How many moves a GM can think?


Ok, plz don't kill me for the question, but take it as simple amateur question. I am very much curious to know how many moves a grandmaster can think. It will help me understanding GM games. For example, i was looking at 2 GM games...I found a move which is good but i just thought 4-5 moves head ,but GM played some different move and i later realized, i think after 10th move that why he played his move and not mine. So i am finding difficulty in analyzing games of top players simply because there current moves are process of deep thinking. 

So i am just curious to know how many moves ahead GM or top players can think


Ya i know....Positional skill is most important. As i said before i am just curious to know. Will Anand answer my question or any other GM ? Yes!!! They will simply lie at least till retirement. Carlsen once told 13-15 moves, but i am sure it is a lie, isn't it ?


There are some positions in which very deep calculation is needed, and GMs are able to see that deeply.

But it's more important to see broadly (see only three moves deep, but don't miss any reasonable reply by the opponent) and to evaluate the position at the end of the line correctly.


This is a very common question -- here is a good article on it, it's about a page long but has good information:

To put it briefly, the most common misconception among amateurs is that GMs win because they must calculate a lot.  The key to playing good chess is their long term memory and positional understanding (correct evaluations at the end of any calculation).

Remember 1.e4 e5 2.Nf3 Nc6 is only two moves, what an amateur or non chess player calls "one more" is actually a half move :) so when carlsen says 15-20 it may really make your head spin... and no, he's not lying.  When there are a lot of forcing moves, or especially in pawn endgames it's not too hard (at least for GMs) to calculate 20 full moves ahead.

I'm a class player (not even an expert) and in long tactical puzzles I've solved correctly 10-12-14 moves ahead (but remember these are more or less forcing combinations which makes it easier) and I'm light years away from grandmasters. 

Again the secret though has nothing to do with how many moves you can calculate (visualize an opening where each side moves a pawn one square forward down the line each in turn you've just perfectly visualized 8 moves ahead) the secret is accurate evaluation of the position you've reached.

tonydal wrote:

First off, you have to clarify something.  "Moves" equals full moves (not half-moves), one move by White and one by Black=one move.

Then you have to get something else straight:  true calculation involves seeing all the best moves for both players, not just some plausible continuation for yourself.  I bring this up because it's been a bone of contention on these forums before.  If you're looking at (or diligently trying to anyway) all the best moves for both players, that's real analysis; anything else (or less) is more akin to daydreaming than calculation.

Having said that, I would say that GMs can routinely see 10 or more (full) moves ahead.  Whether they do so all the time (or even a lot of the time) is another matter...the "record" (as far as I know) is approx 20 (in Reti-Alekhine, Baden Baden 1925); although his notes to the game actually cast doubt on whether he foresaw everything from the beginning of that combo.

Which leads to another point:  the role of intuition in the game.  Kasparov always highlighted it (in his own play at least).  It's hard to believe that anyone could've foreseen every twist and turn in his masterpiece against Topalov, for example, but win that one he did (as he did with so many others of course).

Yes i mean Full moves..both for white and black...I understand that for good analysis opponents good and best reply need to be considered. 

Thanks for your replies :)


From my own experience I can say that grandmasters do not do an inordinate amount of calculating. Tests support me in this claim. If anything, grandmasters often consider fewer alternatives; they tend not to look at as many possible moves as weaker players do. And so, perversely, chess skill often seems to reflect the ability to avoid calculations . . ."
        -  Artur Yusupov


A quote I like to give out when I can Tongue out

tonydal wrote:

Yet Watson says they crunch like howitzers...maybe this is all a reflection of style.

I've heard some GMs emphasize the importance of intuition and some calculate like mad.  I enjoy and focus on Yusupov's quote because it appeals to me personally, however that GMs have different styles is a very good point.

ankitthemaster wrote:

Ya i know....Positional skill is most important . . . "

 But do you really?  Because its the very heart of what makes a GM. 

They don't just know positional theory - they know the positions.  This how they win 40/40 in simuls, they don't stand in front of you calculating for 5 minutes, they look at the board for a second and a position from memory pops to mind.  The GM then sacs his bishop on h7 because he already knows that he has enough material and tempo to win regardless of what you do.

This is why GM's often resign games that we amatures think are still playable.  Its a position that they both already know (and - they both know the other guys knows!). 

