|
|
|
The Skittles Room
Translate this page
|
The Bxh6 Sacrifice by Steven B. Dowd Introduction LeMoir, in his book Essential Chess Sacrifices, notes that Bxh6, typically followed by the recapture with the queen, is probably the sacrifice with the greatest success rate. We've recently seen two books published about the Greek gift, Bxh7+. That sacrifice has been a frequently seen topic in the chess literature for years. Why no such analysis on taking the h-pawn advanced one square? Probably because the Bxh7+ sacrifice is flashier. The Bxh6 sacrifice may be more successful, but it also requires more work. White receives two pawns for the piece, and often a lasting initiative based on mate threats, but the outcome is not as clear, especially in the face of dedicated defense. My goal is to provide an introduction to this sacrifice. I hope other players take up an analysis of this sacrifice as well, so it's potential can be better understood by the chess-playing public. Often, I've noted, players make such sacrifices intuitively – but intuition combined with specific knowledge is a powerful tool, as the great Mikhail Tal showed us! We are looking at a specific kind of Bxh6 sacrifice here. The bishop captures on h6 and the queen then captures back immediately, or soon after in most cases, giving two pawns for the bishop and an attack. We will look at a few exceptions to the recapture on h6 by the queen, for example, in the Bronstein-Keres game, where a different Black defense would mean that not recapturing with the queen immediately would have been best. One of the earliest examples I could find of this sacrifice showed the great Anderssen on the receiving end: Von Kolisch – Anderssen 1.e4 c5 2.Nf3 e6 3.d4 cxd4 4.Nxd4 Nf6 5.Bd3 Nc6 6.Be3 d5 7.exd5 exd5 8.0-0 Bd6 9.h3 h6 10.Nc3 0-0 11.Qd2 Re8 12.Rad1 Bc7 13.Rfe1 Qd6 14.Nf3 a6
An innocuous position, or so it seems. But we have the "usual line-up" for the sacrifice, a battery with the white queen on d2 (sometimes c1) and white bishop on e3, ready to strike on h6. Kolisch decides to mix it up with the sacrifice. 15.Bxh6 Rxe1+ 16.Rxe1 gxh6 17.Qxh6 Ne4 18.Qh5 f5? 18...Be6! and White has at best a slight advantage. 19.Nxd5!! Qxd5 20.Bxe4 Qd7 21.Bd5+ Kg7 22.Qg5+ 1-0 After 22...Kh7 23.Qg8+ Kh6 24.Bf7!, Black not only loses the queen, but will be mated soon anyway. Back to LeMoir. His first example is an instructive one: Riumin – Budo
White's advantages after 13.Bxh6 gxh (declining the sacrifice simply loses a pawn but is in rare cases the best defense) 14.Qxh6 are many, according to LeMoir:
Your engine will probably indicate that White is ahead the equivalent of nearly three pawns here. Black has three possible defenses: 14...Re8, 14...Bd6, and 14...Qe7. LeMoir considers 14...Qe7 to be the best (in the game Black practically surrendered immediately with 14...e5?, although this does stop the g4 advance) and believes White should then go after the knight with g4-g5. That is certainly good enough, but after 14...Qe7 15.Ng5!, just adding more pieces to the attack, what does Black do? If 15...Rd8, then 16.Bh7+ Kh8 17.Rad1!! (highlighting the weakness on f7), Black is lost. At best he can trade the rooks 17...Rd5 18.Be4+ Kg8 19.Rd3! Rxd3 20.Bh7+! Kh8 21.Bxd3+ Kg8 22.Bh7+! (yes again, and this is not repeating the position) 22...Kh8 and now 23.Rd1! White will now lift this rook without any interference from Black. The most recent, and by far the best book on the Greek gift, Edwards's Sacking the Citadel, notes