An old chestnut, or rather two of them

From the diagrammed position, Black's fatal blunder 9...Ng4?? occurred 58 times in Mega Database 2013. The position arises from a number of different move orders, including the Sicilian Accelerated Dragon, Maroczy Bind, as below; the Symmetrical English (1.c4 c5 2.Nf3 Nf6 3.Nc3 Nc6 4.d4 cxd4 5.Nxd4 g6 6.e4 Bg7 7.Be3 d6 8.Be2 0-0 9.0-0); and the King's Indian (1.d4 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 0-0 5.Nf3 c5 6.Be2 cxd4 7.Nxd4 Nc6 8.Be3 d6 9.0-0).

I remembered seeing this trap in GM Larry Evans' 1970 book Chess Catechism (p. 72), or thought I had. It turned out that Evans was referring to a very similar trap in the Sicilian Dragon, where White's pawn is still on c2 and Black hasn't castled. Alekhine, in his famous book of the New York 1924 tournament (Lasker's immortal triumph, at age 55), annotating the game below, wrote of Réti's eighth move, "More cautious is 8.P-KR3, for now Black can play ...N-KN5." Evans' correspondent had dutifully played the recommended 8...N-KN5 (8...Ng4) in his own game, only to be rudely awakened by 9.Bxg4!, which works just as well as in the other version. This version of the trap is seen much less often (8...Ng4 was played five times in Mega Database 2013), and, oddly, the blunderer is much more successful! In the first version of the trap, White played 10.Bxg4! 56 out of 58 times. On the 52 occasions that Black played 10...Bxg4, White found the winning follow-up 11.Nxc6! 36 times. In the present version, incredibly, White found the winning 9.Bxg4! only once in five games, and only scored 30%! Vepkhvishvili-Wiedersich, Pardubice Open 1993 (not in Mega) is one of the rare games where White found 9.Bxg4!

Confusion in the Symmetrical English

The Symmetrical English (1.c4 c5) is an opening rich in possibilities for both players. After 2.Nf3 Nf6, two ways (among many) for White to play are 3.g3, aiming to get a reversed Maroczy Bind after 3...d5 (again, many other moves are possible: 3...b6, 3...g6, 3...Nc6, etc.) 4.cxd5 Nxd5 5.Bg2 Nc6 6.Nc3 Nc7 (6...e5? 7.Nxe5!) 7.O-O e5 8.d3, and 3.Nc3 d5 4.cxd5 Nxd5 5.e4 Nb4 6.Bc4, not fearing 6...Nd3+. However, combining g3 and e4 in the manner that White does in the game below is much weaker. After 3.g3 d5 4.cxd5 Nxd5 5.e4? Nb4! White already had a bad game. Kabat-Misicko, Czechia 2007 continued 6.d3 Bg4 7.Na3 N8c6 8.Be2 Bxf3 9.Bxf3 Qxd3, winning a pawn (0-1, 28). My opponent's 6.d4 cxd4 7.Nxd4?? was no improvement, dropping a piece to the well-known trick 7...Qxd4!, winning a piece (8.Qxd4 Nc2+ and 8...Nxd4). After 11...Bh3+ he resigned in light of 12.Kg1 Nf3# or 12.Ke1 Nc2+, winning the rook.