Play of the master

Roa Zubia, Guillermo

Elhuyar Zientzia

Winning always has not much grace. That is why the game of the shepherd is not a national sport, it is easy to know what to do to never miss. Once studied, you could also play against the world's biggest expert without risk of losing a game. It would be good that it was something like this also to play chess. Well, according to mathematics, there is, and not only for chess, but also for many other games and for many activities that function as games.
Play of the master
01/12/2006 | Roa Zubia, Guillermo | Elhuyar Zientzia Komunikazioa
(Photo: D. D. Solabarrieta)

In the shepherd's game (three in stripe), almost all matches start the same way: the first player places his tile in the central position. It is the best option. Starting from the central position, the line could be formed in four directions, placing the tile in a corner, in three directions, and placed in the central position on one side, only in two. There is no need to be a mathematician to detect it.

However, the fact of starting with the central position does not ensure that the first player wins the match. And if the other player plays well he won't lose it. (Of course, that good behavior implies, for example, that the second players put the first tile in a corner, etc. ). To understand it, it is also not necessary to be mathematical.

But yes to analyze what is the most appropriate strategy of games. XX. In the first half of the twentieth century, mathematicians developed a theory to seek perfect strategies for the shepherd's game and other similar games. What's more, it was a theory to know if there is a perfect strategy: the theory of games.

Chess Chess Chess

The game of the shepherd is simple and therefore it is a good starting point to begin to develop the theory of games. But the theory was later carried by experts; for example, it can be applied with chess, with complex games.

(Photo: D. D. Solabarrieta)

Chess and the game of the shepherd have many differences, the majority very remarkable. One of them is the number of moves. In the game of the shepherd is limited, with a maximum of nine moves, since by then all positions are full. In chess, however, there is no limit to moves. When there are few pieces left, in theory they can remain "in return", without the game being finished.

However, for this to not happen, several tie rules have been invented in chess. For example, if in forty moves peons are not moved and pieces of higher value are not eaten, the result is a draw. No one wins. Taking into account the tie rules, mathematicians have calculated that there is a limit of moves that exceeds five thousand players. (However, this limit is much higher than the number of moves of any game).

On the other hand, there are some similarities between the shepherd's game and chess. Both are games between two players. In addition, for one to win, the other has to lose, so it makes no sense to establish alliances. And there are no hidden moves, that is, all games do them in sight.

John von Neumann

The Hungarian mathematician John von Neumann studied the games that meet all these characteristics, and in that study he showed that the theorem that initiated the theory of games, or at least, the modern theory of games that we know today. Von Neumann showed that for this type of games there is a perfect strategy. Not only in the game of the shepherd, but also in chess.

What is the perfect strategy to play chess? At the moment no one knows, mathematicians have not yet calculated this strategy. The only thing they have done has been to show that there is a perfect strategy. If they found it, they could schedule a computer that does not lose in chess, for example, but for the moment they cannot. On that way, the matches between the computer Deep blue and Gary Kasparov were very famous; the computer did not have an irresistible system to conquer Kasparov. Sometimes he won the machine, other times Kasparov.

The game of the shepherd.
D. D. Solabarrieta

It should be noted that mathematicians do not know anything about this perfect strategy. It exists, but they do not know what would be the result of a game applying the perfect strategy. It is clear that with the application of this strategy you could not lose, but perhaps neither win. In the shepherd game it is clearly seen; if the two players use the perfect strategy no one wins. Tie. Perhaps the same thing happens in chess, and perhaps not. Mathematicians assume that if the perfect strategy is the winner, white pieces would win matches because they have the advantage of the first move. But who knows.

This does not mean that the theory of games leaves chess worthless. Although the perfect strategy is calculated, it would only work in some conditions. In the game of the shepherd is clearly seen. According to the theory of games can not be lost, but there are those who lose it.

This is because not everyone always uses the perfect strategy. The theory of games only serves when the two players play as best as possible. It doesn't work with a player who doesn't mind losing or starting to experiment.

