August 13, 2009
Thoughts on Information in Strategy
"In fact, all the enemy's movements should be determined by the signs that we choose to give him."
-Tu Mu, notes to Sun Tzu's "The Art of War"
I was playing gin this evening when I got to thinking about the role of information in the game. To hear the really good gin players talk, the key to the game is information: what is in my opponents hand? If you can deduce what cards your opponent is holding, then you have a distinct advantage, and this is no different from any other card game. In fact, the importance of information is seen in virtually any game or sport where something is known to one side and not the other. For example, in baseball, if the defense knows that the batter will bunt, then they can bring infielders in to nullify the bunter's advantage.
To my way of thinking, there are really three types of information: What I have, what my opponent tells me, and what I tell my opponent.
Notice I didn't use the questions "what does my opponent have?" or "what does the opponent think I have?" There's a reason for this. You must always assume your opponent is of equal or better talent, or you will underestimate them and eventually make unsound decisions relying on the fact that your opponent is weaker. This is how upsets happen. If I am capable of choosing what information I give to my opponent, then I have to assume that they can choose what information they give to me. In poker terms, if I can bluff, then my opponent has the ability to bluff as well.
Type number one: what do I have? In card games, this is very easy: what cards am I holding? In other games and sports, the question is more difficult, especially when what you have are people. You have to know and understand your team's strengths and weaknesses so that you can form your strategy along the lines that give you the best chance.
Type number two: what does my opponent tell me? This is usually the step that distinguishes the beginner from the advanced player in many games. Every single action your opponent makes tells you something about what they have and/or what they want to do. In gin, this consists of paying attention to what cards your opponent picks up and which cards they discard. In other sports it's less transparent. An example in football is knowing the tendencies of your opponent. If the opposing team has run the ball in the same situation nine times in a row, it might be safe to bet that they will run when that situation occurs again. Of course, it's important to remember that your opponent has control over virtually everything they tell you, so you must take each piece of information with a grain of salt. How skeptical you are, or more appropriately, whether you can act on the information shown, depends on the arena and the situation. DIfferentiating between which of the information is true or false is a matter of skill and experience, however, and I have little to offer that can be applied to a wide range of endeavors. That is, except for considering how you yourself would act if you were in your opponent's shoes.
Which brings us to the third type of information: what information do I choose to give my opponent? This applies to how much or how little information you choose to give as well as what specific information you convey. This is, to me, the essence of strategy when information is involved. If you know all of the information that your opponent has about your situation, then you can use that to your advantage. If you know that your opponent knows your tendencies, then you can choose to break that tendency in a critical situation. Of course, the corollary is that you have to keep track of what information you give. As difficult as it is to keep tabs on the information coming in, you must also keep tabs on the information going out. Manipulating your opponents expectations can be a powerful advantage.
In other words, in many situations you have to allow your opponent to be right. In general, I don't believe in always giving away as little as possible. If you know what your opponent expects, and you give it to them, then it simply reinforces their expectation. Josh Waitzkin, in his excellent book The Art of Learning, uses the metaphor of martial arts. If you can get your opponent to lean heavily on one leg, then you can sweep that leg from beneath them when they try to rely on it. The key here is to realize that once you do choose to sweep that leg away, then your opponent will now be aware that you may attempt this tactic. Once you have done this one time, it loses some of its potency. It may very well work again, but likely not with one hundred percent success in the future. Therefore, it's important to wait until the opportune moment for maximum effect.
As you may have already realized, this discussion is not limited to games. The ideas and themes are definitely applicable to any other arena where strategy is necessary, which happens to be everywhere in my eyes. Perhaps not always to deceive and defeat, but also to give yourself the best chance in whatever you do.