Tuesday November 24 2009
Counter-Strike
Official Design Partner
Story Header

Counter-Strike: Checkmate: Good Game! Issue #3

By: Michael Lau - Published January 19, 2006 at 1:51 AM EST - Writer Archive
In a game of chess, there are 64 squares and each player has 16 pieces. The fact that white starts first means that white has 20 possible moves - the 8 pawns moving either 1 or 2 squares up the board, or the 2 knights moving either left or right. For the black player, there are also 20 different possible moves after white advances its piece. Again, back to white's second move, white has around 20 moves to make depending on its first move, and black has 20 moves or so to make after white takes the second move.

If we were to think 3 moves ahead right from the beginning, that means that there are 8000 (20*20*20) possible moves we can make in only 3 turns. Depending on what black does in its turn, this number increases. To fully extend the number of moves possible in a game of chess there are about 1,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000 or 10^120. To provide a comparison, according to my science book, if we were to exclude dark matter and other quasi-factualized matter then there are about 10^66 atoms in the WHOLE universe.

What makes computers so much better than humans at playing chess, then, is the fact that they can calculate the probability of moves using a method known as minimax algorithm, in which the computer creates tree diagrams (yes, the stuff we learn in high school) to calculate around 15 ply ahead. A ply in chess means 1 turn, when white and black both makes their move. Chess players at my level can think around 8 ply ahead (25600000000 moves) whereas Grandmasters like Gary Kasparov can think 13 ply ahead.

As a matter of fact, all these numbers are meaningless, they are only percentages of probability on what the opponent might be doing based on simple probabilistic simulations. How we act on this information is an entirely different level of abstract thinking that makes humans able to defeat computers (Kasparov - X3D Fritz), because players such as Kasparov can either play 'Kasparov Chess' or 'Anti-X3D Fritz Chess' creating something even beyond information processing.

How, then, is this whole mumble jumble related to Counter Strike? To be quite honest with you, the main reason this whole article is delayed is because I had problems finding an answer to my own question. What I realized, however, is that by taking Probability into consideration, it works as both a mindset and as an easier method of playing the game.

Thinking about this deeply, a game of Counter-Strike consists of 10 players and a near inifite number of squares to move on. Likewise, the moves of one round do not greatly influence those of another round. However, by knowing an opponent to be at a certain location, we know immediately the probability of adjacent moves it can make from that area. Extrapolating on this, we can consquently minimize the proability of the moves of the rest of the opponents. By looking at this new piece of art, we can act upon it to maximize our own probability of success.

Take this, for example: consider you are in a two versus one situation on a map such as Dust2. You and your teammate are on the Counter-Terrorists side and neither of you have a clue where the last enemy may be located. If both of you split into opposite bombsites, the probability of the enemy ending up at either bombsite is 99% (1% if the enemy decides to save). However, what is your percentage chance of success?

Now, look at it this way. If you and your teammate stick together and choose one bombsite by making a probabilistic simulation from past experiences of similar situations, which leads you to believe that the enemy has a higher probability of going to that site, then your chance of having that enemy coming to the site you are at is approximately 50.5 - 70%. With this in mind, your chances of success would be much higher as you and your teammate would be together when the gunplay starts.

Of course, I’m not here to teach the game of Counter-Strike, I’m only trying to raise awareness. However, many a times making a probability simulation such as the one above is done unconsciously, based on experience and game sense. Developing this thought process in individual game play as well as team strategies is only a matter of taking this art into consideration.

By playing your own game or the 'anti-enemy' game you have roughly a 50-50 chance of success and vice versa, but realization of the factors involved allows us to manipulate this percentage in our favor.

There will always be that one strategy, one stack or one move with the highest probability of success, so of course we'll always want to take the higher possibility than the lesser, right?

Or do we...
Page:

User Comments

- 54 Comments

» This story has had 54 comments posted since January 19, 2006 at 1:51 AM EST.

Latest Poll