In addition, to apply the theory of games, all players must know all possible moves, remember the previous ones, etc. For a human being it is almost impossible to do it in chess, so there is no fear of finding a perfect strategy. Although they find it, chess will be an interesting game.

Three players

Not all games are between two players. On many occasions, three, four, ten thousand or one million people can participate. Of course, the more players participate, the more complex the mathematical analysis of a game will be.

If the perfect strategy of chess was known, it would have to be left to start the game, because the result of the game was known beforehand.
From file

Among them, the alliances. In the three player games, if an alliance occurs between two, the game becomes a game between two: a player against a partner. The increase of players implies more strategies of alliances. It's not a nonsense, in games with many players there are always alliances.

This can be due to economic issues of real life, in which millions of people defend individual economic interests. But the individual strategies are not used, but the group ones, since the money raised is often a winning strategy and a way to unite a lot of money is to combine the interest and money of many 'players'.

Real Games Real Games

The theory of games was applied immediately in what was not a game, especially in the economic and military. The negotiation between unions and companies is a game between two participants, the fight between two armies.

This does not mean that the theory announces a perfect strategy, but mathematical analysis contributes to the development of a convincing strategy. Neumann himself pointed out clearly, for example, that the theory of games does not serve to make money on the stock market, while in negotiations it may be useful.

At the same time, in the case of war, the theory does not guarantee that it will win an army, but it can help to choose objectives. For example, many believe the game theory helped Americans decide where to launch the atomic bomb during World War II. Logically, Kyoto was a very strategic city, but it was launched at Hiroshima. Why? It could be a matter of strategy, probably the biggest bombing the Japanese expected in Kyoto.

The economy is the ideal framework for applying the theory of games. However, it is not possible to earn money simply by applying mathematical analysis.
From file

The theory helps to perform the analysis of the game (situation) by means of a mathematical analysis. However, for various reasons it does not give a perfect strategy in the economy, war and many other 'games' of real life.

The first reason is that real games are very complex. They are not as simple in the approaches as the game of the shepherd or chess. As discussed above, alliances and the high number of players are involved. But it's not just that.

The second reason for not giving the perfect strategies is that the theory of games takes into account the ideal players, that is, the players who make the perfect moves. But in reality people are not like that, they make mistakes, use poorly designed strategies, and in many cases they do not take into account everything that needs to be taken into account. Therefore, the theory of games is not a mathematical model of reality.

Two children, a cake
(Photo: From archive)
Distributing a cake between two children is a serious matter. Just like how the cake is cut, one of the two children will say that it has touched the smallest part. One or maybe both. But there is a solution, there is a perfect strategy for children not to complain. Using the theory of the game you can reach the solution that a child cuts the cake and that the other choose the part you want.
The game of biology
At present, the most important applications of the theory of games are not economics and war, but have taken the most out of the theory in biology and sociology.
(Photo: From archive)
In biology, for example, it has been used to understand symbiosis. Why should the two living beings help themselves? Why does a crocodile decide not to eat in the mouth a bird that cleans its teeth? The biologists have analyzed this situation and know that both animals benefit from this activity and that, in the long run, not respecting the symbiosis would be counterproductive for both species. However, the crocodile does not make such general considerations. What does the future of the species matter to you? However, the crocodile does not eat a bird. Why?
In the 1980s, biologist Richard Axelrod found an answer by applying the theory of games to this problem. According to the results, the crocodile would get the best part to eat bird, if she ate alone and the other crocodiles would not. But if all crocodiles ate birds, they would be to the detriment of the whole group, they would disappear or never approach the crocodiles. And that's what crocodiles know. And for this reason, the team uses a punishment strategy to control all members of the team. Finally, crocodiles do not eat birds.
Puente Roa, Guillermo
Services Services Services
226 226
2006 2006 2006 2006 2006
Description Description Description Description
031 031 031
Mathematics; Biology
Article Article Article
Services Services Services
Babesleak
Eusko Jaurlaritzako Industria, Merkataritza eta Turismo Saila