Let's start with the conclusion: at present, we can't draw the conclusion that chess wins first. Chess may be the first to win, the first to draw, or the first to lose (the specific situation is still unclear to human computers, but it must be one of them). The following is a detailed explanation:
First of all, two conditions are required to use the complete minimax strategy: 1. There are no random factors in the game, and every game is the same (just like poker draws different cards in each game, that won't do). 2. The game will definitely end in a limited number of steps. Chess meets these two conditions: 1. Obviously, the beginning of every game of chess is the same, and there is no process of shuffling and throwing dice in the middle. 2. Add some rules and restrictions, such as the repeated calculation of the draw, or how many steps to calculate the draw later, to ensure that the chess will end in a limited number of steps.
Then start building the policy tree. The first start is the root node, assuming that A is put first and B is put later. Since the beginning, A has moved many times, and each move forms a child node of the root node. For each child node, B has several coping styles, and the situation corresponding to each style becomes the child node of the child node ... and so on, and a "tree" is constructed. Because the game ends in a limited number of steps, all branches will win or lose in the end. Assign values to each end: A wins 1, B wins-1, and the draw is 0. Then calculate from the end forward and assign a value to each child node: assuming that it is a node's turn, A will of course take the largest value among several child nodes as the value of the node (that is, A is smart enough to make the situation the most favorable branch with corresponding coping methods); Suppose it's B's turn to be the next node, and of course B takes the minimum value among several child nodes as the value of the node. ..... Finally, if we push back step by step, the value of the "open position" situation must be 1 or-1 or 0, which can always be found in theory. If it is 1, it means the first hand wins; If it is-1, it means that the first hand will fail; If it is 0, it must be a draw.
There is a game in which the first hand wins and the first hand loses, and the game is bound to be a draw. Chess must also belong to one of them. It is impossible to determine which one through existing observations.