Output Randomness: Game Designers should know how to use it.
And ways to make it feel "fair"
It’s natural to assume that randomness is inherently unfair. After all, randomness is the degree to which factors outside your control affect your performance. I love putting randomness in my games, but I also want my players to feel that the game is fair. How?
I’m not sure if it’s possible for me to write anything without referencing Slay the Spire.
In his foundational gametek segment Geoff Engelstein defined the incredibly useful terms Input and Output randomness. Input randomness is when something random happens before you make a decision. In Slay the Spire, you draw a hand of cards and then decide what to do with them. That’s input randomness. Output randomness is when something random happens after you make your decision. When armies clash in Risk, you roll dice to see who wins. That’s output randomness.1
Input randomness is a powerful way to make things feel fair and everyone uses it. Secret objectives in board games, variable starting conditions, hands of cards. Everybody loves input randomness. It’s harder to love output randomness. It feels bad to make a good play and get completely screwed over by an unlucky roll of the dice. However, there are great rewards waiting for designers who are able to harness output randomness in their games. With output randomness, you get to hold the player in a state of uncertainty until the last moment. It’s exciting! So let’s set aside Input Randomness. We know that works. Is there a way to make Output Randomness feel fair?
Pictured: Me getting absolutely destroyed in Slice and Dice.
Slice and Dice is a game with, shall we say, a LOT of output randomness. Every turn hinges on the roll of the dice. Yet, Slice and Dice feels fair. Why? Slice and Dice allows you to influence the odds of getting different results. Between each fight, you get the chance to level up and add equipment to your heroes, changing the faces of their dice. Within each fight, you also get to “lock” your dice to keep their results. When you’re really pushing your luck with a final roll that has a chance to go great or terribly, you’ve usually passed on a chance to guarantee a medium result. That makes the result feel more fair.
Slice and DIce has another trick up its sleeve. Seeing the literal physical dice rolling on the screen makes it feel more fair. Zach Gage said it best on the eggplant podcast:
“I think there's a magic to a visual understanding of a dice that cracks the problem that humans have with understanding odds, because things that we can feel and touch and hold work differently in our brains than things that are in the realm of pure mathematics.“
- https://eggplant.show/into-the-depths-slice-dice-part-1 (timestamp 1:30:00)
the tower is a beautiful game that lives in my mind always.
the tower is an absurdist game about climbing a tower of stochastic chutes and ladders. There is a magic door on each floor that determines whether you climb to the next floor or slip back to the previous one. The doors work thusly: roll a 100 sided die and compare your result to the number of the floor you’re on. If you roll higher than that number, you proceed. If you roll lower, you slip. Due to the rules of probability, tower climbers tend to congregate around floor 50 (there’s a party there) and it is practically impossible to reach the top without cheating. The “dice rolling” in the tower is imaginary, but it still feels fair to me. First it displays the goal, then it reveals what number you got. The typing effect makes it feel more weighty, more real, even though it doesn’t have a physical dice simulation like Slice and Dice. Simply put, players are more likely to accept randomness in the form of a visual process.
This is slipways classic made in pico-8 with, I assume, some kind of black magic.
Slipways is a game about building an interplanetary trade empire. You start with nothing but 100 space-bucks to invest in colonies across the universe. Before you colonize a planet, you must first find a planet to land upon. To do this, you locate gray maybe-planets and scan the vicinity around them. The maybe-planets can turn out to be a planet or just nothing. You must scan them to find out. Mathematically, it would be equally correct to say that these maybe-planets have a random chance of disappearing when you scan them, but it doesn’t feel that way to me. It feels like I’m discovering information that was already true. This subtle difference in framing makes the game feel more fair. Minesweeper is a classic example of this principle in action. It feels like the bombs were hidden there before you started playing, and you’re merely discovering them.
To conclude, these are four ways to make randomness feel fair:
If you’re a coward, input randomness (allow players to make decisions after the results of a random event).
Allow players previous decisions to influence the odds of the random event
Show the random event happening as a visual process
Frame the event as discovering something that was already determined
Go forth and make random games!
- Ezra Szanton
Subscribe for more thoughtful game design deep dives.
As Isaac Shalev points out, all input and output randomness bleed into each other at some level. The output randomness of last turn is the input randomness of this turn. As a nod to this cyclical understanding, he suggests referring to them as “clockwise” and “counterclockwise” randomness. While I appreciate the underlying truth these words are trying to convey, I find them to be more difficult to think about and explain (why is clockwise input randomness and counterclockwise output randomness? I find it hard to remember which is which, and I want the words I use to help me think not be a hinderance.) Here is a link to his article if you’re interested in reading further: https://www.kindfortress.com/2018/10/03/design-patterns-random-loops/