Thursday, August 21, 2014

Grandmaster Thinking and The Rest of Us

In 1946 Dutch Master and psychologist Adriaan de Groot published Though and Choice in Chess. It was based on studies he conducted from 1938 to 1943 and was designed to see the way chess players think. GM's used in the experiment were Keres, Alekhine, Flohr, Fine, Euwe and Tartakower. The IM’s were Prins, Cortlever, Landau and van Scheltinga and two Dutch women champions, Heemskerk and Roodzant. In addition five players in the ‘Expert’ (2000-2199) category and five players ranging from 1600-1900 were also used. The players were asked to speak their thoughts out loud during their analysis of a number of positions. Here’s the thought processes of Euwe, Keres and one of the class players on the following position (White to move). I have presented the players’ thoughts in a somewhat condensed form. I will leave it up to the readers to draw any conclusions but it’s clear how quickly GM’s zero in on the essence of the position while the rest of us, like the class player, sometimes have really muddled thinking.


Dr. Max Euwe took 15 minutes
     First impression: isolated Pawn. White has more freedom. Black threatens …Qxb2. Is it worthwhie to parry that? It probably is; if he takes the Qa3 is also attacked. Can White then take advantage of the open file? Does not look like it. Still again, 2.Nxc6 and then by exchange the P at a3 is defended by the Q.
     Indirectly in connection with the hanging position of the N on c6 and possibly because of the overburdening of the B on e7. But wait a moment, no, …Qxb2 is rather unpleasant after all because the B at a2 is undefended. Can I do something myself? Investigate that first.
     The pieces on f6 and d5 are both somewhat tied down. Lets look at the consequences of some specific moves.
     1.Nxd5, possibly proceeded by 1.Nxc6 then 1…Rxc6 is probably impossible because of White taking on d5. Black has a number of forced moves; it may be possible to take advantage of that. It’s not yet quite clear. Let’s look at other attacks.
     1.Bh6 in connection with f7, but I don’t really see how to get at it. 1.b4 in order to parry the threat. But then exchange on c3 will give some difficulties in connection with …Bb5; Oh, no. That is not correct; one can take back with the Q. So far a somewhat disorderly preliminary investigation. Now let’s look in some more detail at the possibilities for exchange: 1.Nc6 or 1.Nd5 or maybe 1.Bxd5 or maybe first 1.Bxf6.
     Euwe then analyzed 1.Nxc6 and came to the conclusion there was no immediate advantage. He then analyzed 1.Nxd5 and came to the conclusion that White gets a good position, but there was no way to make anything out of it. He then moved on to 1.Bxd5 and his thoughts continued…
     1.Bxd5 this must be looked in to. Does it make any difference? 1.Bxd5 Bxd5 is again impossible because of 2.Nd7. That is to say we will have to look out for 2…Bb5, but that we can probably cope with: the worst that can happen to me is that he regains the exchange, but then I have in any case some gain of time.
     1.Bxd5 Nxd5 Same difference as just before. No, that is now impossible: 2.Nxd5 wins a piece. 1.Bxd5 Bxd5 2.Bxf6 Bxf6 3.Nd7 Qd8. Let’s have a closer look at that: 4.Nxd5 exd5 and I’m an exchange o the good. But that’s good for White. The N on f6 is weak, the B on e6 hangs and the B on c6 stands badly. On positional grounds one could already decide on 1.Bxd5. Is there some immediate gain? 1.Bxd5 exd5; it looks bad for Black. 
     Probably some more accidents will soon happen. Much is still up in the air. One plays 2.Qb3. Defending the N on f6 is not do easy; 2…Kg7 looks very unpleasant. Yes, I play 1.Bxd5.

     Paul Keres took 6 minutes looked at 1.Bh6, 1.Bxc6, 1.Nxc6 and quickly concluded ‘White wins after 1.Bxd5. Notice how the GM’s homed in on the correct move 1.Bxd5 and how concise there analysis was.  Looking three moves deep seems to be the norm. Compare that to a typical class player’s analysis. 
     A Class player who took 28 minutes summarized the position as being a maze of pieces with White attacking and having a concentration of Q, B’s, N on e5 on the Black K’s position and he didn’t think there was anything for Black to fear because of his slightly weakened K’s position due to the P on g6.
     He then stated he had to begin thinking of a combination even though he did not think there were any immediate winning possibilities. He began by looking at ‘combinational stuff.’
     He first looked at 1.Bh6 and 1. Qh3 quickly concluding that 1.Bh6 was not satisfactory because Black could easily defend and 1.Qh3 allowed Black to capture on d4.
     He then thought he should try to open up Black’s K more and so first had to play 1.Nxd4 Nxd4 then he would have to trade the with 2.Bxe7 but that allowed too many exchanges.
     Then he went back to 1.Bh6. Next he looked at 1.Ne4 but quickly concluded it also allowed too many exchanges which was not good for him. 
     He then looked at 1.Nxd5 and almost immediately switched back to 1.Qh3 reasoning ‘to start kind of a pin on the P on e6 – after the P takes on d5, the diagonal was open, then he suddenly remembered the d4 Pawn was hanging and was somewhat disgusted with the situation. At that point the examiner asked him what he was thinking about and he replied the d-Pawn again then finally observed the b-Pawn was hanging but didn’t think it was important, but on second thought he could not allow it to be captured.
     He then voiced the opinion that everything he had looked at resulted in simplification and his attack was gone so perhaps he should look at positional move, but if he did that then Black himself could simplify.
     He suggested 1.b4 (safe), 1.h4 (attacks, but too slow), 1.Bb1 (but with the Q in front of the P no sacrifice on g6 was possible), 1.Ne2 (and then Ng3 followed by h4-5 but then Black captures on b2 and White needs to keep the attack going) so it was back to 1.Bh6 and 2.b4.
     He then considered 1.a4 to prevent …Bb4. Then he examined the possibility of getting rid of Black’s B on c6 with 1.Nxc6. If 1…Qxc6 then 2.b4 is unnecessary and on 2…Rxc6 or bxc6 then he has lost the Bishop pair. After that White could get his N on c3 into the game.
     He then mentioned again that simplification was not a good idea and concluded 1.Nxd4 and then 2.Re1 ‘in case holes turn up on the K-fie.’ He concluded the best move was 1.Nxd4.
     I let Stockfish, Houdini 2, Critter, Deep Rybka 4, Gull 3, Fritz 12 and Komodo 5 examine this position and they all immediately selected 1.Bxd5 as the best move and that White had an approximate 1.0 to 1.25 P advantage. Fritz and Naum 4.2 thought Black should play 1…Bxd5 while the others selected 1…exd5. 
     After 1.Bxd5 exd5 then 2.Qf3 was selected by all the engines except Houdini 2 which recommended 2.Rfe1.
     I was curious about how the class player’s move of 1.Nxd4 turned out. The evaluation dropped to -0.25 after 1…Nxd5. Now White can’t play the intended 2.Re1; he has to move the B. Simplification starting with 2…Bxe7 does mean his advantage has dissipated and Black is better so White needs to play 2.Bh6 but after 2…Rfd8 3.Rfe1 (3.Qd2 is roughly equal) 3...Qxb2 4.Qf3 Nf6 5.d5 exd5 the advantage has swung to Black!

No comments:

Post a Comment