Rock Paper Scissors - A Method for Competitive Game Play Design
January 23, 2007 Page 3 of 5
At this point we have the ingredients for an interesting game. Depending on how well we implement the three criteria outlined in the previous section, we may have a very enjoyable game with a good amount of depth in the competitive game play. However, as players become extremely skilled at a game there is a definite optimum strategy.
First, let's review aspects of the game:
Each attack has a known and effective defense or counter attack.
Each attack presents a signal that is detectable.
Each attack occurs after the signal at a time interval which allows the defender to react.
And let's consider qualities of a skilled player:
He is able to identify an attack by its signal.
He is able to react in time to a signal with an effective defense or counter attack.
In this situation, if two skilled players are fighting, the defender is
at a clear advantage. This is problematic because there is no
incentive to initiate an attack - doing so puts the attacker at a
disadvantage as he "opens himself up" for a counter attack.
The fighting game Killer Instinct presented such a situation. A defender could defend against an attack simply by pushing the back direction to block. Most attacks gave some form of signal to the defender by hesitating or displaying an animation as the attacker prepared. Therefore, the defender could simply block and counter attack. The end result is called "turtling", or simply waiting for the opponent to attack. Therefore, The end strategy of a Rock Paper Scissors game with detectable signals which can be reacted to is to not initiate attacks.
Is Turtling So Bad?
At first thought it may seem like designing a game with the end strategy of turtling is undesirable; that is, designing a game in which players are encouraged to attack rapidly and randomly may seem like a better idea than designing a game where players are encouraged to not attack at all. However, consider when the strategy of each game is realized.
A simple RPS presents a system which has a very obvious strategy. Attacking randomly and quickly is very natural. Young children often fight by punching and kicking wildly rather than thoughtfully analyzing their opponent. For a player to reach the end strategy of turtling, he must become very skilled at the game. He must know the counter attack for every attack, be able to associate the signal to every attack, and be able to recognize the signal and recall the counter attack before the actual attack occurs.
Although Killer Instinct is notorious for encouraging turtling, it presented enough variety to keep players interested for many hours before realizing this strategy. If the amount of time that players spend playing a game is an indication of the game's quality, then the end strategy of turtling is more desirable than random, fast attacks from a designer's point of view.
Separation of Signal and Attack
A simple RPS game results in a strategy which is adopted without much learning and which is ultimately boring. A RPS game with signals also results in a boring end strategy, but this end strategy is only realized after a considerable amount of play time and learning.
This brings us to the question of whether it is possible to make a game which both encourages players to play and learn the game while resulting in an enjoyable end strategy. Of course, the answer is yes.
The biggest problem with a simple RPS game is that it doesn't involve any learning or skill. Players attack randomly and quickly. Of course, there is some skill in performing tasks quickly, but the best competitive activities offer more to the participants. The reason for adding in signals is to allow players to learn and utilize skill. While solving one problem by encouraging players to learn, signals introduce another problem of reducing the effectiveness of attacking - eventually to a point where initiating an attack should not be done.
Therefore, the next "addition" is one which will empower the attacker and restore balance to the game. This addition is the separation of the signal from the attack. A common term for this is "faking". The separation of the signal from the attack allows the attacker to give the defender the indication that a particular attack is to follow without being forced to follow through with the attack.
Any actual attack should still require a signal; that is, signals without attacks should be possible, but attacks without signals should not, otherwise the end strategy would be to randomly display one signal while performing a different attack. If the signal is completely separate from the attack, then the signal is no longer a signal.
Also, separating signal from attack should require some skill. Specifically, there are two considerations which should determine an attacker's skill at separating the signal from the attack:
A more experienced player should be able to separate the signal from the attack more effectively than a player who is not as experienced.
A player who is under less pressure when performing his attack should be able to more effectively separate the signal from the attack.
NBA Jam implemented separation of signal and attack by allowing players to initiate a shot but interrupt the shot by passing while in the air. The player cannot shoot the ball without jumping in the air, so attacks without signals are not possible. Passing while in air is not something which is commonly performed by beginners. This requires dexterity and some understanding of the game's mechanics. Therefore, beginners generally associate the signal of jumping with the attack of shooting. More advanced players are able to fake defenders by passing while in the air. However, when under pressure, it can be more difficult for an attacker to perform a fake shot, especially if his teammate is on the ground from a foul.
NBA Jam by Midway
The following is a likely "skill path" of two players who are learning an RPS game with signals and separation of signal and attack:
The players experiment with the game. They learn how to attack each other.
The players realize that each attack has a counter attack or defense.
One player may find one attack which is effective against his opponent.
After numerous attacks, the defender will recognize a signal for the attack.
By recognizing the signal and knowing the counter attack or defense, the defending player is able to react to the attack and stop the attacker.
The attacker realizes that the defender has detected a pattern, so he performs a different attack.
The defender learns the signals for other attacks and is able to effectively counter them. At this point, without separation of signal and attack, the players would reach the end strategy of turtling.
The attacker realizes that the defender is reacting to his signals, and with the knowledge that signals can be separated from attacks, he attempts to "fake" his opponent.
Initially, if the defender is conditioned to react to the signal, the fake will be successful, and the attacker will be able to take advantage of his opponent's conditioning.
The defender realizes that the attacker can separate his signal from his attack, and cannot rely purely on reaction to defend against attacks.
Page 3 of 5