Catalan
1.c4 Nf6 2.Nc3 e6 3.g3 d5 4.cxd5 exd5 5.d4 c6 6.Bg2 Bd6 7.Nf3
 |
| What's in a name - or, to put it another way, what would you call this opening? |
*****
*****
*****
*****
After 4.cxd5 ChessBase calls it "A17: English Opening: 1...Nf6 with ...Bb4." After 7.Nf3 it changes to: "D35: Queen's Gambit Declined: Exchange Variation." I reckon it is a Catalan, albeit one in which White has captured on d5 at an unusually early stage.
7...0-0 8.0-0 Re8 9.Re1 Bg4
This is the most popular move (just) in ChessBase's 2024 Mega database, but Stockfish16 and Komodo14.1 are unimpressed, preferring 9...Bf5.
10.h3 Bh5 11.a3 a5 12.Bg5 Nbd7 13.e4!?
There is a game in Mega24 in which a 2163 played 13.Qc1. The engines much prefer the text.
13...dxe4 14.Rxe4!? Rxe4 15.Nxe4 Be7
 |
| How should White proceed? |
*****
*****
*****
*****
16.Nc3!?
The engines like this retreat, although they also like 16.Nxf6+ Bxf6 17.Bxf6 Nxf6 18.Qb3, claiming a tiny pull for White.
16...h6 17.Bf4 Nb6 18.g4 Bg6 19.Ne5
 |
| This is rather a good example of a position in which dynamism more than makes up for the structural weakness of having an isolated queen's pawn |
*****
*****
*****
*****
19...Nbd5
Preserving the bishop-pair with 19...Bh7?! looks problematic after 20.Qb3.
20.Nxg6 fxg6 21.Bg3 Bd6 22.Be5!
 |
| How should White respond? |
*****
*****
*****
*****
22...Qc7
This may be better than 22...Bxe5, when best-play. according to the engines, runs 23.dxe5 Nxc3 24.bxc3 Qxd1+ 25.Rxd1 Nd5 26.Bxd5+ cxd5 27.Rxd5, after which White has won a pawn, but the position is not clearcut (White has the upper hand - Komodo14.1; White has a tiny pull - Stockfish16).
23.Nxd5 Nxd5 24.Bxd5+ cxd5 25.Qb3 Qc6 26.Bxd6 Qxd6 27.Qxb7 Rxb2 28.Qa7 Rxb2 29.Qxa5
 |
| White has won a pawn, but Black is active and so is completely equal, according to the engines |
*****
*****
*****
*****
29...Rb3
Arguably more clearcut, but requiring a huge belief in one's calculating abilities, is the engines' 29...Rxf2!?
30.Kg2 Qe6?!
Probably better are 30...Kh7 and 30...Rd3.
31.Re1 Qf7
The only move to keep the game going.
32.Re3 Rxe3 33.fxe3
 |
| How would you assess this queen-and-pawn ending? |
*****
*****
*****
*****
As I understand it, two factors stand out in a queen-and-pawn ending. The first is that a passed pawn nearly always gives excellent winning chances, but these can be cancelled by having an exposed king. The engines reckon this position is equal.
33...Qe6 34.Qd2 Qa6 35.Qb2 Qd3 36.Kf2
 |
| The practical problem, from the defender's view, is that the player a pawn up can keep on probing in the knowledge that, barring a horrific blunder, the worst that can happen is the game is drawn |
*****
*****
*****
*****
36...h5?!
Trying for counterplay, but I probably should have stayed pat with, for example, 36...Qc4.
37.Qe2!?
The engines agree this is strong.
37...Qxa3 38.gxh5
 |
| White no longer has the 'holy grail' of a passed pawn, but is nevertheless winning, according to the engines |
*****
*****
*****
*****
38...Qd6!? 39.hxg6 Qh2+ 40.Ke1 Qg1+ 41.Kd2 Qxg6
 |
| I find it hard to comprehend how this position is so much better for White than the position with a passed pawn after 36.Kf2, but the engines are insistent |
*****
*****
*****
*****
42.Qg4 Qc6 43.Qf5 Qc4
Nevertheless, it is far from obvious, at least to me, how White should continue.