Dr. De Groot showed that a typical Master probably has a positional memory of about 10k positions, while a GM has a memory of around 100k.  That alone is enough to trim a lot of dead branches off of the 'candidate move tree'. 


I think most GMs can calculate when needed, in forcing lines, as far as play can be forced.  Of course they calculate, but they can weed out the inferior choices more quickly and devote their time to the best candidates.  Tactics are more obvious to them, so less time is spent divining them and more on the key features of the position.


Capablanca once lost a game to a patzer. Some reporter asked him how many moves he thought ahead, and Capa replied "10" - then same journalist asked the patzer how many HE thought ahead. The patzer replied "One". The amazed journalist then asked him how he could possibly have beaten the world champion when he only thought one move ahead. The patzer replied "Only one move....but it is always the BEST move " Laughing

smileative wrote:

Capablanca once lost a game to a patzer. Some reporter asked him how many moves he thought ahead, and Capa replied "10" - then same journalist asked the patzer how many HE thought ahead. The patzer replied "One". The amazed journalist then asked him how he could possibly have beaten the world champion when he only thought one move ahead. The patzer replied "Only one move....but it is always the BEST move " a rather silly statement. (Besides, I thought that quote was by Capablanca himself?)


Capablanca used to tell this story:

"I was playing in a tournament in Germany one year when a man approached me.  Thinking he just wanted an autograph, I reached for my pen, when the man made a startling announcement.  'I've solved chess!'  I sensibly started to back away, in case the man was dangerous as well as insane, but the man continued: 'I'll bet you 50 marks that if you come back to my hotel room I can prove it to you.'  Well, 50 marks was 50 marks, so I humored the follow and accompanied him to his room..."

"Back at the room, we sat down at his chess board.  'I've worked it all out, white mates in 12 no matter what.'  I played black perhaps a bit incautiously, but I found to my horror that white's pieces coordinated very strangely, and that I was going to be mated on the 12th move!"

"I tired gain, and I played a completely different opening that couldn't possibly result in such a position, but after a series of very queer-looking moves, once again, I found my king surrounded, with mate to fall on the 12th move.  I asked the man to wait while I ran downstairs and fetched Emmanuel Lasker, who was world champion before me.  He was extremely skeptical, but agreed to at least come and play.  Along the way we snagged Alekhine, who was then world champion, and the three of us ran back up to the room..."

"Lasker took no chances, but played as cautiously as could be, yet after a bizarre, pointless-looking series of maneuvers, found himself hemmed in a mating net from which there was no escape.  Alekhine also tried his hand, too, but also to no avail."

"It was awful!  Here we were, the finest players in the world, men who had devoted our very lives to the game, and it was all over!  The tournaments, the matches, everything - chess had been solved, white win in a dozen or so moves!"

About this time Capa's friends would break in, saying "Wait a minute, I never heard anything about all this!  What happened?"

"Why... we killed him, of course..."


Capa was a great story teller -- and by the way you mixed it up, Capa was the one who only looked 1 move, the best move, ahead Wink


orangehonda, great story, I've read it before so I know it's true Laughing

tyzebug, get urself a sense of humour - I know that in short supply in Singapore so try the local MacDonald's Smile


Only one move, but usually the best move


Chess takes lots of thinking. Grandmasters must think upto 25 depth to see long tactical moves.

GetClub Chess thinks 10-12 depth in one sec.

I play: Here Baby level takes 2 sec to think arround 15-16 depth.

While for some positions it can think more than 20 depth long. All depends on position of game.


smileative wrote:

tyzebug, get urself a sense of humour - I know that in short supply in Singapore so try the local MacDonald's

I tend to look for it in the Life section of the supermarkets, myself. :P (You can't deny that 1. it's a silly statement and 2. it was from Capablanca, not the patzer; though.)


When Reti was asked how far ahead he calculates, he replied, "One move."

mkirk wrote:

Only one move, but usually the best move

For a player to find the best move he first needs to think ahead and see why his move is best, otherwise he is just recalling familiar positions.


Well, he doesn't allways need to see the following positions, sometimes a outpost for a knight might be a goal, even though the player doesn't necessarily see what he can do with it later on.


Would they actually –and really – calculate 20 moves ahead, they surely would then play 20 moves very fast. This is not the case. They usually play one move at the time, only openings (no calculations needed) or various exchanges (maybe 3 moves) they play fast.   

So, usually, they calculate one move ahead.

ps. I don’t know, but this is logical?