Home Page arrow Chess Life Online arrow 2012 arrow October arrow Lessons From the 18th US Chess School: GM Alejandro Ramirez
Lessons From the 18th US Chess School: GM Alejandro Ramirez Print E-mail
By Matan Prilleltensky   
October 25, 2012
Christopher Wu, Deepak Aaron and USCS founder IM Greg Shahade, Photo Elizabeth Spiegel

The 18th session of the US Chess School opened with a bang: As IM Greg Shahade, the school's organizer, and GM Alejandro Ramirez, the instructor for the day, moved a table with laptop and projector, the table broke, with the projector almost following Greg's blueberries onto the floor. Thanks to the quick reflexes of all concerned (and Greg's liberal interpretation of the 10-second rule) all important items escaped lasting harm.
The US Chess School group hard at work, Photo Elizabeth Spiegel

The US Chess School, which first met in 2006, is a program designed to provide top-level instruction for the strongest young players in the country. The school is made possible by the generous support of Jim Roberts and the Schein-Friedman foundation: These individuals and their supporters are making a lasting contribution to the infrastructure of chess training in the USA. This edition was also indebted to the support of the Marshall Chess Club, which served as an excellent venue. Usually, a single distinguished trainer conducts the session; multiple star Grandmasters have assumed this role at least once. For this New York City session, IM Shahade decided on something else: A different Grandmaster would lead each day, offering students a variety of perspectives and styles. Greg also assumed a great deal of the coaching responsibilities himself.

Day 1

GM Alejandro Ramirez
After brief introductions, GM Alejandro Ramirez got directly to work. He opened the day with three puzzles, his preferred introduction to a day of coaching. These incorporated a mixture of endgame and geometric themes, with students working toward a different result in each study. Each one was ultimately solved, either through individual effort or a collective process. GM Ramirez also used this time to set forth his expectations for the day: If a position was concrete, students were not to reply with generalizations; precision and calculation were required. "You're not beginners, so I don't have to tell you not to call out the first move in your head. Calculate."

After the students had warmed up, GM Ramirez moved into the theme for the week: Learning from your mistakes. IM Shahade had directed each trainer to focus on tough losses and lessons learned. Alejandro started with one of the most memorable games of his career. "There was a lot of hype surrounding this game, because it was first time (my opponent) was playing in the Olympiad."

Anand,Viswanathan (2781) - Ramirez Alvarez,Alejandro (2519)