44.Qc2
The engines suggest the counterintuitive 44.Ke1!?
44...Qb4+
Apparently 44...Qf1!? gives better drawing chances.
45.Qc3 Qb8 46.Kd3
And here the engines like 46.Qc6.
46...Qb5+ 47.Kd2 Qb8 48.Kc1
Again Qc6 is the move, according to the engines.
48...Qb7 49.Qc5 Qa8 50.Kb2 Qb7+ 51.Kc3 Kh8!?
Played so that White cannot capture on d5 with check.
52.Qf8+ Kh7 53.Qf5+ g6!? 54.Qf4 Qc8+ 55.Kd2
 |
| What should Black play? |
*****
*****
*****
*****
55...Qc4
I thought 55...Qxh3!? 56.Qf7+ Kh6 57.Qxd5 made White's task easier, thanks to White having a pair of connected passed pawns.
 |
| However the Syzygy endgame tablebase shows the position is drawn, but there is only one move that draws |
*****
*****
*****
*****
In the diagrammed position Black needs to find 57...g5!! as everything else loses. So objectively I should have grabbed the h pawn at move 55, but I believe the practical chances of Black drawing from the diagram are low.
Back to the game:
56.h4 Qa2+ 57.Ke1 Qg2 58.Qg5 Qe4 59.Kf2 Qe6 60.Kg3 Qf7 61.Qf4 Qe6?
The queen should stay on the seventh rank.
 |
| White is winning anyway, but now has a sequence that puts the result beyond doubt |
*****
*****
*****
*****
62.Qe5
After 62.Qc7+ Kh6 63.Qe5 Qg8 64.Qg5+ Kh7 65.h5 Qb8+ (pawn-endings are lost for Black) 66.Kh3 Qc8+ 67.Kh4 gxh5 68.Qxh5+ Black is pretty much hopeless, but 62...Kg8 63.Qe5 Qf7 64.Kg4 Qd7+ 65.Kg5 Kh7 66.h5 Qd8+ 67.Kg4 Qc8+ 68.Kf4 Qf8+ 69.Kg3 is little better.
62...Qf7 63.Kg4 Qd7+ 64.Kg5 Qd8+ 65.Kg4 Qd7+ 66.Kg3 Qf7 67.Qg5
Stockfish16's 67.h5!? seems strong, the point being that after 67...gxh5 68.Kh4 Qf2+ 69.Kxh5 White has a tablebase win.
67...Qe8?!
Almost certainly better is 67...Qf1.
68.Kf2
This time both engines suggest h5, the move I have been trying to prevent. After 62.h5!? gxh5 63.Kh4 White will recover the sacrificed pawn, leaving the black king with no cover.
68...Qf7+ 69.Ke2 Qe8?
Lack of concentration, but, as I pointed out earlier, the practical chances very much lie with the attacker in this type of ending.
70.Qxd5 Qb8 71.Qf7+ Kh8
Not 71...Kh6? as the queens come off after 72.Qf4+.
72.Qf6+
The position is a tablebase draw after 72.Qxg6? Qh2+ (or 72...Qb2+).
72...Kh7 73.Qe7+ Kh6 74.Qg5+ Kh7 75.Qf4 Qb7 76.Kf2 Qb2+ 77.Kf3 Qd2 78.Ke4 Qc2+ 79.Kf3 Qd1+ 80.Kg2 Qe2+ 81.Kh3 Qd1!?
 |
| White to play and win (ie avoid a draw by repetition) |
*****
*****
*****
*****
82.Qg3?
This seems the only move that does not win (apart from giving the queen away with, for example, 82.Qf3??). The engines like 82.Kg2!?, but perhaps easiest for a human to calculate is 82.Qf2.
The game finished:
82...Qh1+ 83.Kg4 Qe4+ 84.Kh3 Qh1+ 85.Qh2 Qf3+ 86.Qg3 Qh1+ ½–½
No comments:
Post a Comment