Maybe I should apologize for my incessant use of two-part titles, but I often find it congenial here to treat one topic in the main text and a second, not necessarily related topic in the linked games. Particularly when I've recently been laboring on some analysis or other, I like to share it here even if it doesn't always relate closely to the previously advertised theme of the column. And such indeed is the case now, the subject of my investigations being a theoretical problem that arose recently in the play of some school-age players of my acquaintance (the analysis presented here is, of course, aimed at more advanced players).

In my role as a volunteer chess coach of school children aged 6-11, I've discovered that minds innocent of chess wisdom are nevertheless capable of posing some surprisingly difficult chess problems. A line of the Three Knights Game has been coming up in our scholastic chess team practices and presents a not-so-easy question for chess theory, or at least, for my understanding of it

Many chess kids like to open with the moves 1. e4 e5 2. Nf3 Nc6 3. Nc3, but here the reply 3...Nf6 is far from universal. It's always nice when 3...Bc5? is played, because it gives me a chance to demonstrate the fork trick, 4. Nxe5 and so forth, in my post-game comments. But 3...Bb4 is also highly favored by chess kids, and it isn't a simple matter for White to establish an advantage against it.

Leading to a roughly balanced system of the Vienna game is 4. Bc4 Nf6 5. d3. This and the similarly boring Italian with d2-d3 are frequent visitors in scholastic chess events, but I like my kids to play more incisively than that. So I've encouraged them to meet 3...Bb4 with 4. Nd5, the theoretically optimal reply. The books give the common-sensical 4...Ba5 5. Bc4 Nf6 6. O-O Nxe4 (7...d6 8. d6 and White will play Bg5) 7. Re1 f5 and now 8. b4!

Quite good for White is 8...Nxb4 9. Nxe5. And after 8...Bxb4 9. Nxb4 Nxb4 10. Nxe5 White advantageously threatens both 12. d3 and 12. Qh5+. If 12...Qf6, then 12. Bb2. Similar is 12...Qh4 13. g3 Bb2. (Readers interested in chess coaching should not assume that I expect my chess kids to master such specific variations; I just show them the moves and point out the key ideas.)

But instead of theory's 4...Ba5, the chess kids like to play 4...Nf6. After all, they've been taught to develop their pieces! This is not such an easy move for White to meet.

It's a little strange that this position, arising from plausible play after only the fourth move of the Open Game, isn't currently considered in the literature. Maybe at one time it was, and I would check the Handbuch des Schachspiels, 1916 edition (Karl Schlecther, ed.), except that the only pages I have of that work are those relevant to the Italian Game. The position is, in fact, a Rubinstein Defense to the Four Knights, with colors reversed. Here "Black," with pale pieces, has an extra tempo.

But how to use it? White can probably obtain some advantage with 5. Bc4 and even the adventuresome 5. c3!?, but I will consider here only the direct 5. Nxb4, which makes the most sense to me and which I believe leads to an advantage for White. After 5...Nxb4 White has 6. Nxe4, 6. d4 and 6. c3.

I. 6. Nxe4 Nxe4 (6...Qe7?! 7. d4 d6 8. a3 leads to trouble) 7. d3 (7. Qe2 Nxc7+ 8. Kd1 Nxa1 9. Qxe4 Qe7 favors Black, because his advantage in time more than makes up for the slight material deficit he may suffer after his a1 knight is finally captured) 7...d6 8. dxe4 (8. Nxf7 Nxf2) 8...dxe5.

Position after 8...dxe5

White has a little advantage in his two bishops, but it's hard to see how he'll be able to make much progress.

II. 6. d4 exd4 (6...Nxe4 7. dxe5 favors White) 7. e5 (7. Nxd4? Nxe4 8. Qe2 O-O) 7...Qe7 8. Qe2 Nfd5 (too frisky is 8...Ng4 9. h3 d3 10. cxd6 Qc5 11. hxg5 Nc7+ 12. Kd1 Nxa1 13. b3 and White has excellent play for his material deficit, for example, 13...Qd5 14. Bb2 Nxb3 15. axb3 Qxb3+ 16. Kc1) 9. a3 (9. Bg5? Qc5) 9...Nc6 10. Bg5 Qc5(10...Qe6?! 11. O-O-O h6 12. Bh4 b6 13. Nxd4 Nxd4 14. Rxd4 Bb7 15. Qe4 and Black is in a bad way).

Position after 10...Qc5

Black's queen is deployed somewhat awkwardly, but his game seems playable enough. If 11. Qc4, then 11...d6.

