Thursday, February 15, 2007

PCG Letter from Melchor

In April 1998, I received a letter by air mail from Alejandro Melchor of Barcelona, who describes himself as "an expert chess player from Spain". Besides the letter, there were several typewritten pages containing the moves from dozens of games in the Philidor Counter Gambit, most of them by transposition from the Latvian Gambit. Incidentally a dozen games in which Melchor himself participated can be found in New Developments in the Latvian Gambit by Kon Grivainis and John Elburg.

Melchor writes in his letter that he recently purchased my book The Dynamic Philidor Counter-Gambit and was surprised to find so many points of contact with Latvian Gambit variations, mentioning that Chapter Five in my book (1.e4 e5 2.Nf3 d6 3.d4 f5 4.exf5 e4 5.Ng5) is in fact a Latvian Gambit. From a data base of more than 6,000 games on the Latvian Gambit, Melchor has sent me only those games that transpose into lines in the Philidor Counter Gambit. Some of these I have never seen before and are of theoretical significance.

The first game of importance includes a subvariation that overturns analysis by Staunton of a line first suggested by Lowenthal in his notes to the famous consultation game Staunton & Owen vs. Morphy & Barnes, London 1858. That game began 1.e4 e5 2.Nf3 d6 3.d4 f5 4.dxe5 fxe4 5.Ng5 d5 6.e6 Nh6. Now instead of the game continuation 7.Nc3, Lowenthal recommended 7.f3. But Staunton himself felt that after 7.f3 e3 8.Bxe3 Be7 the position would slightly favor Black despite being a pawn down. Until Melchor wrote to me, I was unaware that Lowenthal's 7.f3 had actually been played over the board in N.N.-Sedley, USA 1871. After 7.f3, Sedley avoided Staunton's variation in favor of 7...Bc5!?.

Melchor cites the following line of analysis by Polugayevsky: 7.f3 e3 8.Bxe3 Be7 9.Qd2 O-O 10.Bd3 with clear advantage to White, as opposed to 10.Nxh7 Kxh7 11.Bd3+ Nf5 12.g4 Bh4+ 13.Kd1 Bxe6 which would be equal according to analysis by Suhle. After Sedley's seventh move, the game proceeded 8.fxe4 O-O 9.Qxd5 (or 9.exd5 Bf2+ 10.Ke2 b6! with compensation) Bf2+ 10.Kd1 Qe7 11.Bd3 Ng4 12.Nxh7!? Bxe6 13.Qh5 Rd8 14.Bg5 Rxd3+ 15.cxd3 Qd7 16.Nd2 Nc6 17.Rf1 Nge5 18.h3 Nxd3 19.Rb1 Nce5 20.Bf6 Ng4 21.Bg5 Nde5 22.Rxf2 Nxf2+ 23.Ke2 Nxe4 24.Nxe4 Qd3+ 25.Kf2 Qxe4 26.Bd2 Nd3+ 27.Kg1 Qxh7, and here White resigned, a piece down. This game illustrates the theme of rapid development by Black.

In a similar fashion, the correspondence game Henderson-Jackson played in 1868 interpolated the moves 7.Nc3 c6 before continuing with 8.f3 Bc5!? 9.fxe4 O-O 10.exd5 Bf2+ 11.Ke2 b6! 12.Qd3 Ba6 13.Nb5 g6 14.dxc6 Nxc6 15.Qxd8 Raxd8 16.Nf3 Ng4 17.c4 Rfe8.

Here Melchor evaluates the position as clearly favoring Black and gives no further moves.

The next game was of particular interest to me because, to the best of my knowledge, it is the first time that a suggestion of mine 9...Nf6! and 12...Ne4! was put into practice. It happened in Geenen-Henris, Belgium Championship 1995. Here are the moves: 1.e4 e5 2.Nf3 d6 3.d4 f5 4.dxe5 fxe4 5.Ng5 d5 6.e6 Bc5 7.Nxe4 Be7 8.Qg4 g6 9.Ng5 Nf6! 10.Qa4+ c6 11.Nf7 Qb6 12.Nxh8 Ne4!

