POISONED PAWN

[Annotator "Tyson Mordue"]

[WhiteTeam "Braille Chess Association"] Mordue AT 2292
[BlackTeam "KJCA Kings"]Jones,Victor 1991 Hinckley, Englang 2008
1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Bg5 e6 7. f4 Qb6
{The famous ‘Poisoned Pawn’ line. Black forsakes his development to grab a Pawn with his Queen. The main line is 7... Be7, while 7... Qc7, 7... Nbd7 and 7... b5 [the Polugayevsky variation] are all viable alternatives. After the text White can go 8. Nb3, but it really is much better to play a gambit.}
8. Qd2 Qxb2 9. Nb3
{In the famed Spassky-Fischer match Spassky played 9. Nb3 in the 7th and 11th games. In the 7th he was lucky to escape with a draw after some untypically sloppy technique by Fischer in an endgame where the American had an extra Pawn. However, Spassky’s powerful approach in the 11th game brought him a crushing victory in 31 moves, although Fischer could have resigned on move 24 with his Queen trapped and a Rook and Knight en prise as well. This game put 9. Nb3 well and truly on the theoretical map. I was so impressed by that game that I’ve never played the previously popular 9. Rb1. Many years later I was doing some preparation in anticipation of a Poisoned Pawn and reading through John Nunn’s book on the Sicilian Najdorf when I got a pleasant surprise - one of my own games had been quoted! One of the characteristics of the 9. Nb3 line is that there are tactical chances to trap the Black Queen. Right now White threatens 10. a3 and 11. Ra2, hence Black's reply.}
9... Qa3 10. Bxf6 gxf6 11. Be2 h5
{This move prevents Be2-Bh5 and was Fischer’s choice in the 1972 match. Black can instead choose to play -Bg7, -0-0, -Kh8, and -Rg8, while White goes Bh5 and f4-f5 to pressure the light squares. This actually happened in Mordue-Shapland, Hereford 2008 [I won in 29 moves]}
12. O-O Nd7
{The text was new to me and only features in four games in my database. I played the usual continuation of tucking the King into the corner before committing my pieces.
Spassky-Fischer 11th match-game 1972 continued
12... Nc6 13. Kh1 Bd7 14. Nb1 Qb4 15. Qe3 d5 16. exd5 Ne7 17. c4 Nf5 18. Qd3 h4 19. Bg4 Nd6 20. N1d2 f5 21. a3 Qb6 22. c5 Qb5 23. Qc3 fxg4 24. a4 h3 25. axb5 hxg2+ 26. Kxg2 Rh3 27. Qf6 Nf5 28. c6 Bc8 29. dxe6 fxe6 30. Rfe1 Be7 31. Rxe6}
13. Kh1 Be7 14. Bf3
{Clearing the E-file and pointing the Bishop at Black’s Queenside. My opponent now emulated Kasparov's Pawn advance and I decided that I didn't want the thorn in my side of -h4-h3. The weakness on g3 doesn’t seem important, as a black Knight is hardly likely to arrive there in the near future. Alternatives are 14. Nb1 and 14. Qd4, but both sides are keeping their cards close to their chests here and avoiding really committal moves.}
14... h4 15. h3 Nc5 16. Rad1
{This move is a partial bluff. Ideally I’d like the Rook on the C-file, but I have to wait for Black to exchange Knights on b3. Whether he wants to is a moot point, but I thought that by playing Rad1 it might put Black off playing the natural -Bd7. It did! Victor now settled in for a long think while I settled in a comfortable chair at the other end of the room. My opponent was clearly suffering from a cold and I didn’t want to catch anything - other than his King or his Queen, that is.}
16... Rb8
{Victor discovers that 16... Bd7 17. e5 is not appealing. With Black’s King in the centre any line-opening sequence is to White’s advantage. If 17... dxe5 18. Nxc5 wins a piece. Better is 17... Nxb3 18. cxb3 d5 19. exf6 Bb4 [not 19... Bxf6 20. Bxd5 and Black could be in trouble]. He decides to get his Rook off the h1-a8 diagonal and venture -b7-b5, but now castling Queenside is no longer possible. Meanwhile White continues to build up in the centre.}
17. Rfe1 b5 18. e5
{For better or worse White forces Black to define the central structure. Both of us thought that 18... fxe5 19. fxe5 d5 was asking for a sacrifice on d5, Victor suggesting 20. Nd4 first. Hence his choice to open the D-file in the belief that with his two Bishops controlling d6, d7 and d8 he should be safe.}
18... dxe5 19. fxe5 Nxb3
{Exchanging Knights before White gets the chance to play Nb3-d4 and into the weak square on c6. Around here White deliberately avoids Bc6+, because it would just present Black with a tempo to centralise his Queen with -Qc5.}
20. cxb3 f5
{Black has the two Bishops, but opening lines with 20. -fxe5 is just playing into White’s hands because of his superior development. With the text he hopes to shuffle his King to g7 and some sort of shelter. At this point I determined that the Black Bishop on e7 was the glue holding Black’s position together. There is only one White piece that could realistically hope to eliminate it, so my remaining steed sets off on a journey. The target square is c6.}
21. Ne2 Kf8
{This move threatens the skewer 22 -Bb4 now that Qd8 is no longer mate. A crucial alternative is 21... Qc5 bringing the Queen back into play. Fritz suggests the line
21... Qc5 22. Nd4 Bb7 23. Bxb7 Rxb7 24. Rc1 Qd5 25. Rc8+ Bd8 26. Rd1 0-0 27. Nxe6 fxe6 28. Rxd8
and White regains his Pawn. The resulting ending is probably drawn though.}
22. Nd4 Bb7
{This time 22... Bb4 is answered by 23. Qg5 Bxe1 24. Nxf5 and Black is crushed on the dark squares that his Bishop has just abandoned. An obvious improvement is 23... Be7 but after 24. Qe3 White has gained a useful tempo - here 24... Bc5 25. Nxf5 makes the point quite forcefully. The regrouping of the Queen doesn’t fare so well here. If 22... Qc5 23. Nc6 Rb6 24. b4 Qc4 [the only square!] 25. Nxe7 Kxe7 26. Qg5+ mates again.
