Ftacnik - Rohde, US Open 1991, Rd 9
1 d4 Nf6 2 Nf3 e6 3 c4 Bb4+ 4 Nd2 d5 [More popular are 4 ... b6 and 4 ... 0-0. With the text Black attempts to obtain a Queen's Gambit formation in which White's knight is misplaced on d2.] 5 Qa4+ [To cross up Black's plans by forcing him to block his c-pawn with his knight. In round 5 of this event Igor Ivanov played instead 5 a3. After 5 ... Be7 6 g3 0-0 7 Bg2 Nbd7 8 0-0 c5 Black had few problems. Igor tried 9 cd ed 10 Re1!? Now, most logical would be 10 ... Re8 11 Ne5 cd 12 Nd7, etc., but I tempted him with 10 ... b6?! Black threatens to get a very good game with 11 ... Bb7, so White must take action: 11 dc! bc 12 e4 Ne4 13 Ne4 de 14 Nd2 Nf6. Now 15 Ne4 would just be a draw, so Ivanov played 15 Qc2. I confused him with 15 ... Qa5!? pinning the Nd2. He responded quickly with 16 Re3! and offered a draw. Since Black cannot stop Ne4 I accepted (not 16 ... Nd5? 17 Re4 Bf5 18 Qa4).] Nc6 6 a3 Bd2+ 7 Bd2 Ne4 8 Rd1 [Best for White here is the simple retreat 8 Qc2! with which Timman defeated Salov in the decisive game of their 1988 Candidates Match. In Browne - Rohde, 1988 American Open the continuation was 8 Be3!? 0-0 9 g3 Qe8?!.] 0-0 9 g3 [New to me. Lputian - Rohde, St. John 1988 featured 9 e3 Bd7 10 Qc2 f5! and thanks to the unfortunate position of the Rd1, Black is assured that his bishop maneuver to h5 will be meaningful.] Nd6! [Immediately taking advantage of the White bishop's apparent desertion of c4.] 10 c5 Nc4 11 Bc3 [if 11 Bc3 b6!!.] Bd7 [Not 11 ... b5? 12 Qb5 Rb8 13 Qa4 and wins.] 12 Qc2 b5 13 cb! [Trying to take the sting out of the threatened ... a5, ... b4, etc.] ab [A difficult decision. 13 ... cb renews the idea of ... b5, ... a5, but 14 e4 b5 15 Bd3 or 14 e4 de 15 Qe4 Nd6 16 Qf4 is unclear.] 14 e3 [Not necessarily admitting that 9 g3 was wrong - if Black can invest so much time in the Nc4, then White is justified in switching plans to deal with it.] Ne7 15 Bb4 c5! 16 Bc4! [Weak was 16 dc bc 17 Bc5 Rc8 18 Be7 Qe7 19 Bc4 Rc4 20 Qd2 Qf6.] dc [Thematic. 16 ... cb 17 Bd3 ba 18 Bh7+ Kh8 19 Bd3 gets Black nowhere.] 17 dc Qc8!! [17 ... Nd5 18 c6! Bc6 19 Bf8 Ba4 20 Qc1 Bd1 21 Qd1 is OK for White, and if 17 ... Qc7 18 Qc4! Qb7 19 Rd7! Qd7 20 c6 and White is winning.] 18 Qc4 bc! [18 ... Qb7 19 Rd7! loses, and 18 ... Bc6 19 e4 is unclear.] 19 Qc5 [Not 19 Bc5 Qb7 and wins.] Bc6 20 Ne5?! [20 Qe7 Re8 21 Qc5 Bf3 22 Qc8 Rec8 23 0-0 Bd1 24 Rd1 Rcd8 and Black is slightly better.] Bh1 21 Qe7 Qa6 22 Nd7? [Clearly better was 22 Qd7 but after 22 ... Bg2! Black is doing very well.] Bf3 23 Rd2 Rfc8 24 Bc3 Bg2 25 Kd1 Qf1+ 26 Kc2 Be4+ 27 Kb3 Qc4#

