Seven Rules of Game Design
that everyone forgets
The design of videogames, as a discipline, has been around for four decades now. There are hundreds of books, lectures, articles, blog posts and YouTube videos on the subject. You can find templates for concept documents, pitch decks, GDDs, TDDs, media plans, staffing breakdowns, swimlane charts, kanban boards. The list is nearly endless. But all of that technical and procedural acumen sometimes leaves things out of the equation that matter far more than whether your high concept statement is punchy enough.
Here are seven things that almost never make it into a game design document, but always should.
It’s their fifty bucks
In 1987, a lead developer and a game designer on SSI’s Pool of Radiance got into an… energetic… debate over the design of the final dungeon. In frustration, the developer finally asked the designer, “Dude… What do you think your job actually is?” Without a moment’s hesitation, the designer said, “To beat the player.” The developer thought for a moment and said, “I don’t think so, man. I think your job is to show players a good time.”
People like to say that the purpose of a game is to deliver an ‘experience’. That’s true, but it’s also fundamentally true of all entertainment. The word is too broad. Watching paint dry and getting kicked in the shin are experiences too. That doesn’t mean they’re experiences people actually want to have. So how do you narrow the field? That’s where remembering who’s paying the bills can help. That dev had a pretty good point. The contract entertainers make with their audience is simple. They give you money. You show them a good time. But that leads to some interesting questions. What, exactly, is a good time?
Oh here we go… we’re going to start theorizing about fun. Kind of… Let’s tighten it up a little. What we’re really talking about is two organic molecules called neuromodulators, specifically dopamine and serotonin. We know dopamine is connected to immediate short term rewards, what one designer likes to call ‘Push a button. Win a cookie’. It’s at the heart of every gambling mechanic ever devised, from loot drops to loot boxes. It’s quick. The spike is sharp. It fuels addiction loops.
But the other one, serotonin, comes into play differently. It’s connected to the brain’s mechanisms of learning and memory in ways that complement dopamine. Put simply, serotonin is what feeds a player’s sense of accomplishment and what designers have known instinctively for a long time is now being borne out by the science. Dopamine will get a player’s attention. Serotonin will keep them coming back.
And now you can see the machinery that makes games like Diablo, Path of Exile and so many others so potent. Little short term hits from killing a monster, getting a loot drop, finding a treasure chest. Pacing those moments about 10 to 30 seconds apart keeps that dopamine flowing. But then it starts to saturate. Every successive hit has less impact. Boredom looms. And that’s when beating a boss, leveling up or picking up a new ability resets the system. Serotonin wipes the slate clean and helps cement the memory of that moment of victory.
But there’s more in play here than simply punching the player’s buttons. This rule is really a cautionary note about who a game is actually for. When that designer said his job was to beat the player, what he was forgetting is that the game didn’t exist for him. His victory over the player is irrelevant. Its purpose wasn’t to showcase his cleverness. It had one job to do... show the player a good time.
RNG is a harsh mistress
Since the very earliest days of the pen and paper military simulations, when games like Tactics emerged from think tanks like the Rand Corporation in the late 1940’s and early 1950’s, random numbers have been used to introduce uncertainty into the simulation. Unlike chess or go, these games were deliberately non-deterministic because they were meant to force consideration of the unexpected into the decisions players had to make. Essentially, ‘What do I do if this goes horribly wrong?’ And it works. It turns out that the uncertainty of random numbers is incredibly useful for injecting doubt into a player’s mind. But at what cost? It turns out that RNG can be both a powerful motivator and incredibly frustrating for players. Here are a few things to consider when contemplating the use of RNG in a design.
An action that produces a variable result can have a strong and immediate impact on player psychology. Seeing huge critical hit damage numbers fly off of an enemy feels great. So does opening a chest and finding a rare coveted item. It’s pretty common for designers to keep these kinds of RNG events in the positive range. In other words, the player can ‘succeed better’ but they can’t ‘fail worse’.
Random victories feel hollow. This has mainly to do with how dominant RNG is in the outcome. Imagine a boss fight where there's a .0001% chance of just one-shotting the boss for an instant victory. The first time it happens the player might feel great, but what will happen is they’ll start intuitively expecting that insanely rare die roll and, if there’s a way to save scum their way to getting that result, there’s a very good chance they’ll use it. In other words, highly desirable, but rare random outcomes become the standard and the game stops being about playing and, instead, becomes an exercise in finding a way to induce those rare outcomes.
Random losses feel unjust. This is just the reciprocal of random wins. If, every once in a while, the player just gets clobbered by a rare failure. The only thing the design is doing is setting up moments of frustration. You can see this phenomenon all the time in games like Dungeons & Dragons where a player can critically fail a save and suffer some catastrophic outcome. Whether or not the player can recover from disaster, they’ll still walk away from the event feeling ill used. Which leads to…
Don’t piss ‘em off
Obviously… Except it’s probably the most overlooked consideration in game design. Mechanics and worldbuilding aside, what designers are designing is a user experience and UX has some well established principles that apply just as effectively to the design of games as they do in any other interactive implementation.
What just happened? If something goes wrong and the player doesn’t know why, that’s a guaranteed moment of frustration and, if it’s severe enough, it has a good chance of inducing a rage quit.
Too many choices. This comes up in RPGs all the time. When presented with a bunch of options, even if those options are explained, players can fall into a kind of decision paralysis that’s annoying instead of interesting. A good rule of thumb is to keep the immediate choices under five.