Calvia ol (Men) Mallorca (2), 16.10.2004

Alejandro started by discussing his first problem: Which opening to play against his illustrious opponent? The Sveshnikov, then his main opening, was going through a crisis ("The Sveshnikov loves to go into crisis", he mused). Playing the Dragon against Anand was, in his eyes, tantamount to a forced loss. But he had some ideas in the French Defense, and chose this bd 1 encounter between India and Costa Rica to play 1...e6 for the first time. 1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.e5 c5 5.a3 Bxc3+ 6.bxc3 Ne7 7.Qg4 0-0 8.Bd3 Nbc6 9.Qh5 Alejandro mentioned here that students should understand why they are making theoretical moves in the opening; this mate in 1 threat, for example, is prophylaxis against ...f5. 9...Ng6 10.Nf3 Qc7 11.Be3
Alejandro used this move to probe the students' understanding: What does this do, other than guard against ...cxd4 cxd4 ...Nxd4 Nxd4 ...Qc3+? By guarding f2, he explained, this makes Ng5 a threat: 11...c4 This defends indirectly against Ng5; now it can be met by ...h6. [11...-- 12.Ng5 h6 13.Nxf7 Qxf7 Now g6 hangs, and because of Be3, black cannot play ...Qxf2+.] 12.Bxg6 [12.Ng5? h6 13.Nxf7 cxd3] 12...fxg6 13.Qg4 Bd7 14.h4 Rf5
This move plans to meet h5 with taking on h5 and trading off the h1 rook immediately. 
 Interestingly, Alejandro pinpointed this as an example of a super GM mistake. "A bad move, but a very good idea." Anand is preparing Ng1-e2, and also thinking about pushing his kingside pawns. Black now has an important decision to make. Should black double rooks on the f-file, or begin with .. Qa5? Alejandro's choice in the game led to serious problems. After the game, James Black asked why white couldn't play the knight retreat in the game a move earlier. Alejandro indicated that he should do so, and referenced the following game: Bacrot,E-Apicella,M/FRA t-ch final 2004 15.Ng1!? Qa5 16.Ne2 Raf8 17.Kd2 Ne7 18.Ng3 Rxf2+ 19.Bxf2 Rxf2+ 20.Ne2 Nf5 21.Raf1 Rxf1 22.Rxf1 Qxa3 23.Qg5 Be8 24.Nf4ƒ Bacrot,E-Apicella,M/FRA t-ch final 2004/1-0 (52);
[15...Qa5! 16.Kd2 Raf8 is the key improvement. In the game, white implemented the idea of Ng1-e2, solidifying c3 while unblocking his kingside pawns. This more forcing move order prevents white from carrying out his plan so easily: 17.Ng1? fails to 17...Rxf2+! 18.Bxf2 Rxf2+ 19.Ne2 Nxd4 and black is better. Note the key difference: In the game, white was ready to meet ...Qa5 with Ne2 (rather than Kd2). To force Kd2, black had to play ...Qa4, but without his Queen pinning the c3 pawn in the game, the ...Nxd4 shot was impossible.] 
16.Ng1! Qa5 17.Ne2 
See the previous note for why this is so important. 
17...Qa4 18.Kd2 
Here Alejandro realized the situation was dire. "White will play f4, g4, h5, and crush me." He cautioned students against relying on engine evaluations in closed games: "If you give this to the computer, it will prefer black; even today, they don't understand many closed positions. . . I think black is strategically lost". He went on to note the absence of a clear plan for black, and observed that in addition to being slow, aiming for ...b5-b4 is not particularly dangerous. 
Black does this anyway, but now the exchange sacrifice is insufficient. 
19.Bxf2 Rxf2 20.h5! Rxe2+ 
20...gxh5 21.Qxh5 h6 22.Raf1 Rxf1 23.Rxf1 Nd8 24.Qh4 Nf7 (24...Nc6 25.Qf4!+-) 25.Qe7 Be8 26.Nf4+-; 20...Be8 21.Qxe6++-
21.Kxe2 Qxc2+ 22.Kf1 gxh5
23.Kg1! Qg6 
23...Ne7 24.Rf1 Qg6 (24...Nf5 25.g4!) 25.Qxh5 
24.Qxh5 Qxh5 25.Rxh5 Ne7 26.Rh3 Be8 27.Rf3 Ng6 
27...Bh5 28.Rf2 Nf5 29.Rb2 b6 30.a4+-
28.a4! b6 
28...a5 29.Rf2 Bc6 30.g3 Ne7 31.Rb2 Kf7 32.Rab1+- 
29.Rf2 a6 
29...h6 30.a5 b5 31.a6+- 
30.g3 h5?! 31.Rb2 b5 32.axb5 Bxb5 
32...axb5 33.Rf2! Bc6 34.Ra7 
33.Rh2!+- h4 34.gxh4 Ne7 35.Kf2 Nf5 36.Kf3 Kh7 37.h5 Kh6 38.Kf4 

For his second loss, Alejandro turned to the worst tournament of his chess career: The B group at Wijk Aan Zee. He introduced Ramirez-Karjakin by discussing the value of the exchange. "You have all been to camps and seen games where one side sacrifices the exchange and wins a beautiful positional game with no resistance. In real life it is not always like that. . ."

