Solving Treblecross

During the summer, my project is involved in researching the field of game theory, primarily focusing on a game called “Treblecross”. This game is played with two players and starts off by drawing a 1xn board on a sheet of paper.  Then the first player places an “X” in one of the empty boxes in the board followed by player 2. The first player to get three “X” marks in a row wins. The objective of this project is to figure out if player 1 or player 2 has a winning strategy with a certain sized board. At first I tried figuring this out by hand but then quickly realized as the board got to be 1x10 that it was very difficult to do so, an adaptive learning computer program that would play the game was devised. An adaptive learning program begins by having the computer place the “X” at a random position. The computer learns how to play the game based on a move that it has made. If an outcome consists of a win, then the computer learns to do that move more often otherwise, the computer learns to avoid making a move that leads to a loss. I programmed two computers so that they can play against each other thousands of times. So far, a pattern has emerged that tells me who wins with a certain sized board. I will continue to investigate this pattern and see if it continues throughout the summer.