13.f3 Bh4+ 14.Kd1 Qf2 15.Be2 Qxg2 16.Rf1 Nf2+!? 17.Kd2 Bf6?!, draw. At first glance, I found it incomprehensible that Black could have followed my analysis through move 15 (with a slight transposition at moves 13 and 14) and then have missed 16...Bxe6! which not only strands the White knight on h8 but prepares to answer 17.fxe4 with 17...Bg4, after which Black has sufficient compensation for the sacrificed material. The only explanation I can think of is that the player of the black pieces must have seen my analysis in Macon Shibut's 1993 book Paul Morphy and the Evolution of Chess Theory which stops after 12...Ne4! with the evaluation of "unclear complications". Even more remarkable is the fact that White agreed to a draw in a position where Black is down a rook with little to show for it after 18.Nf7 Bxe6 19.Nd6+ Kd7 20.Nxb7. At least, 17...Bg5+!? 18.f4 Bf6 would have made more sense than the immediate 17...Bf6. For those of you who are interested, the ratings list in Informant 66 shows Geenan and Henris with FIDE ratings of 2330 and 2280 respectively.

Instead of 8.Qg4, the move 8.Qh5+ was played in Locmaria-Gallighan, Paris 1872. The game continued 8...g6 9.Qe5 Nf6 10.Ng5 O-O 11.Nf7 Nc6.

Now two games of mine followed with 12.Qe2 Qe8, but Black won after 13.c3 Nd8 14.Bh6 Nxe6 15.Bxf8 Bxf8 16.Ne5 Nf4 17.Qe3 Bh6 18.Kd1 Nh3! (0-1, 27, Sunil Weeramantry (USCF 2345) - JimWest (USCF 2232), Somerset NJ 1995) and 13.Bh6 Nd4 14.Qd2 Bb4! 15.c3 Qxe6+ 16.Kd1 Rxf7 17.Qxd4 Ng4 (0-1, 22, Asuka Nakamura (USCF 1984) - JimWest (USCF 2224), Somerset NJ 1996). Here Locmaria tried 12.Qg3 but met the same fate after 12...Qe8 13.Bh6 Nh5 14.Qe3 Ng7 15.Ng5 Bxg5 16.Bxg5 Bxe6 17.Bb5 Nf5 18.Qd2 Bd7+ 19.Be2 Nfd4 20.Be3 Nxe2 21.Qxe2 Ne5 22.Nc3 Bg4 23.Qd2 Rd8 24.O-O d4 25.Bxd4 c5 26.Qg5 Rxd4 27.f4 Rf5 28.Qh4 Rdxf4 29.Rxf4 Rxf4 30.Nd5 Rf5 31.Ne7+ Kg7 32.Nxf5+ Bxf5, and 0-1 in 45 moves.

A Latvian Gambit that transposed into a sharp line in the Philidor Counter Gambit was seen in the correspondence game Malmstrom-Downey, 1994. It began 1.e4 e5 2.Nf3 f5 3.Nc3 d6 4.d4, reaching a position usually arrived at by 2...d6 3.d4 f5 4.Nc3. Here Morphy played 4...fxe4 5.Nxe4 d5 against Bird (London, 1858) and won brilliantly after 6.Ng3. In more recent times, Arhipkin has championed 4...exd4. Instead Downey played 4...Nf6, a move frowned on by theory. But after 5.dxe5 Nxe4 6.Nxe4 fxe4 7.Ng5, Downey avoided 7...d5 8.e6 Bc5 9.Nxe4 Be7 10.Qh5+ g6 11.Qe5 with clear advantage to White and came up with an improvement in 7...Nc6!?.

Malmstrom declined the offer of a pawn following 8.exd6 Bxd6 9.Nxe4, perhaps because 9...Qe7 10.Qe2 O-O gives Black some play for it. But it is not clear what Downey would have played on 9.Bc4, although the exchange sacrifice by 9...Bf5 10.Nf7 Qd7 11.Nxh8 Ne5 12.Bb3 O-O-O (followed by an eventual ...Rxh8) might be worth considering. In any event, the game proceeded 8.e6 Qf6 9.Bc4 Be7 10.Qh5+ g6 11.Qe2 Nd4 12.Qxe4 Qe5 13.Qxe5 dxe5 14.O-O Nxc2 15.Nf7 Nxa1 16.Nxh8 Nc2 17.Nf7 Nd4 18.Nxe5 Bxe6 19.Bxe6 Nxe6 20.Rd1 Rd8 21.Rxd8+ Kxd8 22.Be3 Bf6 23.Nd3 Kd7 with an equal position that was drawn in 46 moves.

