Predictability – Embrace The Chaos!

A while ago I mentioned that chance is an element of game design.

I was totally wrong.

Chance does not qualify as an element of game design… It doesn’t have a big enough effect on the game. PREDICTABILITY is another matter.

Ok, let’s split this up, CHANCE is an approximation of how much your game relies on the generate random number function. UNPREDICTABILITY is an approximation of how accurately the player can predict the next events in the game.

Unpredictability is far more important then chance, in fact, it is an extension of chance as it covers everything chance does and more.

To remove ambiguity the definition I am using is effected by not just about how many ways an event can go but how different those paths are. If a man came up to you and tossed a coin, heads you live, tails you die, then that would be a highly unpredictable situation, even though there are only two options the difference between those options is massive. If he were to come up to you and say “Heads, you live – tails, I kick you in the nuts” or “heads, you get beaten to near death – tails, you die” then the uncertainty is decreased, because one way or another you are going to either live or suffer immense pain. If this was a videogame, the adrenaline would probably be higher the larger the stakes are (because in real life the threat of death is gonna make you panic one way or another).

I am using unpredictability rather than predictability because it is possible to have 0 unpred (one outcome has a 100% chance of happening FTPP (From The Players Perspective)), but it’s impossible to have 0 pred, (every single possible outcome has a 0% chance of happening FTTP.)

Unpredictability comes in 4 flavours.

PRE-SET
“Pick a card, any card…” But you know which card is the ace. Pre-set chance is destined to happen the exact same way one way or the other (provided the choices made are the same)… This is apparent in storytelling. The player doesn’t KNOW how the book is going to end (unless the book sucks or you have already read it) – it is unpredictable, but it is always going to end the same way. In most text adventures this also happens – let’s say there is a blue door and a red door. You know where each door leads but the player does not, therefore to the player the outcome is unpredictable. However, once the outcomes are seen this chance evaporates away.

This is the only real occurrence of unpredictability in storytelling, and it must be noted that twists do not count in this. If a player is nearly certain the outcome is going to go one way, there is a lot of predictability. If it happens to not go that way, that doesn’t matter. (Though it does heavily decrease the predictability of future events, because the player is now unsure of whether they will be faced with another twist or not). What I am trying to say here is any four of prediction rests entirely in the user’s head. It is about HOW CERTAIN THEY ARE that things will go one way or another.

As an extension to this there comes ACTIVE unpredictability – the kind of chance you see in most action games. Remember when you got stuck on a level, and you end up repeating the same sequence of moves over and over? This is a form of pre-set chance, as it wears away after repetition, but unlike the kind associated with storytelling it is not thought out and decisions are usually made on the fly. This sits somewhere between Pre-set and Human error. I’ve decided to put it with pre-set.

CHAOTIC
The least predictable of the unpredictabilities, this is basically what chance was. The use of the generate random number function. Though it could be described a bit better, chaotic chance is unpred that does not change at all between games, it does not vary between other people and the player cannot do anything about it. It is also the easiest form to manipulate as a game designer as it’s not that hard to add chaotic chance in the form of random dice roles and stuff…

Basically, if you want to add unpredictability and can’t think of a better way to do it this is your best option. Unpredictability adds tension into games, keeps the player on their toes, and just in general makes the game more fun. And this is by far the simplest way to add that little “zest” games need.

HUMAN ERROR
You get this flavour in the heaviest doses when playing platform games. Remember that panic when you are not sure if you are going to make that jump or not? The existence of uncertainty means there must be some form of unpred at work here. There is.

No human can do anything perfectly – there is a margin of error to every action you make. This is natural human error and it is often hard to separate from pre-set.

Let me put it this way – you making the jump in a platform game is mostly pre-set chance with a bit of human error. You know this because the more times you repeat a jump the better you get at it. That’s the pre-set chance wearing off, (pre-set wears with repetition) however, some of the super-hard jumps you do just don’t seem able to be done as this is your natural human error shining through. Human error is evil, and any game that relies on it too much should be burnt in a pit of fire.

H.E. varies massively between players, the more practice you get at it the better you become, (note that this is not the same as Pre-set, it doesn’t necessarily have to be the same jump you practice on, – also, this happens over a very large amount of time.) and because of how closely bound it is with Pre-set, a jump that may be near impossible due to H.E. will be practised again and again by the player who assumes that this jump like the others will become easier. This makes the game very stressful for the player, and destroys the fun.

If even 10-20% of that jump is human error you’re getting pretty hard, especially if this jump is going to be made a lot of times. At 50% you had better make sure that jump is only made once, I don’t think you can get 100% but if you could it would be insanely difficult.

And finally…

OPPONENT
This is the best for of unpredictability, this is the unpred that occurs when faces with a live opponent.

Now, some of you may have heard of game theory, I’m not talking about articles to do with game design, I’m talking about an actual psychological study that was done. Let’s talk about the prisoners dilemma.

Two criminals are arrested and they are taken to separate rooms and both told if they snitch on their mate they will get to go free while their mate will get 3 years. However, if both of them snitch they will get 2 years, and if neither of them do they will get 1 year. It is beneficial on both sides to co-operate, but doing so makes you vulnerable to attack, because if you know you’re buddy isn’t going to snitch you should. If you get people to repeat this game over and over you begin to notice something. People’s decisions are totally random.

What does this mean? Well, people spend so much time trying to get into each others heads that what they do actually boils down to random selection, however, that doesn’t make it any less demanding. Opponent unpred gets players to use their full mental capacity but is still totally random, thereby challenging the player and rewarding them. Often in games with this flavour one player does predict and outmatch the other player (primarily when the other player isn’t paying attention) and therefore wins via skill.

What does this all mean? Well, unpredictability is required for everything. You can have too much, yes, but you always need some. If you can see things before they happen the fun drains away, everything drains away, and you are left with nothing.

Embrace the chaos, because it is hiding in the places you would least expect.

Thank god it is.

About The Author: Joseph, or Lawsome, as the internet folk call him. He spends a lot of time making games, most of which fall apart or don’t work and are never published, but the few that survive can be found on his account at Yoyogames – http://sandbox.yoyogames.com/users/Lawsome1997. He mainly enjoys writing about game theory but you’ll see him do a few reviews. He avoids games that look generic and would rather play something original than something fun. He has strong opinions on games and can hold his own in an argument, if you tell him that COD MW3 is the best game ever he may bite your head off.