If instead 22... Qc5 23. Nc6 Rb7 then 24. Nxe7 wins material, because 24... Rxe7 loses to 25. Rc1 and the loose Bishop on c8 falls. These lines demonstrate that Black has problems with the co-ordination of his pieces due to his lack of development and his misplaced Queen [a characteristic of the Poisoned Pawn line]. White is also dominating the open lines so Black must be careful. With the text Black prepares to take the Knight if it lands on c6. Naturally White is happy to exchange Bishops as his f3 Bishop is probably his least effective piece.}
23. Bxb7 Rxb7 24. Nc6 Kg7
{Once again 24... Qc5 to regroup the Queen is an important alternative. After 25. Rc1 Black has two choices, the endgame with 25... Qd5 and the middle game with 25... Qb6. After 25... Qd5 26. Qxd5 exd5 27. Nd4 [avoiding pins on the C-file] followed by 28. Rc6 White is doing very well and can expect to regain his Pawn minus soon. This is probably Black’s best line. Instead 24... Qc5 25. Rc1 Qb6 26. Nxe7 now works well as Black can’t recapture with the Queen. Taking with the King is a disaster after 27. Qg5+ so instead 26... Rxe7 27. Rc8+ Re8 28. R1c1 Qb7 29. Rd8 leaves Black short of moves. Fritz has the nice line 29... f4 30. Qxf4 Rxd8 31. Qf6 neatly emphasising the lack of co-ordination between Black’s pieces. White regains his Rook with a strong initiative.
Even worse is 28... Rxd8 29. Qxd8+ Kg7 30. Qf6+ Kh7 31. Rd1 [threat 32. Rd4 and 33. Rxh4+ mating] 31... Qa7 [the only square that defends f7 and controls d4] 32. Rd3 and there is no defence to 33. Qxh4+ and 34. Rg3+. Note how Black’s h8-Rook does nothing in any of these lines. Also the semi-open G-file is a big problem for Black. If his f5 Pawn was back on g6, Black would have a reasonable position. The immediate transfer of the King to g7 instead of bringing the offside Queen back into play meets with an intriguing riposte. Take a good look at this position.}
25. Re3
{This surprising move enhances White’s options tremendously. Victor took twenty minutes over his reply. This note shows why. He had a lot to think about. The main strategic point is that White threatens 26. b4 after which the black Queen is forced to an even worse square on a4. The immediate 25. b4 is answered by 25... Qg3 and now Black’s Queen assists in the defence of the Kingside. This is why I am refraining from Nxe7 while the Queen can recapture. White has good prospects on the Kingside as long as the black Queen isn’t helping out, see the variations in the last note. Also the Rook can go to d3 or c3 or possibly b3 [after b3-b4 Qa3-a4] to wall in the black Queen. It can even go to g3 in certain lines because the pin 25 -Bg5? is refuted by 26. Rg3 hxg3 27. Qxg5+ Kh7 28. Rd4 and the only move preventing 29. Rh4~ mate is 28... f4. However, White just needs to avoid 29. Rxf4 Qc1~ by the accurate 29. Qxf4 after which 29... Qf8 30. Qg5 Qh6 31. Rh4 wins. Note once again the collapse occurs after the elimination of the black Bishop, and how the black Rooks have no constructive role in the defence here. It is now too late to bring the Queen back. After 25... Qc5 26. Rc3 Qb6 27. Nxe7 Rxe7 28. Qg5+ Kf8 29. Rc8+ Re8 30. Rd7 clearly emphasises the disparity between attack and defence. Even agreeing to enter a disagreeable endgame with 26 -Qd5 is no good as White declines with 27. Qc1! Qe4 28. Nxe7, and you can guess the rest.}
25... Kg6
{Victor tries to force -Bg5, but after the following incarceration of the black Queen the white pieces rush to attack her exposed partner. Fritz suggests activating the Rooks with either 25... Rc7 or 25... Rc8 and even hints that 26. g4 might be the answer to the latter. This is certainly a powerful indication of White’s attacking potential and something we’ll come back to.}
26. b4 Qa4 27. g4
{The striking blow indicated in the last note. Open lines are imperative when attacking. White is almost certain of opening the G-file here. However, the position is far from simple. You may wish to make a separate note of the current position before starting on this long annotation. First of all the question of whether White can actually trap the Queen here has been raised. Here’s a line suggested by Fritz.
