Input vs output randomness

If we’re making systems driven games, there is a very high chance that there is randomness involved. However, when it comes to randomisation, there is a lot of nuance in how we apply it without frustrating players.

I think one very useful tool in learning to use randomness with intention in our designs is to understand Input and Output randomness, and it isn’t something I see talked about often. Let’s dig in!

Input Randomness

Input randomness is a random chance of something that happens before player input. For example, the hand of cards you draw in Slay The Spire, or the map layout of Into The Breach.

Input randomness can be great to promote strategy, asking players to react to a certain situation. Even if they get a bad run in this model, it can feel like their own mistake, as they had all of the information up front, but failed to solve the puzzle. Therefore this method can hide a fair bit of imbalance.

This method also encourages replayability. With a good system, each random input should add up to quite a different experience. We can keep things feeling fresh, but still skill based for the player.

Output Randomness

Output randomness is a chance of something happening after player input. The infamous example here being the hit chance in games like XCOM.

The risks here are obvious. We have all played a game like this and had the dice not go our way, which rarely feels fun, and humans have a psychological bias here too. We aren’t good at remembering odds, but remember bad outcomes more clearly. Therefore we don’t remember the 99% of shots that hit, only the 1% that miss.

However, this doesn’t mean that we shouldn’t be using output randomness. XCOM just wouldn’t be as tense without the dice rolls, and some games rely on chance implicitly, like gambling mechanics.

The Balance

Both serve quite different purposes, but go hand in hand when making a game that feels well balanced and skilful. Video games are largely about managing risk, so introducing chance is a great way to do that, a player simply has to make the most of a situation with the tools in front of them.

The key here is to match the desired player experience. If, for example, you are hearing from players that the game relies too heavily on RNG, then you consider the balance between the two methods. Maybe if players feel that they are missing too many shots, but you want that to happen so that they don’t win too frequently, consider adjusting the input randomness to make fights more difficult, but increasing hit chances.

Player Emotion

Really this all comes down to directing player emotions intentionally. Do we want the player to hold their breath (output randomness tension) or scratch their head (input variation).

Does it thematically make sense now that the player has only partial control over the outcome of a decision? Are there factors outside of their own abilities and characters?

Can we surprise players and make them feel good, without frustrating them when the poor outcome happens? Should we be allowing them to mitigate bad outcomes with rerolls, or does that lose the power of the moment?

One approach isn’t better than another, but understanding the definitions will allow you to have more in depth conversations with other designers, and use our resources with more intention, which to me is what makes a great designer.