Bad Luck?

About random elements in games

It's been a number of times that our group entered into a discussion about the amount of luck that is involved in a particular game and whether an element of luck is good or bad for a game. Our recent experience with "Roaring 20's" triggered me to write this article. First of all I want to take a look at the various kinds of "luck" that may have an influence of the outcome of the game.

An often used element of luck introduced in games by the designers is that of rolling dice. Here, the designer explicitly decided that some kind of non-deterministic element will enrich the game. Very often this concept is implemented in order to reflect real life situations where the probabilities of events are not 1 (i.e. 100%). Good examples are role-playing games or military simulations.

In other games dice rolling is used as a main feature of the game. Good examples here are "Can't Stop" or "Liars Dice/Bluff". In these games dice suit the purpose to introduce a kind of uncertainty in order to create challenges and suspense in the game.

Dice in these games introduce a probabilistic factor which can be calculated using straight forward math. For any given game situation all players are able to calculate their chances of a "good" roll. Since we are dealing with probability rather than certainty there is of course a risk involved: it's either good or bad luck what you rolled...

The concept of probabilities can be implemented in as simple a way as flipping a coin: there's a 50% chance that it will show your choice. A small extension can make this mechanism far more interesting and challenging: "Rock, Paper, Scissors". The probability of success still is 50% in a single throw but the whole thing provides a lot more suspense than just flipping a coin. First of all both players have to select between three rather than two choices. And then there is the additional possibility for an extended game: by throwing several times (say 10) and counting victory points for every throw players are tricked into believing that there may be some strategy that can be applied for improving one's overall chances. By the way, a nice site with lots of details on RPS is http://www.worldrps.com/. Games like "Can't Stop" and "Liars Dice/Bluff" have brought the concept of probabilities to near perfection by adding psychological effects like greed (Can't Stop) or bluff (Liars Dice) into the designs.

Another very common element for adding probabilities to a game design is a deck of cards. Cards provide game designers with unlimited possibilities for their designs. Considering a deck of say 52 different cards the probability of drawing a particular card is now also dependent on the cards known to be in play. In addition, players usually have a hand of more than one card providing them with a choice of which card to play. Probabilities still can and must be calculated, but the calculation is somewhat more complex than it is for dice.

Be it cards or dice, there are games which simply are fun to play although players are not or not fully in control of their chances to win. In my opinion these games all have single characteristics: they are very well balanced and they provide a "feeling of being in control". "6 nimmt/Take 6" for cards and "Can't Stop" for dice are perfect examples. Unbalanced games usually have a very uneven distribution of winning combinations, i.e. there are cards/dice rolls which lead to certain victory or defeat. Situations like this make players feel that they are being played by the game rather than that they play the game, since their fate is completely dependent on that "lucky draw/roll". A good example of an unbalanced deck of cards in a game is "Roaring 20's", which we reviewed recently.

Then there is a completely different set of luck elements that can be present in a game. A name for this set could be "player interactions" and the most important elements of the set are: seating order and start player, players' gaming experience and skills and last but not least players' diplomacy skills (like bluffing, negotiating, whining).

All of these elements add the possibility of introducing random effects into the game which are in many cases hard to predict and often not easily taken care of.

Seating order and start player usually is determined randomly either by the gaming group themselves or by some means the game designer has defined. The mere fact that there is some randomness involved in determining both does not mean that seating order or start player position introduces any kind of luck element on their own. But in combination with the other two elements of the set does seating order pose a great potential for randomness of the game result in any game with more than two players.

To clarify the point let's look at an example. Assume the game group plays a four player game and the player sitting behind you is a novice to the game. In this scenario any blunder made by the novice most like will be to the benefit of the player seated behind him rather then to you, simply because two other players will have the chance to react to the blunder before it's your turn again. So the random (luck) element in this situation is whether or not the player you are leading is going to make a blunder and whether or not the two players leading you will be able to exploit it. This situation is what Walter quite rightly termed the "chaos element" in a game. There is no way to determine the exact probability of a blunder. And even if there would be it is not really advisable to take the possibility of blunders into consideration as undoubtedly you will not play to your best ability. If you do and the blunder is not made your turns are suboptimal just as well as if you don't and the blunder occurs. This leaves you with a 50% chance of a suboptimal turn while the two players leading you have a 100% chance to exploit the blunder.

I dare to say that most gaming groups are inhomogeneous in terms of skill and experience regarding a particular game and therefore this situation will arise often enough to need seriously consideration by game designers because it is almost impossible to devise a sound strategy for it.

A second source of randomness due to player interactions is diplomacy in its widest definition. If and when a player decides to whine about a bad move or position just to convince other players to play less offensive towards him is totally beyond the control of the other players - just as well as the results of this on one's own position in the game. Threats, bluffs or "gentlemen agreements" raised by a player against others cannot be planned for either and they just as well generate a random, chaotic element in a game.

In my opinion good games (designers) will take these factors into account when designing a game and in fact the introduction of clear rules regarding player interactions is a prerequisite to make sure that the game is played as intended. On top of this, game designers often will deliberately introduce random event in there designs to lessen the random effects of player interactions.

In conclusion let me summarize:

View/add comments here
Aaron Haag
04.05.2003