Ramirez Alvarez,Alejandro (2507) - Karjakin,Sergey (2599) [B90]

Corus-B Wijk aan Zee (10), 26.01.2005

1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.a4 e5 7.Nf3 Be7 8.Bc4 0-0 9.0-0 Be6 10.Ba2 
At this point, Alejandro commented on the position's strategic ideas: A fight for d5 will develop. For this reason, Bg5xf6 is a natural idea, removing a piece that defends that square. In general, white wants to end up with a piece rather than pawn on d5. However, there is an exception: If his a-pawn is on a5, white may want to play exd5 with a queenside bind. If white plays this way, black can look for kingside counterplay with ...f5/...e4/...Ne5, but it won't necessarily work. 10...h6 
Preventing Bg5 
11.Re1 Nc6 12.Be3 
"No idea why I played this. . . not correct". 
12...Nb4 13.Bxe6 fxe6 14.a5 Rc8 15.Ra4 Nc6 16.Bb6 Qe8 17.Qd3 Nh5 
Alejandro identified this moment as an indication something had gone wrong, since he is forced to play an unpleasant move. The retreat Be3 looks ugly, while the move played in the game undefends the rook on a4. . . 
18.Ne2 Nd4 19.Nfxd4 Qxa4 20.Nxe6 
White has sacrificed the exchange. His idea was that black will struggle to activate his rooks, and the position has some visual appeal for white: Alejandro was intending to bring his knight to d5, meet ...Nxd5 with exd5, put the other knight on the new e4 outpost, and push on the queenside with c4-c5. But Karjakin had other ideas! 
20...Rf7 21.Nc7 Nf6 22.Nc3 Qc6 23.N7d5 Nxd5 24.exd5 Qc4! 
A strong move. With the queens on the board, white's compensation is sufficient, but a queen trade makes it easier for black to activate his rooks. As Alejandro said, "An exchange is still an exchange!" 
25.Qxc4 Rxc4 26.Re2 Rf8 27.Ne4 Rfc8 28.c3 Ra4 
Karjakin consistently strives for activation. 
29.f3 Kf7 30.Kf2 Ra1 31.Rd2 Ke8 32.Ke2 Kd7 33.b4 Rc1 34.Rb2 Bg5 35.Bf2 Bd8 36.Ra2 Rb1 37.Be1 h5 38.h3 Be7 39.Bf2 g5 40.Be1 g4 41.hxg4 hxg4 42.fxg4 
How can black return the exchange to reach an advantageous ending? 
42...Rc4! 43.Nd2 Rxe1+ 44.Kxe1 Rxc3 45.Ke2 Rg3 46.Rb2 Rxg4 47.b5 axb5 48.Rxb5 Rxg2+ 49.Kd3 Rg3+ 50.Kc4 Rg4+ 51.Kb3 Kc7 52.Rb4 Rg3+ 53.Kc4 Ra3 54.Nb3 Bh4 55.Kd3 Be1 56.Rc4+ Kd7 57.Kc2 Ra2+ 58.Kb1 Rg2 59.Kc1 Rg3 60.Kc2 Re3 61.Ra4 Rc3+ 62.Kb2 Rd3 63.Re4 Bf2 64.Rb4 Kc7 65.Rb5 Be1 66.Kc1 e4 67.Kc2 Kc8 68.Kc1 e3 69.Kc2 Rc3+ 70.Kd1 Bf2 71.Ke2 Rc2+ 72.Kf3 Bg1 

Lessons: Do not underestimate the activity of a rook; once one rook reached the back rank white could not successfully defend against all the threats. 0-1

After this long fight, Alejandro turned to chess preparation and a one-sided loss. In the same unsuccessful tournament, he was paired with American GM Alex Onischuk as black. He had previously used a Benko sideline to draw easily as black against GM Moiseenko. When given the chance, he decided to repeat the line against an extremely well-prepared Grandmaster. I have incorporated some of GM Postny's notes in CBM in this game.