Why is this so slow? The standard for web content used to be three seconds. Now it’s down to one. Speedy response isn’t an option. It’s a mandate. But here’s the thing. If you include some kind of progress indication and connect the response to something the player is doing to get something they want then slower responses make the result feel more consequential. Crafting is the perfect example of this phenomenon.
That’s not fair. This is the random losses problem writ large. If a player fails because of a random number and there’s no way to ‘try again’ there’s a good chance they’ll just give up and find something else to do. And if there is a way to take a mulligan, they’ll probably do it, in which case your random negative result is meaningless. The way to mitigate this is to keep the consequence of random numbers contained so that they create some uncertainty, but never induce disaster with a single roll of the dice.
The best story in a a game is the one you tell your friends
Since the 90’s, AAA games have become increasingly narrative heavy and more cinematic in their presentation. There’s nothing fundamentally wrong with that except research shows that the stories that get people excited about a game have less to do with worldbuilding, dialogue or character development than a player’s own story about what happened to them in the game. It’s about where they went and what they did when they got there. It’s about their personal moments of surprise and delight. Those tales told by players to other players that are unique to games and they far outweigh whatever story the game is trying to tell.
The most important question you can ask about a game is…
“Why would my friends think I’m cool for playing this?” This doesn’t preclude other things that make a game great, but it speaks to something that, especially in the high noise world of digital distribution, can mean the difference between a great game that nobody ever sees and a breakout hit: word of mouth. In the post-influencer era, where consumers are increasingly wary of mavens, the impact of high profile content creators is waning fast. What’s growing is the importance of smaller groups of people who share a common interest. What’s really important about this question is the way it speaks to how people communicate among friends and acquaintances. If you’re afraid your friends will think you’re weird for playing a game, you might be a loyal fan but that loyalty isn’t going to find its way into your friend group because you aren’t going to talk about it. Word of mouth simply never happens.
Cheat like crazy, but don’t get caught
Anyone who’s played a pen and paper RPG like Pathfinder or Dungeons & Dragons knows that great game masters improvise all the time. Because those games aren’t about winning and losing. They’re about being in the moment and a skilled GM understands that fun is more important than correctness or adherence to the rules. But there’s a delicate balance to strike. Get too loose and nothing has consequences. There’s no risk which means rewards don’t mean anything. The dopamine and serotonin hits go flat. But the alternative, if it drives the player away through frustration, can be just as bad.
So… you cheat and the number of great games that find clever ways to bend the rules to improve the player experience is surprisingly high. Take the original Halo for example. It was the console shooter that revolutionized FPS gameplay on a controller… by cheating. Find a copy of Halo or Halo 2 and pay very close attention to the enemies. You’ll eventually notice that they are deliberately stepping in front of your gunsight. Why? Because aiming with your thumb can be frustratingly imprecise. So the game gives you a hand. In fact, Halo employs a whole suite of cheats to make controller gunplay as intuitive and fluid as the old mouse and keyboard model. A couple of examples:
Friction - As you traverse your view left and right with the right thumbstick, the game actually slows your slew rate just a tiny bit as your sight passes over an enemy.
Adhesion - When you have an enemy under your sight and you start turning, the enemy will deliberately move to stay under your sight for a fraction of a second. Just long enough to help you get off a shot.
And yes there are sometimes reasons to cheat on behalf of a computer opponent. This comes up a lot in strategy games where the complexity of situational and spatial awareness can be difficult for the AI to handle.
Ultimately, cheating is a powerful tool for enhancing the player’s experience but once the player catches you, the illusion shatters. Wins begin to feel cheap and losses begin to feel pointless. TLG is currently doing some experiments with LLM moderation that’s intended to fill parts of the role a live DM would fill. At the heart of that problem is the need for subtlety. There’s no doubt that, in some games, we’ll cheat. But we don’t dare get caught.
Real sucks but authentic is awesome
Games aren’t about realism. They never have been. And they shouldn’t be. They are, however, very much about feeling real. A few cases in point…
On Microsoft’s Combat Flight Simulator titles, because players were looking at the world through a screen instead of their eyes, the scene in front of them had to be rendered with a field of view that was wider than the actual angle between their eyes and the edge of the screen. That’s the same projection that fisheye lenses offer. You can see more but it’s distorted and, importantly, it makes things at the edge of the screen bigger and anything in the center of the screen smaller. In an air combat simulator that puts players in the position of chasing tiny little dots most of the time because the enemy planes in the middle of the screen appear smaller than they really are. The fix? Scale things in the center of the screen to be larger and make things as the edge of the screen smaller. It’s not real but it feels more like what you’d see if you were actually sitting in the cockpit. It feels more authentic.
Call of Duty has always cheated the physics of bullet trajectories, weapon recoil, grenade lethality radii and a ton of other properties of gameplay because making them numerically correct would make the game feel slow and clunky.
EA’s F1 titles deliberately bend physics in favor of the player to maintain playability, especially on consoles, to keep cars controllable. Tweaks include reducing steering sensitivity, greatly increasing lateral adhesion in turns, ignoring or greatly reducing the effects of curb hits and other events that would utterly destroy a real car.
The goal, of course, is to keep the illusion of realism while respecting playability. Even games like DCS, ARMA and the venerable Flight Simulator routinely simplify controls, physics and instrument displays because, even if they aspire to be ‘as real as it gets, they’re still games.