Did the old chess masters know something the modern champions don't? The question occurred to me when I was reading grandmaster Valeri Beim's recent book "How to Play Dynamic Chess," issued by Gambit Publications. Beim is a great storyteller and as usual he gives a good workout to the old classic Soviet games. His game selection is very good, sometimes too good for a regular tournament player. The champions, Beim presents, often play too brilliantly to be understood. The following game is a good example.

Old Masterpiece Revised

One of the most exciting players in chess history was three-time Soviet champion Leonid Stein (1934-1973). When he was asked which game he considered his best, he pointed to his encounter with the legendary Estonian Paul Keres from Moscow 1967. In my opinion, by not following the old masters, Stein misplayed the opening, but he corrected himself wonderfully at the right time. Less talented players would find themselves caught in a maze.

Stein - Keres

1.e4 e5 2.Nf3 Nc6 3.Bb5 a6 4.Ba4 Nf6 5.0-0 Nxe4 6.d4 b5 7.Bb3 d5 8.dxe5 Be6 9.c3 Bc5 10.Nbd2 0-0 11.Bc2 f5 12.Nb3 Bb6 13.Nfd4 Nxd4 14.Nxd4 Qd7 (David Janowski's novelty dating back to 1900. After the white knight takes the bishop on e6, white may have problems protecting his pawn on e5. The famous Mackenzie's piece sacrifice from 1882, 14...Bxd4 15.cxd4 f4 15.f3 Ng3!?, was analyzed to death in the 1950s, but white can avoid it with 15.Qxd4.) 15.f3 Nc5 16.Kh1 (Praised too highly. The white king simply leaves a dangerous diagonal.) 16...Nb7 ("This move looks unfortunate. All black's favorable results in this extremely rare variation have involved 16...Rae8," says Beim.)

17.Be3?! (There is a point many commentators, including Beim, missed. It seems more logical to soften the queenside with 17.a4! as it was played in the inaugural game Billecard-Janowski, Munich 1900. Could the old masters have grasped the position better than the Soviet champion? After 17...c5 18.Nxe6 Qxe6, however, Billecard continued with the boring 19.f4?!, allowing 19...c4! with the idea to bring the knight back to e4 via the square c5.

What else can white try on his 19th move? We know that 19.Be3 transposes to our game. But I have no doubt that in this position Stein would come up with the following exciting piece sacrifice:19.axb5 axb5 20.Bxf5!!. Black is in dire straits either after 20...Qxe5 21.Rxa8 Rxa8 22.Qd3!; or after 20...Qxf5 21.Qxd5+ Qf7 22.e6! Qe7 23.Rxa8 Rxa8 24.Bg5 Rd8 25.Qe4 Qxg5 26.Qxb7, threatening both 27.Qxb6 and 26.Qf7+ Kh8 27.e7 and white should win. Clearly, 17.a4! is the right answer to 17...Nb7.) 17...c5 18.Nxe6 Qxe6 (A critical position.)

19.a4! (Better late than never. Stein corrects himself and Keres is in trouble. The move probably made a great impression on commentators because white leaves his pawn on e5 unprotected. But who will play like that? Grandmaster Vladimir Simagin wrote in 1967 that the overwhelming majority of masters would have made the appealing move 19.f4 without thinking.) 19...Na5 (Simagin pointed out that after 19...Qxe5 20.Re1! Bc7 21.Bg1 Qd6 22.axb5 axb5 23.Rxa8 Rxa8 24.Bxf5 white is better.) 20.Bf2! (Highly admired, but Stein could have played 20.Bg1! as well, because taking the e-pawn is not good for black; for example 20...Qxe5 21.axb5 axb5 22.b4 or 20...Nc4 21.b3 Nxe5 22.axb5 axb5 23.Bxf5 Qd6 24.Rxa8 Rxa8 25.Qe2 with white's advantage in both cases.) 20...Kh8 (Keres removes his king because in some variations his d5-pawn may fall with a check. After 20...Qxe5? 21.axb5 axb5 Beim correctly states that 22.b4 Nb7 23.Rxa8 Rxa8 24.Re1 Qd6 25.Bxf5 is better for white and after 22...Nc4? 23.Rxa8 Rxa8 24.bxc5 Bc7 25.Bg3! f4 26.Bxf4 white wins. Moscow grandmaster Evgeny Vasiukov wrote in the tournament bulletin that after 20...Nc4 white can play 21.b3! Nxe5 22.axb5 axb5 23.Rxa8 Rxa8 24.Re1 with an unpleasant pin. But 24.Bxf5 is simpler, for example 24...Qd6 25.Bg3 Re8 26.Re1 Re7 27.Be4! d4 28.cxd4 cxd4 29.Qe2! and white should win.)

21.Re1 ("It is well known that the queen is not a good blocker of passed pawns. Besides, White has two bishops, and Black's pawn structure is quite weak. It is not surprising that White wins without major problems," summed up the position grandmaster Michal Krasenkow.) 21...Ra7 22.Qe2 b4? (According to Vasiukov, even after 22...c4 23.axb5 axb5 24.Rad1 white has a clear advantage.) 23.cxb4 cxb4 24.Bxb6 Qxb6 25.Rad1 Qc5 26.Bd3 (Stein's friend, grandmaster Eduard Gufeld, remarked: "Black's pawns remind one of a jaw full of bad teeth. Stein assumed the duties of a dental surgeon.") 26...Qb6 27.Bb1! Qc6 28.Qd2 Qxa4 29.Qxd5 Nc6 30.Bxf5! (The final strike. On 30...Ne7 comes 31.Qc5 attacking the rook on a7 and 30...Rxf5 is met by 31.Qd8+! and white mates.) 30...Qb5 31.Qd6 ("31.Qxb5 axb5 32.Bd7 wins another pawn. But Stein decides to win the game with an attack," says Vasiukov.) 31...Qb8 32.Qxc6 Rxf5 33.e6 Re7 34.Rd7 Re8 35.Rb7 Qc8 36.Rc7 Qb8 37.Qd7 Rg5 38.f4 Rg6 39.f5 Rg5 40.f6 Black resigned. Solution to today's problem by A. Mann (White: Kd8,Qd1,P:e2; Black: Kd5,P:c4,c5,c6,d4,d6,e4,e5,e6): 1.Qa1! d3 (or 1...e3 2.Qh1+ e4 3.Qh5+ e5 4.Qf7 mate; or 1...c3 2.Qa2+ c4 3.Qa5+ c5 4.Qa8 mate.) 2.Qc3! e3 (2...dxe2 3.Qd2 mate) 3.exd3 e4 (3...e2 4.Qxc4 mate) 4.dxc4 mate.

White mates in four moves.