 Onischuk,Alexander (2652) - Ramirez Alvarez,Alejandro (2507) 
Corus-B Wijk aan Zee (13), 30.01.2005

1.d4 Nf6 2.Nf3 c5 3.d5 b5 4.Bg5 d6 5.Bxf6 
Giving the bishop, but spoiling black's pawn structure.
5...exf6 6.e4 a6 7.a4 

The idea of this move is to control the c4 square. 
7...bxa4 According to Alejandro, this is now considered slightly better. 
8.Bd3 g6 9.Nbd2 Bg7 10.Nc4 0-0 11.0-0 Bb7! 

At the time, computers did not mind this position for black, and Alejandro had some ideas he thought were interesting. This move brings up an important position. Black has only one idea: ...f5. How should white deal with it? The answer was a surprising one: He should force it. If black is permitted to complete development and improve his position, ...f5 will only gain in strength. Therefore, white should force the move, and the ensuing opening of the position, while the second player is still underdeveloped.   
12.Qd2! f5 
otherwise 13.Qf4 and d6 is weak. 
13.exf5 Bxd5 14.Be2 Bxc4?! 
Black wants to eliminate the strong knight, but now white will dominate the light squares. 
15.Bxc4 Nc6 
After 15...Bxb2? 16.Rad1 black's position is nearly collapsing. 
After 16.fxg6 hxg6 17.Qd3 Ne7 18.Bxa6 d5 black is fine.
16...Ne5 17.Nxe5 Bxe5 18.fxg6 hxg6 19.f4N 
19.Qh6 Qf6 20.Rd3 Rad8 21.b3 a5 22.Bd5 Bd4 23.Rh3 Qg7 24.g3 Be5 25.Kg2 Bd4 26.Re1 Rde8 27.Rd1 ½-½ Moiseenko,A-Ramirez Alvarez,A/Dallas 2004/CBM 105 (27) was Alejandro's original game in this line. However, his opponent in this game came prepared with a major improvement.
After this obvious check black is in trouble! The critical line is: 19...Bxb2 20.f5 Bd4+ 21.Kh1 Kg7 (21...d5 22.Qh6! dxc4 23.Rxd4! Qf6 24.Rd6! Qg7 25.Qh4) 22.fxg6 fxg6 23.Rxf8 Qxf8 24.Rf1 Bf6 25.Qd5 Ra7 26.Bxa6 Qe7 and black holds. 
20.Kh1 Kg7 21.c3 bxc3 22.bxc3 Bf6 23.g4! Qa5 24.Rf3 Qxa4 25.Bd5 Rab8 26.Rdf1

White is about to play g5 and f5, while Black has no plan. 
26...Bd8 27.f5 Qxg4 28.f6+ Bxf6 29.Rxf6 Rb1 30.Rxf7+ Rxf7 31.Rxb1 Re7 32.c4 Qf5 33.Rg1 Re5 34.Qg2 Qd3 35.Qf1 Qxf1 36.Rxf1 Re7 37.Ra1 Ra7 38.Ra5! Kf6 39.Kg2 Kg5 40.Kg3 Kf5 41.h4 g5 42.h5 g4 43.Kh4 Kf4 44.Ra1

Lessons: Just because something worked once doesn't mean it will work again! One way to catch GMs is to find an improvement on a line they used in a win or easy draw as black. 1-0

After lunch, IM Greg Shahade began his first session. Greg brings a unique perspective to the US Chess School. Having recently returned from a competitive chess hiatus, he is currently studying intensively and competing in major tournaments. Like the students, he is in a stage of his life where chess improvement is a primary goal. He began by inquiring about study habits, having the students talk briefly about how long they studied each day and what they worked on.

Luke Harmon-Velotti, Photo Elizabeth Spiegel
Luke Harmon-Velotti studies 4-5 hours a day over the summer, maybe 2 hours a day during the school year. He identified openings as his main weakness, and works on them by looking at his lines in the database and aiming to come up with improvements or new ideas. When Greg asked the students how many of them saw openings as their main problem, he was relieved to see only two raise their hands; Greg thinks many young players devote too much attention to this phase. He was surprised to hear Luke's self-assessment, and thought Luke's openings were quite good. Luke believes his tactics are stronger than his positional understanding, and enjoyed studying Volokitin and Grabinsky's "Perfect Your Chess".