III. 6. c3! Nc6 7. d4

Position after 7. d4

The position is identical to a key position of the Ponziani (1.e4 e5 2. Nf3 Nc6 3. c3 Nf6 4. d4), except that White's queen knight and Black's king bishop are missing. This difference favors White, because his queen knight is a problem piece in the Ponziani, while Black's king bishop is active there.

IIIa. 7...Nxe4 (identical is 7...d5 8. Nxe5! Nxe5 9. dxe5 Nxe4 10. Bd3, but that way, White also has the option of 10. Be3) 8. Bd3! d5 (somewhat similar is 8...f5 9. Nxe5! Nxe5 10. dxe5) 9. Nxe5! Nxe5 10. dxe5 O-O 11. O-O

Position after 11 O-O

His two bishops and kingside pawn majority offer White a very promising game. An eventual c3-c4 in certain situations, to undermine the knight, is part of his armament.

IIIb. 7...exd4 8. e5 Qe7 (8...Nd5 9. Qb3 is good for White) 9. cxd4 d6 10. Bb5 Bd7 11. O-O dxe5 (11...Nd5 12. Bg5 greatly favors White) 12. dxe5 Nxe5 13. Bxd7 Nfxd7 (13...Nexd7? 14. Re1 Ne4 15. Bg5! and both 15...Qe6 16. Qd4 and 15...f6 16. Qa4 lead to a White win) 14. Nxe5 Nxe5 15. Re1.

Position after 15. Re1

As punishment for his extra pawn, Black is under great pressure on the e-file. Necessary is 15...Rd8 but then 16. Qe2 (16. Qa4+ c6 17. Bf4 f6 18. Qxa7 O-O 19. Qe3 offers White a definite, but smaller, advantage) 16...Kf8 (forced) 17. Bf4 f6 (17...Nc6 18. Qxe7 Nxe7 19. Bxc7 favors White because of his vastly more active rooks) 18. Bxe5 Qxe5 19. Qc2 Qa5 20. Qb3! and Black is severely handicapped by the absence of his king rook. If 20...Qb6 then 21. Rad1. Or 20...Qc5 21. Qxb7 Rd2 and White can calmly play 22. Rf1, threatening Rac1.

IIIc. 7...d6! Of course. The player on the wrong side of the two bishops should generally avoid opening the position.

Position after 7...d6

Black has no obvious problems with his development and White's center is temporarily under pressure, but still it's not easy for the second player. White has the two bishops, and his advantage in space is significant.

IIIc-1. A direct attempt to weaken Black's pawn structure is 8. dxe5 but 8...Nxe5! keeps Black's difficulties to a manageable minimum (worse is 8...dxe5 9. Qxd8+ and White is good after 9...Kxd8 10. Bb5 Re8 11. Bxc6 bxc6 12. Bg5 or 9...Nxd8 10. Nxe5 Nxe4 11. Bb5+) 9. Nxe5 dxe5 10. Qxd8+ Kxd8 11. Bg5 Be6 12. f4 h6 13. Bxf6+ gxf6.

Position after 13...gxf6

Black's h-pawn is a longterm weakness, but he has the better bishop and his position looks playable enough to me.

IIIc-2. 8. Bd3 is a little passive but may be sufficient for advantage. 8...Bg4 9. Be3 O-O 10. h3 (10. O-O? d5! equalizes completely, but 10. d5 Ne7 11. h3 offers White some advantage) 10...Bxf3 (10...Bd7 11. O-O with advantage to White) 11. Qxf3 exd4!? (11...Qe7 12. d5 Nb8 13. c4 a5 14. b3 Nbd7 15. O-O Nc5 16. Bc2 is good for White, who intends a2-a3, b3-b4) 12. cxd4 (12. Bxd4 Ne5!) 12...Re8 13. O-O Qe7 14. Rc1 Nxe4 15. Bb5

Position after 15. Bb5

Black can't prevent White from regaining the pawn on c6, after which White will have good play against Black's backward c-pawn. For instance, 15...Rb8 16. b3 and soon Bxc6.

IIIc-3. Most active and probably best is 8. Bb5 Bd7 9. Bg5! (9. Qd3 O-O 10. O-O exd4 11. cxd4 Qe8 12. Rfe1 Nb4 is not much trouble for Black, and neither is 9. O-O exd4 10. cxd4 Nxe4 11. Re1 d5 12. Qb3 O-O 13. Qxd4 Nf6).