Aseyev - Stangl, Berlin 1992 Finally somebody copied the idea from Lputian - Rohde, St. John (above): 10 Qc2 Be8 11 Be2 f5 12 O-O Bh5 13 Be1 Kh8 14 Kh1 Rf6 15 Ne5 Bxe2 16 Qxe2 Qe8 17 f3 Nd6 18 Bg3 Na5 19 b3?! Nxb3 20 cxd5 exd5 21 Qa2 Na5 22 Qxd5 Qb5 23 Qa2 Nac4 24 Rfe1 c6 25 Rb1 Qa6 26 Nd7 Rg6 27 Bxd6 Nxd6 28 e4 fxe4 29 fxe4 b6 30 e5 Qc4 31 Qf2 Nb5 32 Rf1 h6, ... Schachwoche 1992/34 p 5

Lputian - Rohde , St. John 1988 10 Qc2 Be8 11 b4 a6 12 Bc1 f5 13 Be2 Bh5 14 O-O Rf6 15 g3 Bg4 16 Kg2 1 / 2 - 1/2

Alburt - Rohde, US Open 1991, Rd 10 1 d4 Nf6 2 c4 e6 3 g3 Bb4+ 4 Nd2 [I thought Alburt would play 4 Bd2 here, as in his famous loss to Wilder in the 1989 US Championship. Instead, Alburt has repeated the Nd2 of Ivanov and Ftacnik but with his pawn on g3 instead of with the knight on f3. This difference allows Black a more agressive continuation.] c5! 5 Nf3 [Weak is 5 a3 Bd2+ 6 Qd2 Nc6! with the idea 7 Nf3 Ne4 or 7 dc Ne4 as in D. Gurevich - A. Ivanov, Washington, D.C. 1990.*] Nc6! [On 5 ... cd 6 a3! is good - 6 ... Bd2+ 7 Qd2 Nc6 8 Nd4 Ne4 9 Nc6.] 6 dc Bc5 7 Bg2 d5 8 0-0 0-0 9 a3! [Asking Black to weaken himself with ... a5 before White plays 10 cd ed which would make other good moves available to Black. I thought for a long time here about playing 9 ... Ne4, with such variations as 10 b4 Nc3, and 10 Qc2 Nf2, etc. But I could not find an answer to 10 e3! followed by 11 Qc2, etc.] a5 10 cd ed 12 Nb3 Bb6 13 Nbd4 Re8 14 Be3 Bg4 [Black stands well as White is not exerting any pressure on d5.] 15 Nc2 Rc8 16 Qd3! [White wants to play Qb5 to force Black to exchange dark-squared bishops.] Bf3 17 Bf3! [Not 17 ef Ne5 18 Bb6 Nd3. With the text, White looks forward to 17 ... Ne5 18 Bb6 Qb6 19 Qd4.] d4!! 18 Bf4 [Backing down from the critical line 18 Bf3 de 19 Be8 ef+ 20 Kh1 Qe8 21 Rac1.] Nd5 19 Qd2 [Black stands well after 19 Bd5 Qd5 20 Ne3 Qe6.] Nf4 20 Qf4 d3 [At this point I had about 9 minutes left for 30 moves.] 21 Rad1?! [Playing for the win. White could hold steady with 21 ed Qd3 22 Rac1.] dc?! [Much better was 21 ... Ne5 22 Ne3 Be3 23 Qe3 Nf3+ 24 Qf3 d2.] 22 Rd8 Red8 23 Bg4? [Extremely unclear was 23 e3! Rd2!] Nd4! [The knight hits e2 and b3.] 24 Kg2 Rc6! [I wanted my pieces to defend each other, like a chain. I now had about 3 and a half minutes left to get to move 50.] 25 Qe3 Ne6 26 Qe5 [Or 26 Qb3 a4!] Rd1! 27 Be6 [Setting the trap 27 ... c1(Q)? 28 Bb3!] Re6! [Cleanest.] 28 Qb8+ Bd8 29 Qc8 c1(Q) 30 Qc1 Rc1 31 Rc1 Re2 32 Rc8 Rd2 33 Rb8 Rd7 34 Kf3 Kf8 35 Ke2 Ke7 36 Ra8 Bb6 37 g4 Rd8  0 - 1

