A trap in the Smith-Morra

Probably the best-known trap in the Smith-Morra Gambit occurs after 3...dxc3 4.Nxc3 Nc6 5.Nf3 d6 6.Bc4 Nf6?! 7.e5! (see diagram). After accepting the pawn Black has played three eminently reasonable-looking developing moves, but is suddenly in trouble! In the game below, he actually stumbles into checkmate - after trading queens, no less. In the final position, Black resigns in light of 10...Ke8 11.Nc7#! Marc Esserman writes, "I've lost count how many times I have delivered this comedic finish." (Mayhem in the Morra!, p. 205)

This trap isn't actually as fearsome as its reputation. After 7.e5!, Black certainly must avoid (a) 7...Nxe5?? 8.Nxe5 dxe5? 9.Bxf7+! Kxf7 10.Qxd8. Esserman considers his best line to be (b) 7...Ng4! 8.exd6 exd6! 9.O-O Be7 10.h3 Nge5 11.Nxe5 dxe5 12.Qh5 O-O 13.Rd1 Qa5 14.Be3, when "White's full range of motion more than compensates for Black's extra e5-pawn." Unfortunately the attractive 8.e6!? has, he says, been refuted by Rybka: 8...fxe6! 9.Ng5 Nge5 10.Nxe6 Qa5! 11.Bb3 Bxe6 12.Bxe6 Qa6! with a small advantage to Black. After (c) 7...dxe5 8.Qxd8+, Esserman says that Black's best is the surprising (c1) 8...Kxd8! 9.Ng5 Na5! when the obvious 10.Nxf7+? Ke8 11.Nxh8 Nxc4 will leave Black up material after he wins the beast on h8. Instead, he recommends 10.Bb5! Be6! 11.Nxe6+ fxe6, when Black is two pawns up but White is slightly better in light of his bishop pair, lead in development, and Black's tripled e-pawns. After the game continuation (c2) 8...Nxd8? 9.Nb5!, Black's best is 9...Rb8! (instead of 9...Kd7??) 10.Nxe5 (threatening Nc7#!) e6 11.Bf4! (Rybka) Nh5 12.Be3 Bb4+ 13.Kf1 O-O 14.Be2 a6 15.a3 Be7 16.Na7! Nf6 17.Rc1 Bd7 18.Nxd7 Nxd7 19.Rc7 Nf6 20.Nc8! and White wins. For more details, see pages 204-07 and 219-21 of Mayhem in the Morra!

Typical Smith-Morra craziness

Black's 5...Bg4!? is an unusual move not mentioned in the Smith-Morra Bible (Marc Esserman's Mayhem in the Morra!). It may be a decent move, possibly transposing to the important line 4...d6 5.Nf3 Nc6 6.Bc4 a6! 7.O-O Nf6 8.Qe2 Bg4!, which has scored extremely well for Black. If 5...Bg4!? is any good, 5.Bc4 (rather than my 5.Nf3) might be a more accurate move order. The game later descended into typical Smith-Morra mayhem. Yes, Virginia, 20.Bd5+ was more accurate than the needless piece sacrifice 20.Bb5+?

Back to Siberia

Many players, faced with the Smith-Morra Gambit (1.e4 c5 2.d4 cxd4 3.c3), try for a quick win with the Siberian Trap. A successful example of the trap is seen in the below game. After 9.h3?? Nd4! White gets mated after 10.Nxd4? Qh2# or 10.Qd3? Nxf3+ 11.Qxf3 Qh2#, or loses the queen for two knights after 10.hxg4 Nxe2+ 11.Bxe2.

Sicilian players like me love to ridicule the Smith-Morra. When IM Mario Campos-Lopez played the French Defense against Ken Smith at San Antonio 1972, Bent Larsen gave 1...e6 a question mark, noting that 1...c5 wins a pawn against Smith. Smith played his beloved gambit thrice in that tournament, losing to GMs Larry Evans and Henrique Mecking, and IM Donald Byrne.

However, IM Marc Esserman in his brilliant and entertaining new book Mayhem in the Morra! makes a strong case that the gambit is sound! He also plays it himself with great results, including a crushing win against GM Loek Van Wely.

Against the Siberian Trap, he extensively analyzes 8.Nb5! (improving on 8.Qe2?!) Qb8 9.e5! After 9...Nxe5?!, he analyzes 10.Nxe5 Qxe5 11.Re1! to a forced win for White. One line is 11...Qb8 12.Qd4! d6 13.Bf4! e5 14.Rxe5+! dxe5 (14...Be7 15.Rxe7+!; 14...Be6 15.Bxe6! dxe5 16.Bxe5 Qd8 17.Nc7+) 15.Bxe5 and Black's queen is trapped. After the alternative 9...Ng4 10.Bf4 a6 (he also analyzes 10...Ngxe5? 11.Nxe5 Nxe5 12.Bxe6! dxe6 (12...fxe6 13.Qh5+) 13.Rc1!) 11.Nd6+ Bxd6 12.exd6 b5 13.Bd5!! is Esserman's TN, which he analyzes as leading to at least a large advantage for White. The main line is acceptance of the sacrifice with 13...exd5 14.Re1+, which Esserman analyzes to a forced win.

I played 13.Bd5!! in two blitz games today, but both of my opponents declined the bishop. One game went 13...Bb7 14.Bxc6! Bxc6 15.Nd4! (another line given by Esserman), when Black could already resign in light of 15...Nf6 16.Nxc6 dxc6? 17.d7+ winning the queen. Here is the other game. Note that 15...Qb6!, leaving Black "only" a piece down, would have been a slight improvement.

In case you haven't gotten the idea, I highly recommend Esserman's book for anyone who plays 1.e4 as White or the Sicilian as Black, or just wants to play over some great tactical games.

Midwest Class this weekend in Wheeling

Don't miss this Illinois Tour event!  Details on the ICA website.  The prize fund is guaranteed to be $16,000, and if the 250 paid entry quota is hit, the prize fund will be $20,000. Not bad for a weekend event!

I won't be there to play (I make my living as a tax accountant, and the procrastinator's deadline is October 15th), but hope to stop by on Sunday to kibitz and see friends.

il-chess.org is back up

Link here. Thanks to Tom Sprandel for reloading the site and upgrading Joomla!

Out of the frying pan . . .

In the following game, my 12.Ng5! made it surprisingly hard for my opponent to defend his f-pawn. He did find a way, it's true - but it allowed me to pull off Boden's Mate. Instead of the ugly 13.b3, why didn't I play 13.Be2! with the same idea? You got me.

When the pin doesn't pin

In the following game, Black forgets that a pin against the queen is only a relative pin. Thus, unlike an absolute pin (a pin against the king), the pinned player can ignore the pin if it is advantageous to do so. Compare the Légal Trap and Imbaud-Strumilo, corr. 1922.