Searching for good examples of the Pirc Defence in the FICS database, I've stumbled across the following interesting 15-min game played in 2003 between a 2239 (White) and a 1912 (FICS ratings are inflated but I'm not sure if it applied back then).

As I almost haven't played for 12 years, I didn't rely on my own tactical vision and checked variations with the free Houdini 1.5a , but in such unclear positions engines tend to find repetition draws so the evaluation question remains open, and you're welcome to make analytical comments and train tactical skills - I've spoilered some critical moves, and there are quite a few sacs in variations!

The PGN can be downloaded at ficsgames (end position spoiler alert!), and the replayer is also duplicated at chessvideos (for faster diagram-making).

1. e4 d6 2. d4 Nf6 3. Nc3 g6 4. Nf3 Bg7 5. Be2 O-O 6. O-O Bg4 7. h3 Bxf3 8. Bxf3 Nc6 9. Be3 e5 10. d5 Ne7 11. Be2 Nd7 12. f3 f5 13. a4 f4 14. Bf2 g5 15. Nb5 a6 16. Na3 Ng6 17. c4 a5 (preventing b2-b4, but the knight returns to a comfortable square) 18. Nb5 Rf7 (what a strange plan to place the rook on g7) 19. Qd2 Bf8 20. Rfc1 h5 21. b4 axb4 22. Qxb4 Rg7 23. a5 Nf6 24. Na7 (with a discovered attack on the b7 pawn, 24. c5 = is a calm option) g4!?



What would you play here as White? What comes to mind is Qxb7 or hxg4.

After 25. Qxb7, Black can discover an attack on the queen and x-ray the knight by 25. ... c5, so it doesn't look like a sensible move.

However, White has a good reply:
So Black's best option is to chase the bishop off the protection of the knight by 25. ... g3 26. Be1 c5 27. a6 Rxa7 =+(?), getting an extra piece against the passed a-pawn in an unclear position (it looks like Black will slowly but surely untangle the constrained piece setup whilst maneuvering tediously around the a-pawn ).

Another sensible 26th move for White pointed out by the engine is
Instead, 25. hxg4 hxg4 was played. Now a good continuation is
But in the game there was 26. fxg4?? (this natural move surely doesn't win a pawn, but why is it a grave blunder?!) Nxe4 27. Qxb7?? (the most stubborn line is 27. Qe1 but even then a piece is still lost after

Nxf2 28. Nc6 (28. Kxf2 c5) Qh4! 29. Qxa8? (29. Ra2 Re8 30. Bf3 Nxg4 31. Bxg4 Qxg4 32. Qb2 f3 is more stubborn)



Now Black has a couple of ways to win fast. The actual move was
but an even more effective move is
GL with your chess!

This is a deceptive thread title as the position given is a Bayonet King's Indian Defence position, although it came from a Pirc initial move order.

I think it was a strategic error by White to allow Black this attack, with White having played h3 already.

As a general assessment, Black's attack only works if he has the light-squared bishop to break through on h3, after ...g3; h3. White's successfully removed this, so Black's attack should fail (and White will have a winning queen-side attack).

However, white has played h3 voluntarily, a mistake. Black's attack probably will work here. If hxg4, Black will come play hxg4, then come down the h-file (perhaps with Nh5-g3, or ..g3).

It will just be a matter of calculation as to who wins from here, a good computer should be able to see it through.