that a taxonomy (originally developed by Voellmy) provides us with hints as to whether the bishop sacrifice is sound. The above four advantages of LeMoir's probably start the taxonomy for Bxh6, although it must be noted that White has one potential added benefit. Here he has two pawns for the piece, not a material equivalent, but close, and amongst players of equal strength, trying to win essentially just one pawn up is difficult. Also, White can try, should the attack not break through, to use the exposed position of the black king to win an additional pawn. In many cases he can play for a repetition of position against the black king that Black cannot avoid. In the Greek gift, there is always worry that the black king will escape; this is rarely the case with the Bxh6 sacrifice, where Black must cleverly place his pieces for defense around his king, with the king held tight in a cage. As in the Edwards/Voellmy taxonomy, not all of the above must be present for the sacrifice to succeed. For example, although the bishop ideally sits on the b1-h7 diagonal, it does not have to, especially if it can get there quickly. Nf3-g5 is another plus that sometimes is not needed for the sacrifice to work. And even though the moves Bxh6 and Qxh6 usually occur one after the other, there are times when a delayed capture on h6 by the queen works as well. We'll see this in the next section. A Favorite Classic Game Like most players of my generation, I grew up marveling at the accomplishments of Soviet-era players. Thus, it is no wonder that I remember this game well, as it was the first time I saw the Bxh6 sacrifice played not as amongst patzers, with an almost immediate win for White, but as one in which both sides fought. The sacrifice provided a long-term advantage to White. Bronstein, David – Keres, Paul 1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.e3 c5 5.Bd3 b6 6.Nge2 Bb7 7.0–0 cxd4 8.exd4 0–0 9.d5 h6 10.Bc2 Na6 11.Nb5 exd5 12.a3 Be7 13.Ng3 dxc4
White's eleventh and thirteenth moves were preludes to the sacrifice-to-come. Here White has an advantage that we have not seen yet: the knight on g3 is ideally placed to jump to f5, giving him further control over the dark squares. Engines indicate Black is better here – or perhaps better stated, that White does not have enough compensation for the two pawns sacrificed. 14.Bxh6 gxh6 15.Qd2 Nh7 No battery formation here, but Black cannot protect the h6-pawn, so he starts to regroup, with the knight headed for the now-familiar defensive square f8. Over-the-board, Black did not think he could fight off the attack of the queen, bishop, and knight against his king without sacrificing back material. Many commentators of the day recommended instead 15...Nc5 for Black. That is also the move preferred by the engine, which sees that and ...Ng4 – at least initially – as providing equality. If 15...Ng4 16.Qf4, Black then has the choice 16...f5, which after 17.Nxf5 h5 18.h3 Rxf5 19.Qxf5 Nf6, is similar to the game in some respects, just another doomed fishing expedition. After 20.Qg6+, White is winning. He will bring the rook(s) to the e-file and Black is helpless. Bondarevsky analyzed 15...Nc5 extensively, and thought White was winning here as well. After 15...Nc5, White must delay the capture even further – 16.Rae1!, as capturing on h6 would allow the defense 16...Be4. So 16.Rae1! Nd3 17.Bxd3 cxd 18.Nf5 Be4 19.Nbd4 Re8 20.Nxh6+ Kf8 21.Qg5 Bg6 22.Rxe7! Rxe7 23.Qxf6 Re4 24.Qh8+ Ke7 25.Nhf5+ Bxf5 26.Nxf5+ Ke6 27.Qh3!
