How to Analyze
Notes on a lecture by NM Scott Massey
NM Scott Massey lectured on October 20, 2005 at the Kenilworth Chess Club on "how you analyze" during a game. He began by considering some criteria that have been developed over the years by several chess writers, which show a certain progression from focusing on the opening, to the middlegame, and then to the endgame. Then he looked at a famous game, Steinitz-Rosenthal, Vienna 1873 (see below), where we could see various modes of analysis in action. The following are simply notes on what he covered and cannot possibly repeat more than the basics for those who missed the lecture.
Wilhelm Steinitz famously wrote that "You must attack to win" and "Many advantages are temporary." When these remarks are repeated these days, people generally see him as saying that as soon as you have an advantage you must attack or risk allowing that advantage to slip away. But how did Steinitz evaluate an advantage? He offers the following principles in his writings:
These principles are really the principles we see at the beginning of the game when the fight is over the center and development. They move into the middlegame and ending to some extent, but it is interesting that one of the first writers to offer a systematic approach to analysis gives us principles that are most helpful in the opening stages.
Moving into the Middlegame
A different set of criteria, much simplified compared to Steinitz, are offered by Reuben Fine in his book The Middle Game in Chess:
These are principles that are most helpful in thinking about the middlegame as you are making decisions and planning your next move.
"What if the Queens came off?"
More recent writers have thought about the implications of decisions at every stage of the game upon the endgame. In what he calls the "Static Evaluation of a Position," Iossif Dorfman offers the following criteria, which are invaluable for any concrete evaluation of a position:
The most significant item is "Who has the better position after the exchange of queens?" Clearly, for Dorfman the ending is always in view. He prefaces his criteria by distinguishing between what he calls "static" and "dynamic" issues--"static" issues being those that are most important long-term:
"Candidate moves are chosen in accordance with the static balance. By static are implied factors that have enduring effect, whereas dynamic factors are change in a state of the position--with the energy of a break through, with the coming into contact with the opposing army, with the passage of time, their role diminishes and reduces to naught. Find a critical position (a turning point in the play, a moment when there is a possible change in the hierarchy of stategic elements)--a position in which a decision has to be taken regarding a possible exchange or a possible change in pawn formation or the end of a series of forced moves. If for one of the players the static balance is negative, he must without hesitation employ dynamic means and be ready to go in for extreme measures." --Dorfman
Let's look at a game to see how well it illustrates these principles. The following game has been annotated by several authors, including Chernev in The Most Instructive Games Ever Played and Tartakower in his 500 Games of Chess. Some of their notes have been incorporated.
Samuel Rosenthal - William Steinitz [C46]
Vienna (1) 1873
"Quite listless would be 4. Bb5?! Bg7 5.
"He allows his adversary to complete his development in peace" notes Tartakower. Thus White loses his advantage in mobility and development. In order to keep that advantage, perhaps he must attack. Better instead is:
(Illogical is 5... Bc5?! 1-0 (59) Rosenthal-Stenitz, London 1883 which leaves long-term weaknesses on the kingside.)
(The best move, though it seems counter-intuitive. But Black must hold the critical dark squares in his camp. If instead 6... Nge7? 7. Nxd4!+- Bxd4 8. Qxd4! Nxd4?? 9. Nf6+ Kf8 10. Bh6# and the dark squares have proven fatal.)
(This move may not be best, though it is what Rosenthal played in a later game and it is often given as "Book." Best may be the simple 7. Nxd4 c6 8. Nxe7 (or 8. Nc3 h6 9. Be3 Nf6 unclear, but likely better for White) 8... Nxe7 9. Qd2 h6 10. Bh4 d5 11.
O-O-Og5 12. Bg3 dxe4 13. Qe3 Qb6 (13... Qa5!? unclear ) 14. Bd6 f5!? unclear--but we see that Black must battle back aggressively so as not to fall behind.)
"White rightly prefers to maintain a piece in the center, rather than to strengthen the adverse pawn center by exchanges" notes Tartakower.
As Tartakower notes, this move has the advantage of not blocking the ...f5 break. For a time, however, 6...Nf6 became preferred. In other games with this line, Steinitz himself used 6... Nf6 e.g.:
(11. Bxa7 Rxb2 12. Bd4 Qe7!! 13. Bxb2 (13. Bd3 Qb4+ 14. c3 Qe7 15. Rb1 Qg5 16. Be3 Rxb1+ 17. Bxb1 Qxe5 18. Qxe5+ Bxe5 19. Kd2 d6 20. Be4 Be6 21. a4 Kd7= Nedev - Santos, Batumi 1999) 13... Qb4+ 14. c3 (or 14. Qd2 Qxb2 15. Rd1) and after Qxb2, Qxc3+, and Qxe5+ Black might have enough for the exchange, though this is debatable. The specific opening theory is not as important as the illustration of how the battale for development, mobility, and other dynamic issues often requires the sacrifice of material.)
blocks the QB and leaves it unprotected. The struggle begins to enter a new stage with specific threats and responses.
Better 10. Be2
in order to answer ...c5 with Be4.
threatening 16.... Bd4 17. Bd4 Rd4
note how White's B is blocked by pawns
"Driving away White's only well-developed piece" notes Tartakower.
"Black has a definite positional advantage in that he controls all the important lines" writes Tartakower.
Perhaps 23. b3.
The two Bishops go into hiding to preserve themselves for the ending. If the Queens came off, Black would have the edge.
according to Chernev, White resigns here, but other sources give the following moves:
Tartakower writes: "A full-bodied game! If one feature more than another deserves notice, it is the skillful manner in which Black turns to account the latent power of his two Bishops. The white forces are continually forced to retreat, with corresponding loss in territory."
[Chernev, Tartakower, etc.]
Game in PGN