27. Rb3 Bg5 28. Qb2 Rc7 29. Rd6 R8c8 30. Ra3 Rxc6 and now both 31. Rxc6 Qd1+ and 31. Rxa4 Rxc1+ 32. Kh2 Bf4+ 33. g3 R8c2+ win for Black. This seems convincing enough. Secondly Fritz claims that 27. Nxe7+ Rxe7 28. g4 is a better move order, and it could be right. Certainly after 27. g4 f4 I’d have played 28. Nxe7+ and after 28... Rxe7 almost any sensible move of the e3 Rook (say d3 or b3) followed by 30. Qxf4 is very strong. The exclusion of the black Queen from the defence is the key factor.
Finally there is the intriguing question of what I would have done after Victor’s sensible post-mortem suggestion of 27... Bg5 28. exf5+ Kh7 It’s worth writing these notes just for this point alone. I had originally only looked at 28... exf5 29. Rg1 and the Black Bishop must fall soon. I’ve already said the position is far from simple. There are lots of possible moves on both sides.
Fritz claims 29. Rg1 Bxe3 30. Qxe3 Rg8 31. Rxg8 Kxg8 32. Qg5+ wins. This is based on a combination of getting the white Queen on a square where it rules out -Qd1+ and playing f5-f6. This is a very long calculation and a typical example of a computer program doing some number-crunching. My instinctive reaction was 29. Qd3 Bxe3 [I never entertained 29... Qxd1+ 30. Qxd1 Bxe3 and indeed after 31. fxe6 - threatening 32. Qd3+ picking up the Bishop - Black has no co-ordinated defence against Queen, Knight and two advanced Pawns] 30. f6+ Kg8. At this point I was certain the attack should crash through, but try as I might I couldn‘t produce a convincing line. The best I could find was 31. Ne7+ Rxe7 32. fxe7 Kg7 33. Qd8 forcing promotion, but now 33... Qxb4 and Fritz claims Black is even better. After 34. Qxh8+ Kxh8 35. e8=Q+ Kg7 Black threatens 36... Qe4+ amongst other things.
Time for another look. 27... Bg5 28. exf5+ Kh7 29. Qd3 Bxe3 30. f6+ Kg8 Look again. It’s worth it! 31. Qe2 is a very good try. Victor and I succeeded in convincing ourselves [and I later convinced GM Matthew Turner amongst others!] that 31... Rh6 was a good defence, and it is after 32. Qg4+ Rg6 33. Rd8+ Kh7 34. Qh5+ Bh6. In fact Fritz proves it does win after 31. Qe2 Rh6 32. Rg1+ Bxg1 33. Qg4 Rg6 34. Qh5+ Rh6 35. Qxf7+ Kh8 36. Qg7~ mate. Credit to you if you found this line in full.
However, the really beautiful line is the wonderful yet logical retreat 31. Qb1!! I’m certain I would have found this over the board if we had reached this position. Now Black is faced with the threat of 32. Rd8 mate because the Queen still controls h7. The black Rooks are almost irrelevant here. Note how the Knight on c6 prevents Rb8 as a defence in every line. The desperate 31... Qxd1+ 32. Qxd1 leaves Black facing 33. Ne7+ and 33. Qg4+ and there is no defence to both. The only apparent defence to 31. Qb1 is 31... Bb6. White’s constant problem here is getting an effective check on the G-file but now he can finally force it, and a bit more to boot! 32. Rd8+ Bxd8 33. Qg1+ Kf8 34. Qg7+ Ke8 35. Qxh8+ Kd7 36. Qxd8+ Kxc6 37. Qd6~ mate!
Back to the game after 27. g4! Black decided to capture directly. Taking en passant leads to annihilation after 28. Rxg3+ Kh7 29. Nxe7 Rxe7 30. R1g1.}
27... fxg4 28. Rg1
{Now Black doesn’t even have the desperate -Qxd1+. White’s major pieces converge on the hapless black King and there is now no parry. Both the thematic 28. Nxe7+ and 28. Re4 also win.}
28... g3
{If 28... Bg5 29. Rxg4 Rh5 30. Rf3 [threat 31. Rf6+] 30... f5 31. exf6 Kh6 32. Qd8 storms through.
Fritz adds 32... Qc2 33. Ne7 Qc1+ 34. Rg1 just in case you need convincing.}
29. Rexg3+ hxg3 30. Rxg3+ Kh5
{After 30... Kh7, White ladders his way in with 31. Qd3+ Kh6 32. Qe3+ Kh6 33. Qe4+ Kh6 34. Qf4+ Kh5 and now either 35. Qxf7+ or 35. Qg4+ mate soon. After the text White finally plays the thematic capture of the bishop, and Black has no way of ultimately stopping the mate on g5 except by giving up his stranded Queen.}
31. Nxe7 1-0