Rohde - Yermolinsky, US Open 1991, Rd 12 1 d4 Nf6 2 Nf3 e6 3 c4 Bb4+ [My initial delight at facing this defense in this game for all the marbles wore off as I realized that I did not know all of the lines. For example, what if I played 4 Nd2 and he answered 4 ... 0-0?] 4 Bd2 Qe7 5 g3 Nc6 [On the adjacent board, Browne was playing Goldin in the identical position! Goldin played the simple, and weak, 5 ... Bd2+ 6 Qd2 d6. Gold tied for third place. Yermolinsky's move is more critical as on 6 Bg2 Bd2+ White must play 7 Nbd2 - not 7 Qd2 Ne4.] I crossed him up with a variation which I thought might be unexpected.] 6 Nc3 d6 7 Bg2 0-0 8 0-0 Bc3 [Not 8 ... e5? 9 Nd5.] 9 Bc3 e5 [Better was 9 ... Ne4 and White could choose between 10 Qc2, 10 Rc1 and 10 Be1 with a small advantage in each case.] 10 d5 Nb8 11 Qc2! [Eyeing the f5 square. On the immediate 11 Nh4, Black might recover with 11 ... g5! 12 Nf3 Ne4, etc.] c6 [A completely different situation arises if Black fights to get in ... f5. 11 ... Ne8 12 Nh4 f5 13 Bh3 e4! leads to massive complications.] 12 Nh4 cd 13 cd Na6 14 e4 Bd7 15 f4! [An obvious move, but it had to be accurately calculated. On 15 ... ef I planned, not 16 Rf4 Ng4, or 16 gf Ne4, but 16 Rae1! Then 16 ... fg 17 Rf6 gf 18 Qd3! wins, or 16 ... Nh5 17 Nf5 is good, or 16 ... Ng4 Nf5 transposes back to the game.] Ng4 16 Rae1 ef [Worth considering was going into a crouch with 16 ... f6.] 17 Nf5 Bf5? [Best was the surprising 17 ... Qd8] 18 ef Qc7 [Or 18 ... Qg5 19 gf! Qh4 20 Bf3 Rae8 21 Qg2.] 19 Rf4 Qc5+ 20 Kh1 Ne3 [On 20 ... Rae8 21 Re8 Re8 22 Rg4!!, e.g. 22 ... Re1+ 23 Bf1 Qd5+ 24 Kg1 Rf1+ 25 Kf1 Qh1+ 26 Ke2, or 22 ... Re1+ 23 Bf1 Rf1+ 24 Kg2 Rg1+ 25 Kh3.] 21 Qd2 [Now Black realized that his intended 21 ... Ng2 would be met by 22 Rg4!! f6 23 Rg7+! Kg7 24 Qg5+ Kf7 25 Qh5+ Kg8 26 Re7. And 21 ... Nc4 runs into 22 Rc4! Qc4 23 Re4 followed by 24 Bg7.] Rae8 22 Bd4 Nc4 23 Bc5! Nd2 24 Be3 Nb1 25 Rf3 1 - 0

GAME OF THE MONTH BOUNCING BACK In the penultimate round of the 1992 Philadelphia International, held June 25 - 29 as a type of warm-up for the World Open, I found myself playing former New York Open Champion Alexander Goldin. I had just lost an easily won game against Ibragimov, who went on to win the event. (Dzindzi took second; Browne, Fedorov, Kudrin and I tied for third.) Players have differing reactions to such tough losses. In round robin events  one may have the luxury of drawing his way back into form, but in today's big Swisses, there is generally no time for that. Many competitors switch openings - for practical and psychological reasons. Inevitably, a certain carelessness slips in. Against Goldin, I quickly obtained a dubious opening, which promised to ruin  my tournament for good. While he crashed through the center, I sought refuge in bizarre kingside counterplay, putting some distance between myself and the previous round's disaster. As Goldin endeavored to exchange off some of the attackers, the position just kept on getting more complicated, finally erupting into a wild tactical melee.