Christopher Wu, Photo Elizabeth Spiegel
Christopher Wu consistently studies 2-3 hours a day. He works through books, solves puzzles, and studies openings; he was one of those who believed he was weak in this stage. According to Christopher, his repertoire only contains "a couple not bad lines" and needed to be broadened. Greg begged to differ: He pointed out that "Some top players always play the same stuff", implying that a narrow repertoire was far from the end of the world.

Deepak Aaron studies 3-4 hours a day during the summer, and 1-2 during the school year. He sees his weakness as Closed positions. This was a surprise for Greg, given that he plays the ...a6 Slav against 1.d4! Deepak's chosen study methods include game analysis and ICC.

James Black, Photo Elizabeth Spiegel
James Black earned high marks for his work ethic: He studies three hours on school days, six hours on summer days or weekends. He studies GM games, solves endgame training positions, and plays through games in New in Chess or Chessbase magazine. When Greg asked him about his weaknesses, James was unsure; this started a recurring theme throughout the week with basically every student. Greg tries to have students identify weaknesses as specifically as possible, so that they will be able to rectify them through training and study.

Christopher Gu studies 3-5 hours a day, during which he solves tactics problems and studies his openings. He identified time pressure as a primary weakness.

Arthur Shen provided probably the most open-ended answer to Greg's effort to have students quantify their study time: "It depends if I feel like it!" Some days, this means not at all; other days, it may mean quite a bit. He identified his weaknesses as misevaluating positions and difficulty playing positions which are new for him.

Alex Ostrovskiy and Arthur Shen 

Alex Ostrovskiy had the widest range in his routine: Some days he might not touch chess, but on other days he can study for eight hours. He trains with Yusupov and Dvoretsky books, which can make him feel like he "doesn't know anything". This was a repeated theme: Greg mentioned several times that Dvoretsky books, while being very instructive, can make even the strongest players scratch their heads. Dvoretsky's Analytical Manual was singled out in this regard.

Once the students had described their study habits, Greg presented a mini-lesson with a series of examples from his own games, all pertaining to the same theme. 

Shankland-Shahade, 2012 
Black to play, 12...Qd7 makes sense here. Instead Greg played 12...e4, which he identified as a serious mistake. Play continued 13.dxe4 Bxe4 14.Qa4; with Rfd1 coming next this is very bad for black. Shankland went on to win.

Zlotnikov - Shahade, Manhattan 2011

Instead of a move like ...Qc7 or ...axb4, Greg chose 12...b6, a move that came in for some of his harshest criticism of the week. After 13.Nd4 Be4 14.Nc3 white was already much better. A recurring theme throughout the USCS was Greg's self-critical approach to his mistakes - he worked to identify their sources as specifically as possible. He did this for two reasons: Ultimately correcting them, and supporting the development of this type of metacognition in the students.

De Fotis - Shahade, Continental Class 2011

Greg also showed this position (previously mentioned on CLO in his article on the Continental Class)  Black has some useful moves, notably 15...Qc7 and 15...Qe7. Instead, Greg chose 15...Ne4 after which his opponent can play Qc2, and is threatening f3 followed by Nxc5. Like Greg's previous mistake in this game, he said this move "made my opponent's life easy by making forcing moves". 

After showing students these (and other positions), Greg mentioned that all of the mistakes he had discussed shared a common theme, and invited students to guess.

Show Solution

In the future, he resolved to be very careful each time he made a forcing move, knowing that he has the tendency to make forcing moves when they are not demanded by the position.

Look for the second series in Matan's write up from the US Chess School later this month and find out more about the program at uschessschool.com.