I first learned of the game Cordial Minuet from a post on the gaming blog Kotaku. It was essentially presented as a skill based gambling game, where choices affected outcome more than chance. I was intrigued. Let's be honest here, who hasn't been tempted by the prospect of free money at some point in their lives? But when it comes to gambling, unless you have an edge against the house, it's a losing proposition. And with games like Texas Hold 'em, whatever skill may be involved is significantly offset by luck, and situation memorization (understanding the odds at any given time). Despite that glamorous presentation it is often given, playing it online typically devolves into a grind by which you gradual syphon money off unskilled players by applying your knowledge of the game's fundamentals. Assuming no one is cheating. In other words, you might as well get a real job.
Cordial Minuet however, seemed to be different. To me it actually seemed to have a lot in common with fighting games, which rely on a concept called "yomi". To those not familiar with fighting game jargon, yomi is the concept of predicting an opponent's future actions, based on past actions. For example, if you were playing paper rock scissors, and you knew that the other player had a bias towards picking rock the first turn, you should always pick paper the first you play with them, because this will give you a statistical edge. And then, if/when they figure out you are always picking paper the first turn, they will start being biased towards picking scissors instead, and so on. Essentially, you recognize patterns in the opponent's behavior, and you adjust your strategy to take advantage of those patterns.
To someone who doesn't realize you are doing this, it can at times seem like magic. They will ask, "How do you always know what I am going to do?", and you might reply "I can see it before it even happens." Magic. Except really, you are just building a probabilistic model of the other player's behavior in your head, and you are using that model as a predictor of their future actions. You can even manipulate the other player by doing things which you know they have a predictable reaction to, and then countering their response. For example, if someone has a habit of always using scissors after they use paper twice in a row, you can counter that by picking rock. The more details you can hold in your head regarding the other player's patterns, the more effortlessly you can counter their actions. And when both players understand that this is what they are doing, they will strive to be less predictable, while still taking advantage of patterns in the other player's behavior.
Now let's take a step back and look at game theory as a whole. There are essentially three factors that determine success at any game. The first is luck. In this context, we are just going to define luck as the part of the game we don't have control over, such as the roll of the dice in a board game. In theory, there may be things we could do to take control of these factors, but let's assume we're not going to do that. The second is efficiency. In other words, how fast/accurately you can perform an action in a given amount of time. So say if you had a contest to see who could press the button the fastest in sixty seconds, this would be a contest of efficiency. Efficiency also applies to reflexes, which determine the minimum time it takes you to respond to something another player might do. Then finally, we have prediction (or yomi), which was discussed above. Games are all more or less designed around these three factors, and the balance of how these three factors interact with one another determine what skills will be most applicable to a game.
In a casino game like the slot machine, luck is generally considered the biggest factor. Because you are unable to predict and adapt to the behavior of the game, you are powerless to affect the outcome, and therefore the correct strategy is not to play. It may be theoretically possible to predict the game's outcome at a given time, but since it is not practical for you to do so, you are left at the mercy of the game's odds. In a game like a music game (think Guitar Hero), efficiency would be the biggest factor. How quickly you can interpret the flow of input from the screen into accurate physical reactions will determine your skill at the game, and you improve mainly by training your mental and physical response to the input you are receiving. But then, we have prediction/yomi. This is particularly interesting, because when you talk about competitive gaming, this aspect is often the one that ends up being the most significant. You see, in order to have a skill curve in a game, you need to be able to get better by learning. Luck won't do us much good here, because we have defined luck as things which cannot be predicted by the player. Efficiency can be improved by training, but this alone will not make for a very interesting game. If both players are competing only on efficiency, they will tend to reach a plateau in their skill level and then lose interest. But if we are competing on prediction, the game will turn into a constant battle to one up the other player, with a potentially diverse set of strategies.
There is another factor that should be mentioned, which is the requirement for outside knowledge. This is not the same as the other factors however, because it is basically just an advantage, that will disappear as you play the game. For example, in a trivia game, you may initially be at a disadvantage if you weren't familiar with the subject material, but eventually you would learn the answers to questions, and would be on even footing with a more knowledgeable person. The amount of time this takes of course, would be proportional to just how many possible questions there were to begin with. This could in many ways just thought of another form of experience, which states that you will be better at a game the longer that you play it, because you will learn the probability of the game's outcomes, and become better at predicting them. But this is different than predicting the other player, as it applies to only the game itself, and not the person you are playing against.
Part of the reason Cordial Minuet interested me, was that all the skill of the game seemed to be placed on prediction. Unlike games such as Texas Hold 'em, in which a lot of the strategy has to do with specific situations and the probabilities of the outcomes of those situations, Cordial Minuet was purely a game of anticipating what the other player would do. I later learned that there was a degree of luck involved, since certain board layouts you may play on could be slightly more advantageous to one player than the other, but there is always a way to win, regardless of the board. In other words, think poker, where there is essentially no such thing as a bad hand. Sounds interesting? I thought so, at the time.
So what could go wrong? Stay tuned for Part 2: The Game...
No comments:
Post a Comment