Goldin - Rohde, Rd. 7, Phila. International 1992, E11
1 d4 Nf6 2 c4 e6 3 Nf3 Bb4+ 4 Bd2 a5 [Earlier in the tournament I had seen Goldin as Black play the 4 ... Qe7  variation, and I knew he was a Bogo- player from last year's U.S. Open. But as a general matter, it is a good idea to play the same defence as your opponenet plays, as long as you play a different subvariation, because players tend to be less well prepared against alternative subvariations of their own defences.] 5 g3 d6 6 Bg2 Nbd7 7 O-O e5 [Black is ready to exchange pawns, freeing c5 as a retreat square for the  bishop.] 8 Nc3 [One of the best of White's many choices here. 8 Bg5 proved fruitless after 8 ... exd4 9 Nxd4 h6 in Alburt - Wilder, U.S. Champ. 1988, the game responsible for popularizing this variation. Interesting are 8 Be3 and 8 Bc3, but Black has no particular problems after 8 Be3 O-O 9 a3 exd4, or 8 Bc3 O-O 9 a3 Bxc3. Henley once tried 8 e3 against me, which, be defending d4, threatens 9 Bc1! After 8 ... Bxd2 9 Qxd2 O-O 10 Nc3 White had a slight advantage.] 8 ... O-O {A superficial move, ignoring White's positional threat of 9 a3. The  theoretical continuation is 8 ... exd4 9 Nxd4 O-O, which can lead to a sharp game after 10 a3 Bc5 11 Nb3 a4!?. Note that if Black tries 8 ... Qe7?, envisioning 9 a3 Bxc3 10 Bxc3 e4 11 Nd2 e3, he runs into big problems after 9 Nb5.] 9 a3 Bxc3 [Impossible is 9 ... exd4? 10 axb4 dxc3 11 Bxc3 with too much pressure on the a-pawn.] 10 Bxc3 e4 [A more conservative continuation is 10 ... Qe7, "strongpointing" on e5, but  in the long run the bishop on c3 could turn into a monster. And 10 ... Ne4? is unstable after 11 Be1 f5 12 Nh4!] 11 Nd2 Re8 [The immediate 11 ... e3? 12 fxe3 Ng4 13 Rf3 is unsound.] 12 e3 d5 DIAGRAM [I have worked hard to contain White's bishops, and was looking forward to 13 cxd5 Nb6, etc.] 13 f3! [This shot exposes the poverty of Black's opening stategy and creates the  real probability of sending him spiraling downward in the crosstable. Borrowing an idea common to the Nimzo-Indian and (with colors reversed) the French, Goldin attacks the head, not the base, of the pawn chain. He is williing to accept a backward e-pawn for f-file pressure.] 13 ... exf3 [If Black simply develops with 13 ... Nb6 14 fxe4 Nxe4 15 Nxe4 dxe4, with the idea of 16 b3 a4, then White would have a big advantage after 16 Qe2 c6 (else 17 d5) 17 a4! thanks to having preserved the mobility of his center pawns.] 14 Qxf3 c6 [Building a fortress, awaiting 15 Rae1 Nb6 16 e4 Be6! 17 e5 Nfd7 18 c5 Na4,  and while White has a big space advantage, there are no clear targets, and Black may have queenside counterplay. In this line, Black hunts down and exchanges White's c3 bishop to prevent it from being recycled over to the kingside.] 15 g4! [Giving Black no time to set up a cohesive defense involving ... Be6. Now  Black avoids 15 ... Qe7 16 g5 Ne4 17 Nxe4 dxe4 18 Qg3, as 19 Rf4 will be devastating.] 15 ... h6 16 h4 Qe7 17 g5 hxg5 18 hxg5 Nh7 19 g6! fxg6 20 cxd5 DIAGRAM [Black's pawn structure has been shattered, and White dominates the center.  Any hope for counterplay must be based on White's lack of pawn protection for his king.] 20 ... Ndf6! [At least regaining control of the d5 square.] 21 dxc6 bxc6 22 Rae1 Bd7 [Only an illusion of activity is provided by 22 ... Ba6 23 Rf2 Ng5 24 Qxc6  Rec8. After 25 Qb6 Nfe4 26 Rf4 White is winning.] 23 Qg3  [Trying to disorganize Black. On 23 Nc4 Ng5 and a knight lands on e4, or 23  e4 Ng5 24 Qg3 Nh5 with weird counterplay.] 