Now White certainly is not losing – this is one of those types of positions where the knight is better than "the clumsy rook" – but is he winning? 27...d5! seems best, and after 28.Nd4+ Kd6 29.Qxd3, I would prefer White. He has a pawn for the exchange now, so as Edmar Mednis used to say, he is only "half-a-pawn down," but his compensation is obvious. The a7-pawn will probably fall as well, and the black king isn't really safe anywhere. 16.Qxh6 f5 17.Nxf5 Rxf5 18.Bxf5 Nf8 19.Rad1 Bg5 Black has averted immediate loss, but his king is denuded, with no pawn protection at all. White is winning. 20.Qh5 Qf6 21.Nd6 Bc6 22.Qg4 Kh8 23.Be4 This little trick works, but certainly 23.Rd4! is good here. Some have recommended 23.Qxg5 here as a culmination of the attack, but it seems to me to simply dissipate White's advantage, going "only" into a winning, but still difficult ending. 23...Bh6 24.Bxc6 dxc6 25.Qxc4 Nc5 26.b4 Nce6 27.Qxc6 Rb8 28.Ne4 Qg6 29.Rd6 Bg7 30.f4 Qg4 31.h3 Qe2 32.Ng3 Qe3+ 33.Kh2 Nd4 34.Qd5 Re8 35.Nh5 Ne2 36.Nxg7 Qg3+ 37.Kh1 Nxf4 38.Qf3 Ne2 39.Rh6+ 1–0 He can't take the knight, and 39...Nh7 also leads to mate, just a slower one. Other Sources When I first started putting together this article, I was rather surprised to note that Vukovic, in his classic The Art of Attack in Chess, doesn't really discuss this sacrifice, or consider h6 a "focal point" as he does the squares h7, g7, and f7. Spielmann does, in his section on "King's field sacrifices" in his classic The Art of Sacrifice in Chess (we, of course, consider the sacrifice ...Bxh3 the same as White's Bxh6). Becker – Spielmann
Becker played 19.N4f3?, which Spielmann considered to be a mistake; after 19.Nf1, he thought White was in little danger. Indeed, many years after this, Bent Larsen noted the usefulness of Nf1 or Nf8, and in fact stated, tongue-in-cheek as always, "With a knight on f8 you can never get mated." My engine doesn't see any danger ahead, and in fact notes several plausible moves instead of Becker's. Fritz 11 prefers 19...Nf5 as a reply to Becker's move, considering the position equal. After the game moves 19...Bxh3! (Spielmann's exclam, although he also calls it "an everyday king's field sacrifice," with "KR3 on either side ... a favorite sacrificial point") 20.gxh Qxh3 Here Fritz "sees" Spielmann's recommendation of 21.Bf1! Spielmann himself notes that now 21...Qg3+ 22.Bg2 Nhg4 gives Black an attack, but there is no decisive result in sight. Instead, 21.Re2? followed and Spielmann played 21...Bxf4!?, tossing another piece into the mix. But better was the interpolation, he notes, of the line clearance sacrifice 21...Nhg4! 22.Rg2 d4!!. White took the bishop with 22.exf, but he should have interpolated 22.Qd4! according to Spielmann. Again, I am not so sure, 22.Nf1! looks better to me, and after 22.Qd4 Qg4+! (not 22...Qg3+) and Black is still winning. On 23.Kf1 Ne4! and on 23.Kh1 Nh5! looks to bring a strong Black advantage in either case. The game finished 22...Rxe2! +- 23.Bxe2 Qg3+ 24.Kh1 Nhg4 25.Rf1 Qh3+ 26.Kg1 Qg3+ 27.Kh1 Re8 28.Qd3 Qh3+ 29.Kg1 Qg3+ 30.Kh1 Qh3+ 31.Kg1 Ne3 32.Ne1 Qg3+ 33.Kh1 Qh3+ 34.Kg1 Qg3+ 35.Kh1 Nxf1 36.Bxf1 Re3 37.Qf5 Rxe1 38.Be5 Qg4 39.Qxg4 Nxg4 40.Kg2 Rd1 0–1 I've kept a pretty extensive collection of books on chess sacrifices, and the above two examples were the most prominent ones I could find. For some reason the Bxh6 sacrifice just isn't covered by many authors, despite its success rate and routine appearance in games. A PDF file of this week's Skittles Room article, along with all previous articles, is available in the ChessCafe.com Archives. The ChessCafe.com Skittles Room is open to reader contributions. If you have an article that you would like published, please write to us via our Contact Page. Comment on this month's column via our Contact Page! |
Purchases from our
|
|
|
|
[ChessCafe
Home Page] [Book Review]
[Columnists] © 2012 BrainGamz, Inc. All Rights Reserved. |
|
