One such I recently spotted is My Dream Game by Philip Corbin in the January 2005 issue of Chess.
It concerns his win with the Elephant Gambit at the 2004 Olympiad against an international master.
The gambit starts 1.e4 e5 2.Nf3 d5, which apparently has served PC well, at least judging by his statistics in ChessBase's 2021 Mega database.
In 20 games he has amassed a score of +12=1-7, which is 62%. His average rating in those games was 2226, and his performance rating is given as 2242, albeit only 11 of his opponents had a rating.
However, if you take out the unrated opponents, and a game from a simul, PC's score becomes +3=0-7. His average rating in those 10 games was 2208, and his performance rating is given as 2162.
Nevertheless I thought it would be fun to go through the game published in Chess, using modern engines to cast a critical eye on the key moments.
Olympiad (Mallorca) Round 1
Comments in italics are by PC
1.e4 e5 2.Nf3 d5
At this olympiad I played (the Elephant Gambit) four times, scoring three wins and a draw.
3.exd5
I deem 3,Nxe5 Bd6 4.d4 dxe4 OK for Black.
3...Bd6
3...e4?! is well met by 4.Qe2.
4.d4!?
More popular in Mega21 is 4.Nc3, which has been played by Karpov, but the text scores 15 percentage points higher and is much preferred by Stockfish14 and Komodo12.1.1.
4...e4 5.Ne5 Nf6 6.Bb5+ c6!?
The safer, but more-placid, alternative is 6...Bd7 with an unclear position. I was playing for maximum complications.
7.dxc6 bxc6 8.Nxc6 Qb6!
The only move that makes the line playable. This game is the first time the line appears in Mega21, but the implication of PC's notes is that it was known to Elephant aficionados.
 |
| Position after 8...Qb6! |
*****
*****
*****
*****
*****
9.c4
Later games saw 9.Nc3, which is slightly preferred by Komodo12.1.1. The text is preferred by Stockfish14.
Note that, as PC points out, 9.Nxb8?? loses a piece to 9...Qxb5. Less clear, but probably also good for Black, is 9.Nxa7+?! Bd7 10.c4 Qxa7 11.Nc3, when White has three pawns for a knight, but the engines prefer Black. Another possibility in the latter variation is 10.Nc8, when PC's 10...Bb4+! seems strong, eg 11.c3 Qxb5, although after 12.a4!?, Komodo12.1.1 reckons Black is only slightly better (Black is winning, according to Stockfish14).
9...Nc6 10.d5
The engines prefer 10.Nc3!?, although Komodo12.1.1 gives the nod for a while to the text. After 10.Nc3!? best play, according to the engines, runs 10...0-0 11.c5 Qc7 12.cxd6 Qxd6 13.Bxc6 Qxc6 14.0-0, when White is a pawn up.
10...0-0 11.Bxc6 Ng4!
PC also gives this an exclamation mark, and rightly it would seem; certainly the text is the engines' choice. The crude threat is checkmate on f2, but there is much more to the move than that.
12.Qe2
This is the only way to keep a slight edge, according to the engines.
12...f5!?
Offering a whole rook to gain a tempo in the attack! The idea of the move came to me when preparing for a game in the 2004 Barbados championship. I had not gotten the chance to play it then (so) this game was the first test over the board of my new rook-sac idea!
PC gives the text two exclamation marks, but Komodo12.1.1 prefers 12...Ne5, when it reckons 13.0-0 Nxc6 14.dxc6 Qxc6 15.Nc3 is only slightly better for White. Stockfish14 gives 12...Rb8 13.h3 Ne5 14.0-0 f5, when it reckons Black's initiative at least compensates for his two-pawn deficit. PC rejected this latter line because of 14.Qxe4!?, which wins a third pawn, but Stockfish14 continues 14...f5 15.Qe3 Qc7, claiming Black is winning (Komodo12.1.1 'only' gives Black the upper hand).
13.Bxa8
The engines come to narrowly prefer this over the arguably more-practical 13.Nc3, but in any event the position is very tricky for White and, at least for a while, Komodo12.1.1 prefers 13.Nc3.
13...f4 14.f3
Stockfish14 prefers 14.Nc3, but then 14...f3 15.gxf3 exf3 requires White to find 16.Be3 Nxe3 17.Qxe3 Bc5 18.Na4! Qa5+ 19.Qd2 Qxa4, when White is up the exchange and two pawns but Black still has a strong attack.
14...Ne5!?
This seems even stronger than 14...exf3 14.gxf3 Ne5, when both 15.Kd1 and 15.Bc6 keep White in the game, according to the engines.
15.Nc3
Not 15.Qxe4? Bf5 16.Qe2 Nd3+ 17.Kd1 Rxa8, when Black has a winning attack, according to the engines. Similarly 15.fxe4? Bg4 16.Qc2 f3 is horrific for White.
15...exf3 16.gxf3 Bf5 17.Ne4
Komodo12.1.1 at first suggests 17.c5, but the simple 17...Qxc5 18.Bc6 Nd3+ leads to advantage for Black after 19.Kd1 Nf2+ 20.Ke1 Nxh1. Its second suggestion of 17.Bc6 is dismissed by Stockfish14 with 17...Qd4, the point being Komodo12.1..1's 18.h4 fails to 18...Bc5 19.Qg2 Nd3+ 20.Kf1 Nf2.
17...Bb4+ 18.Kd1?!
White's best chance seems to be 18.Bd2 Bxd2+ 19.Nxd2, although both PC's 19...Nd3+ and the engines' 19...Re8 leave Black well on top.
18...Qd4+ 19.Kc2 Nxf3! 20.Rd1 Bxe4+ 21.Kb3 Rb8!
Black is 'only' the exchange and a pawn down, so he offers his queen.
22.Bc6
This allows a quick mate, but 22.Rxd4 Nxd4+ 23.Ka4 Nxe2 also weaves a mating net, eg PC points out 24.Bc6 Bc2+ 25.b3 Nc3#.
22...Bd2+ 23.Bb5
If 23.Ka4, the engines give a two-bishop mate with 23...Qxb2!? (the prosaic 23...Rb4+ also mates) 24.Bxb2 Bc2+ 25.Ka3 Bb4#.
23...Rxb5+!
Yet another sac.
24.cxb5
Or 24.Ka3 Ra5+ 25.Kb3 Qb6#.
24...Qb4#
An amazing game in which I have been unable to find a move by White that deserves an unadorned question mark. Is the Elephant Gambit really that good?!
No comments:
Post a Comment