23 ... g5 24 e4 [An oversight, allowing a Black knight to hop in to f4. But the position was  agonizingly unclear after 24 Nc4 g4 25 Ne5 Ng5, or 25 Nd6 Rf8 26 Nf5 Qf7.] 24 ... Nh5 25 Qh2 Nf4 26 Nf3 Rf8 27 Ne5 Rf6 [A bizarre position. While White still dominates the center, Black's kingside threats take on real proportions.] 28 Qg3 Rh6 29 Rf2  DIAGRAM 29 ... Bh3 [Tempting was 29 ... Nh3+ 30 Bxh3 Rxh3 31 Qg2 g4, with threats of 32 ... Qh4  and 32 ... Ng5. But White has 32 Nxd7!!, so that if 32 ... Qh4 33 Kf1, or 32 ... Qxd7 33 Rf5! and the attack fizzles. I played the text intending 30 Bh1 Bc8!! 31 Bg2 Nh3+, etc., preserving the light-squared bishop. But Goldin finds a different way to liquidate.] 30 Ng6! Rxg6 31 Bxh3 Nxh3+ 32 Qxh3 g4 [Now the question is whether Black can get his knight to g5 meaningfully. If  White can stop that, or get to an ending, he will have good prospects.] 33 Qh5! [The queen makes herself a pain in the neck. If now 33 ... Rg5 34 Qh4, the  annoying pin enables White to control g5 with moves like Bd2 or Rf5.] 33 ... Rh6! [Jettisoning the g-pawn to ensure the knight's arrival on g5.] 34 Qxg4 Rh4 35 Qg2 DIAGRAM [White has stolen a pawn and is running for cover. Black, who can combine  threats against White's king with threats against the e-pawn, risks not recovering the pawn, or recovering it but losing the initiative. The tactics starting from this position are truly unbeleivable. If 35 ... Nf6, threatening 36 ... Rg4 and 36 ... Rxe4, White has 36 e5!! Then 36 ... Rg4 37 exf6 Rxg2+ comes out great after the obvious 38 Kxg2 Qf7. But White has the vicious 38 Rxg2! Qxf6 39 d5 and wins!] 35 ... Ng5 36 Re3 Qd7!! [Strong and necessary! White's rooklift is decisive on 36 ... Rxe4? 37 Rg3.  And on the expected 36 ... Nxe4? White does not play the routine 37 Rfe2? Re8 after which Black is better, but instead White "hits the cutoff man" (i.e. the e5 square) with 37 Rf5!! containing the paralyzing twin threats of 38 Re5 and 38 Be1. But with the text Black threatens 37 ... Rg4, and White cannot play 37 Rg3 Nxe4, or 37 Kf1 Rg4 38 Qh1 Qe6.] 37 Rf5! Rxe4 [A neat geometrical pattern. 37 ... Nxe4 would lose to 38 Rxe4 as the rook on h4 would hang, but now both White rooks are hanging, and 38 Qxg5 Rxe3 does not help, while 38 Rxe4 Qxf5 leaves the knight finally ready to move in. 38 Rxg5 Rxe3 39 d5 [White has lost the Exchange, but is attacking g7 and c6. 39 ... Rxe3 40 bxc3 cxd5 would be a draw.] 39 ... Re7!  DIAGRAM 40 dxc6!! [Goldin allows me to pin his queen, but if 40 d6? Rf7 my only target on g7 is well defended and my material advantage would be telling.] 40 ... Qd1+ 41 Kh2 Re2 42 Rxg7+ Kf8 43 Rg8+ Kf7 44 Rg7+! [On 44 Rxa8? White loses instructively: after Black wins the queen he can  only win if he can pick up the rook on a8. To do that he must take the pawn on c6 with check. The line is 44 Rxa8? Qd6+ 45 Kh1 Qh6+! (not 45 ... Rxg2? 46 Ra7+) 46 Kg1 Rxg2+ 47 Kxg2 Qxc6+. But White's real idea is to give perpetual check - if the Black king crosses the d-file White will have Rd7+.] 44 ... Ke6 45 Rg6+ Ke7 [Of course not 45 ... Kf5?? 46 Rf6 mate!] 46 Rg7+ Kd8 47 Rd7+! [White could still lose with 47 Bf6+? Kc8! (47 ... Ke8?? 48 Re7+! Rxe7 49 Qg8 mate) as Black has crossed the d-file. Then 48 Rg8+ Kc7 49 Be5+ Kb6 50 Bd4+ Qxd4 51 Qxe2 Rxg8, or 49 Rg7+ Kb6 50 Bd4+ Ka6 ends matters.] 47 ... Qxd7 48 cxd7 [White may be losing after 48 Qxe2? Qxc6.] 48 ... Rxg2+ 49 Kxg2 a4 50 Kf3 Kxd7 [The position is a theoretical dead draw.] 51 Ke4 Kc6 52 Kd3 Kb5 53 Bg7 Kc5 54 Bc3 Rg8 55 Kc2 Kc4 56 Bf6 Ra2+ 57 Kc1 Kb3 58 Bc3 1/2 - 1/2