of game dev. Players that yell “why did THAT happen?” and “but why didn’t it do what I WANTED?”

If your game can avoid that at all times, I think you’re in a good place.

Pandaqi