Site hosted by Angelfire.com: Build your free website today!

Chyss Chess (Home)

Chess Thought

Chess Poll


Are Modern Players Better Than Those From The Past?
view results



How Chess Players Think

One of the most commonly asked questions when people begin to play chess is how far ahead good chess players look or calculate. This varies a great deal, not only from player to player but from game to game and even in different positions during the same game. In general good players look between 3 and 5 moves ahead by both sides but will concentrate only on the plausible moves. It is possible, when very little time is available, to play entirely on instinct, and still play good moves. This instinct comes from experience. It is often useful to compare the way in which human chess players think and the way in which computers calculate. The number of positions which can occur after only three moves by each player is well over 64,000,000, there are far more possible games of chess than there are particles in the known universe, and the number of chess positions, though far fewer, is still astronomical. Some computers can analyse huge numbers of positions; Deeper blue for example was able to consider 200,000,000 positions a second. Quite clearly it is impossible for a human to calculate such a massive number of positions and so he relies on instinct and feeling (which come from experience) in order to cut down the number of positions.

The way in which a human player calculates is fairly straightforward. Players construct what is known as the tree of analysis, where each branch represents a different position. It usually appears disorganized. The reason is that two variations can sometimes lead to the same position and that a player will not always calculate the same number of moves from each position or in the same depth. Only a small number of positions are considered at each level. This rejection of the vast majority of available options is known as thinning the tree. It has been estimated that humans will only consider around three dozen positions before choosing a move.

The actual process of "looking" at positions is not a case of thinking "I have a pawn there and a bishop there and if he goes there then I go here and I will have a rook there and a knight here" etc. After someone has played a number of games the important characteristics of the positions become obvious to the person's mind and little conscious effort is needed to hold positions close to the one in front of you together.

Humans do not choose their moves solely on the basis of cold logic and calculation. Players usually have a plan around which specific variations can be constructed. This plan is usually based on position considerations. This sounds complicated but in fact it is little more than a gut feeling which is usually innate and is extremely difficult to teach or explain. Attempts to do his usually consist of ideas of thousands of positions from previous games flashing past the minds eye and making an almost subconscious impression which can somehow be detected by the player. Positional considerations are things such as weakness of King position, and space available for manoeuvring.

The most important element of a chess player's thinking is how well he evaluates positions. That is, how well he can decide who is winning and why. Again an analogy to the chess computer is useful. A chess computer program will first check if the position is the end of the game for example checkmate, and if not it will add up the pieces giving a numerical value to each position. It will do this crude evaluation in each position it looks at and will choose a move based on which sequence of moves gives it the biggest end number. A human player has a much more advanced evaluation function. In most cases he will carry most of his evaluation over from the last move editing it as appropriate to save time. He will check the material balance, consider the strengths and weaknesses of the relative positions of the two sides, both in terms of long term and short term factors, and will decide what plan each player will form. He will then select the sequence of moves which will either;
a) give him a material advantage (more pieces) or
b) give him the more straight forward and reliable plan.
The conclusion must be that a chess player's thinking is far more general than is usually thought, and far more effective and advanced than the way in which computers play.

For more on how chess players think check out these books at Amazon:
'Think Like A Grandmaster' 'Three Steps to Chess Mastery' 'Secrets of Practical Chess'


Tactics

Relatively speaking, this is the easiest element of chess to teach (which is why for many youngsters this is the strongest area of their game). It relies a great deal on pattern recognition and cool calculation, both of which can be learnt (to a certain extent). The general advice is to play through as many games as possible, thinking for a moment about each move and trying to pick up on is tactical implications. (For a selection of games, click here). Tactical weaknesses which are most commonly taken advantage of are pins, loose pieces, and pieces set up in such a way as to allow a fork or skewer to take advantage of them. Try to avoid them in your own camp and to create them in your opponent's position. To see some examples of relatively simple tactical tricks click here. More tricks and traps can be found here.


The Tree of Analysis

This is how humans and computers calculate complex variations. In the diagram we can see that some branches are taken further than others as some variations need to be calculated deeper in order that an accurate assessment can be made. Clearly this diagram has been greatly simplified for reasons of clarity but it shows to quite good effect how the mind works when considering a move. From the bottom; the right hand branch may be a greek gift (to read more about greek gifts click here) and the left a positional manoeuvre. The next branches on the right represent the possible defences and one has already been evaluated. On the left little research is necessary as positional moves rarely require a great deal of analysis of concrete variations.
(The tree of analysis)


Advice for Chess Players

The best advice for chess players is to play lots of chess, particularly long play chess, as this develops your ability to calculate deeply and reflect on the salient features of the position. Experience is the best instructor. Karpov once said that to become a really strong player you must lose thousands of games against very strong opponents. Thus, when you play, try to play against the strongest players you can find, and if they are willing, get them to explain where you went wrong during your game. Self help books are all very well but their advice is of necessity somewhat general in nature, thus you would gain more from discussing your games with very strong players than from reading books. A piece of advice that everyone is given and which has become almost a cliche is that you should not spend too much time studying openings. As important as they are, it is more important to understand general ideas about the various stages of the game. Play as much serious chess as you can and this will lead to an improvement in your play.
There is one piece of advice which I have found more useful than any other when playing chess. This is that if you feel that one move is the best move, intuitively, and you can't find any reason why it should be, just play it! If you spend time analysing it you will only end up finding some half hearted justification for it and will have wasted time thinking about a move you were going to end up playing anyway. If it all goes wrong you will know next time to have less faith in your intuitions.


Thinking Ahead in Chess

The best way to improve your ability to think ahead is to practice thinking ahead! I know this sounds obvious but it is still a question that is often asked. Take a critical and highly tactical position from a deeply annotated Grandmaster game (for example from a games collection), spend 20 minutes analysing in your head, and then compare your conclusions with those of the Grandmaster in his notes to the game. Do this with as many games as you have time for and you will 'soon' find the calculation of variations much easier. Choosing tactical positions is best as there are deep but narrow lines to calculate which will help you to get 'further from the board', i.e. to calculate to positions which are further from the present position on the board. (See above, the tree of analysis shows one way, [arguably the best way] of structuring your calculations). Once you have mastered a system of calculation such as that described by Kotov you will not find the analysis of complicated chess positions so difficult.


How to Evaluate Chess Positions

Apart from reading books like "judgement and planning in chess" by Dr. Euwe and "My System" by Aron Nimzowich, and playing over annotated Grandmaster games to see what the GMs make of the position, the best way to improve your ability to evaluate chess positions is to play lots of serious games against strong opponents, as this will allow you to see what works and what doesn't, what kinds of position have which features and how these features interact to make the positions advantageous or disadvantageous. As with most things in chess, experience is the best teacher.



Quick links to external chess sites
Chess News       Chess General       Play Chess       Chess Software

You can visit our Amazon books page here, contact us at chyss[at]yahoo[dot]co[dot]uk, or leave a message in our guestbook here.
Unless otherwise stated all material on this page is ChyssChess.
This site contains links to other Internet sites. These links are not endorsements of any products or services in such sites, and no information in such sites has been endorsed or approved by this site.