Position after 9. Bg5

Black is under pressure. 9...h6 (9...exd4 10. cxd4 h6 11. Bxf6 Qxf6 12. O-O O-O 13. Rc1 and White has a significant advantage in space) 10. Bxf6 Qxf6 11. d5 Nb8 12. Qb3! and White has the advantage.

Position after 12. Qb3

If 12...c6, then 13. Bd3 (or even 13. Ba6!? Nxa6 14. Qxb7 Ke7 15. Qxa6 cxd5 16. exd5 Rhb8 17. Qe2 and Black lacks full compensation for his pawn) 13...b6 14. O-O O-O 15. a4. And if 12...b6, the 13. Qc4.

The conclusion of my investigations is that after 1. e4 e5 2. Nf3 Nc6 3. Nc3 Bb4 4. Nd5 Nf6, White can obtain a definite advantage with 5.Nxb4 Nxb4 6. c3! Nc6 7. d4. I invite the reader's comments upon the foregoing variations, which are included in the zipped file linked below.

I now turn to the topic that I advertised in last month's "Hard Chess," the Schliemann Defense to the Ruy. I have advocated this defense before in this space, expressing my view that it's a legitimate try for dynamic play. Black's third move, indeed, implicitly asserts that White's bishop is misplaced on b5! This claim is most clearly borne out after 4. d4 fxe4 5. Nxe5 Nxe5 6. dxe5 c6, when any retreat of the bishop exposes White to the check from a5 and the loss of his e-pawn (White's best idea is to sack the piece with 7. Nc3 leading, like many other lines of the Schliemann, to a difficult game where both sides have winning chances).

The main line of the Schiemann is 1. e4 e5 2. Nf3 Nc6 3. Bb5 f5 4. Nc3! fxe4 5. Nxe4 d5! 6. Nxe5! dxe4 7. Nxc6 Qg5! (Black again points to the "misplacement" of the bishop) 8. Qe2 Nf6 9. f4 Qxf4 10. Ne5+ c6 11. d4 Qh4+ 12. g3 Qh3 13. Bc4 Be6 14. Bg5 (14. Bf4 is a major alternative) 14...O-O-O 15. O-O-O Bd6.

Position after 15...Bd6

White's advantage is chiefly made up of his beautifully centralized knight. Black's king position is a little airy, and a queenside attack, possibly involving the sacrifice of the knight on c6, looms if Black can't get something going. On the other hand, Black is nicely developed, and White's kingside is brittle. A key question will be whether Black's isolated, passed e-pawn is strong or weak. This can become a debate about the e- and f-pawns together if White exchanges on f6. If White doesn't exchange the Black's knight, its natural square is d5. White's eventual push d4-d5 may not be a big problem if Black is prepared to answer with c6-c5. A question for White is whether to push his c-pawn to c4, which gains space but leaves the d-pawn somewhat weak.

I will not treat further the diagrammed position, mostly because I am currently contesting it in my ongoing game against David Burris in the final section of the 13th U.S. championship. I don't want to play here the role of my very strong opponent's devoted research assistant. Instead, I will treat a derivative idea that arises when White, instead of 10. Ne5+ in the line given above, plays 10. d4. Ever since the immortal game Fisher-Matulovic, Herceg Novi blitz, 1970, it has been believed that 10...Qd6 is the required reply. Matulovic attempted transposition into the main line with 10...Qh4+ 11. g3 Qh3 but ran into 12. Bg5!? and was rolled off the board.

Position after 12. Bg5

But recent analysis by Ivanov and Kulagin in their indispensible work, Play the Schliemann Defense, Olbrich 1994, suggests that Fischer's piece sacrifice is good for no more than a draw and, as Fisher chose to play it, even leads to advantage for Black. I present my own commentary on the Fisher-Matulovic game, with generous quotations of Ivanov's and Kulagin's analysis but also with some additions and improvements, in the first game linked below.

A curious echo of that game arose in my own practice when an opponent selected, in the main line instead of 13. Bc4, the move 13. Bg5?!!

Position after 13. Bg5

Maybe White thought he was playing Fischer's line; I don't know. But the move presents some interesting problems that I treat in Guehn-Morss, NAPZ-M62, linked below.

Game 1: Fischer-Matulovic, Herceg Novi blitz, 1970

Game 2: Guehn-Morss, NAPZ-M62

For version 4 browsers: Java replay of all games

Next Month: The Queen's Gambit Point Machine