# How many steps can a grandmaster foresee? [duplicate]

What makes a grandmaster special? Is it the amount of steps the grandmaster can foresee, or the creativity? How many steps can a grandmaster foresee in average in a game (for instance Carlsen and Ivanchuk)? Is there a minimum amount?

• "I see only one move ahead, but always the best move." chesshistory.com/winter/extra/movesahead.html – Noam D. Elkies Mar 27 '17 at 19:27
• For the most extreme variant, I recall the newspaper analysis of Kasparov vs. X3D Fritz game 3 (see here for game but not analysis en.wikipedia.org/wiki/X3D_Fritz ). X3DFritz had achieved 20 move depth and could not see the fall in front him, like shining the world's most powerful flashlight down a black hole. If human depth search likes computer search this puts Kasparov's depth farther than that. – Joshua Mar 28 '17 at 4:01
• @Joshua: At which point of the game do you claim that Kasparov calculated that far? This looks like typical anti-computer chess, where you keep the position closed so that tactics are not important and evaluating the position becomes more important. I doubt Kasparov had to calculate anywhere near to 20 moves deep in this game. – user1583209 Mar 28 '17 at 8:02

This depends a lot on the position and cannot be answered generally. If you have a forced line, i.e. a position where there are only few candidate moves at every step, a grandmaster can calculate many moves. I am reluctant to give a concrete number because it really depends, but to give you a rough idea, let's say 20 moves.

On the other hand if the position is rather closed, of the maneuvering type, with many equal alternatives at every step, the calculation depth will be much smaller, simply because the number of variations will be much much larger in this case. And even if you calculated 20 moves deep you might not come to a conclusion on which line is best, because 20 moves in the future the position might still be equal. In this case a grandmaster would typically only calculate a couple of moves deep (if at all) in order to check that he is not blundering anything to a tactic.

As you can see from this description, pure calculation strength is only needed sometimes in a game and certainly not the only reason that grandmasters are much better than the rest. Unlike computers, humans do not play chess by calculating many moves in advance. The strength of human players over computers is that they can evaluate the position and develop a plan based on the evaluation. Humans focus more on recognizing patterns and play moves based on these patterns.

So, grandmasters are grandmasters, because they are good at all parts of the game:

• evaluating a position correctly
• finding the right plan
• good at defense: using tactical ideas, standard methods, pure calculation...)
• good at openings: knowing main lines, side lines, recognizing wrong moves in an opening and knowing how to react to them, knowing the ideas behind openings...
• good at endgames: as with openings, there is a huge body of theory that you can study; also some grandmasters like Carlsen, are said to be particularly good at endgames beyond the known theory
• pattern recognition abilities: to recognize tactics and other known motives
• creativity
• personal traits, physical fitness
• And you bet that with presential human-to-human games, reading body language is also a big factor. – Mindwin Mar 28 '17 at 12:38

It depends upon the position.

For a simple endgame, grandmasters use shortcuts to see very far. For example, using the rule of the square, a grandmaster can see 7 moves. Combining other shortcuts like this, they can see 15 or more moves ahead.

In middlegames with few tactics, GM Soltis argued in Studying Chess Made Easy that seeing 2 1/2 moves ahead was sufficient for grandmasters to find a good move.

It entirely depends on the position and number of variations. For example, if there is a forced line (where both sides only have one option on each move in a line), then grandmasters can calculate 20 or 30 moves ahead without much trouble.

But if each side has multiple choices on each move, grandmasters can not calculate nearly as much, since the number of possible positions grows exponentially. Think of the number of variations as a tree - if both sides have two choices on each move, the tree grows two new branches at each branch. You can see how this would become very large quickly.

In general though, most positions do not require extensive calculation - on most of their moves, Grandmasters only calculate 4-5 moves ahead... mainly to just make sure their idea works. They would only need to calculate many moves when they're in an intense tactical battle, which usually only happens once or twice in a game (since the winner of the tactical skirmish usually wins the game shortly after).

What makes a grandmaster special is a number of things. The ability to calculate many moves is definitely one trait, but there are many other traits as well:

• Evaluating a position to find the correct plan.
• Pattern recognition abilities (the quickly recognize tactics or well known strategic plans during a game).
• Raw talent - many people work very hard at the game, but only 0.1% of those people actually make it to grandmaster level.
• Competitive spirit and the will to win - this one is definitely important in competitive tournaments. The burning desire to win is important to succeed in chess, even though it may be somewhat unhealthy to have in real life :)

Grandmasters follow many Chess principles which sometimes reduce the amount of Calculations in the game.

1. When the Position is Tactical and lot of forcing moves are compelled they can calculate up to 15-20 moves in different variations.
2. Please note that they do not pick up all the combinations as an amateur tries to do but they calculate the important lines. Now you may ask how they choose the important lines. After the calculation they have already visualized the basic aspects like their own King safety, piece placement, etc.
3. There is a Principle of Elimination method which helps them to reduce the unimportant lines and selection of candidate moves.
4. They choose their openings which they can memorize for 20 moves on average, and they know the clear ideas in the middlegame. For example, if a GM plays the Sicilian Najdorf as Black, he knows that Black's idea is to play on the c-file, attack on the Q-side and play in the Centre. They know the aspects of the middle game when they choose an opening. They even study 100s of games on that particular line.
5. One characteristic is that they believe if their pieces are on good squares and positionally sound, it helps to reduce calculations. As Carlsen said during his WCC match against Anand, "My pieces are on better squares compared to his, so why I should be afraid of losing?" This, I believe, is a great lesson for all.
• The old saw is "one move", i.e. one more than his/her opponent. – CConero Mar 29 '17 at 15:16