Another transposition from a Latvian Gambit occurred in the correspondence game Diepstraaten-Clarke, 1994. The opening moves were 1.e4 e5 2.Nf3 f5 3.d4 d6 4.Bc4, ordinarily arrived at via the move order 2...d6 3.d4 f5 4.Bc4. Twice in his career, Morphy reached this position as Black and developed a piece with 4...Nc6, achieving draws in both cases. In modern times, Mestel and Kosten have favored 4...exd4 which was also the move chosen by Clarke. The game followed with 5.Ng5 Nh6 6.Nxh7 Ng4 7.Nxf8 Kxf8.

This position has been known to favor Black since Nurmi-Mestel, Tjentiste 1975 which ended abruptly with 8.exf5 Qe7+ 9.Kf1 Bxf5 10.Qxd4 Nxh2+ 11.Rxh2 Rxh2 12.Be3 Qe4 13.Qxe4 Rh1+ 14.Ke2 Bxe4, 0-1. Instead Diepstraaten tried 8.Qxd4 Nc6 9.Qd2, an attempt to improve upon 9.Qd5 Qe8 10.Nc3 Nf6 11.Qd1 Nxe4 12.Ne2 Ne5 13.Bd3 Ng4 14.Bxe4 fxe4 15.Bf4 Qg6 16.Qd2 Bd7 17.Nd4 Kg8 18.O-O-O Rf8 19.h3 Qf7 20.g3 Qxa2 (0-1, 23, Sorokin-Maliutin, USSR Championship 1991). But after 9...Nxh2 10.Kd1 Ne5 11.Be2 Qf6 12.Qf4 Rh4 13.Qg3 f4 14.Bxf4 Rxf4 15.Rxh2 Rxf2 16.Nd2 Kg8, Black was clearly better and won in 34 moves.

Finally, a position familiar to both the Latvian Gambit and the Philidor Counter Gambit was reached in the correspondence game Oren-Downey, 1994. The game opened with 1.e4 e5 2.Nf3 f5 3.exf5 d6 4.d4 e4 5.Ng5. Needless to say, the move order might just have easily been 2...d6 3.d4 f5 4.exf5 e4 5.Ng5. Black sometimes plays 5...Nf6 here, but Downey chose 5...Bxf5 instead. After the ensuing 6.Nc3 Nf6 7.f3 Qe7 8.fxe4, Black avoided 8...Bxe4 9.Ngxe4 Nxe4 10.Nxe4 Qxe4+, perhaps fearing Buecker's recommendation 11.Be2! Nc6 12.c3 O-O-O 13.O-O, and continued in true gambit fashion with 8...h6!?.

Following the moves 9.Qe2 Bg4 10.Nf3 Nc6 11.Qe3 O-O-O 12.d5 Nb8 13.Bd3 a6 14.Bd2 g5 15.O-O-O Re8, White was still a pawn up. But Black had pressure on White's backward e-pawn. Now White embarked on a dubious excursion with his queen by 16.Qa7 Bg7 17.Rde1 Nfd7 18.b4 Rhf8 19.Re3 Nb6 20.Rhe1 Bxf3 21.gxf3 Bd4 22.Bf1 Bxe3 23.Bh3+ N8d7 24.Rxe3 Qe5, leaving Black ahead in material with White's queen looking foolishly placed. The concluding moves were 25.Bg4 Rf4 26.h3 Ref8 27.Bf5 R4xf5 28.exf5 Qxf5 29.b5 axb5 30.Nxb5 Qxd5 31.Qa5 c6 32.Rc3 Kd8 33.Nxd6 Qxd6 34.Rd3 Qc7 35.h4 Kc8 36.hxg5 hxg5 37.Qxg5 Qe5 38.Kb1 Qxg5 39.Bxg5 Ne5, and White resigned.

{This article originally appeared in the May-June 1998 issue of Atlantic Chess News}