I really enjoyed annotating these three games and I hope you enjoy playing through them. Mozza plays the French a lot and as one of the worlds strongest players it's very interesting and illuminating to see how he handles it against a super-strong opponent. Lputian is a very strong GM and a French Defence expert so his play is also very instructive. The third game is totally different. It is one of my games from a tournament in 2004. It's only on this page because it does not fit anywhere else on the site. It is not of any great quality, but I still like it. :)

Anand, V 2774 - Morozevich, A 2732
Monte Carlo, 2004
[Alexander]

1.e4 e6 2.d4 d5 3.Nc3 Nf6 4.e5 Nfd7 5.f4 c5 6.Nf3 Nc6 7.Be3 a6

This is one of the easiest ways to play this position for black. The straightforward plan of advancing on the queenside and generating play there is very logical. Another point of immediately pursuing activity on the queenside is that it dissuades white from queenside castling, which is his usual plan in these kinds of positions.

8.Qd2 b5 9.h4 b4

9. ... Bb7 is more normal.

10.Ne2 Be7

Another idea for black would be to try to activiate his light squared bishop by playing 10. ... a5 and 11. ... Ba6.

11.dxc5 Nxc5 12.Ng3 h5

12...b3!?? is a very interesting idea, offering a pawn for white's dark squared bishop.

a) 13.cxb3? Nxb3! threatening Bb4 winning white's queen, 14.Qc2 Nxa1 15.Qxc6+ Bd7-+

b) 13.Bxc5 Bxc5 14.axb3 (14.cxb3? Bb4-+) 14...Qb6 15.Bd3 Be3 16.Qe2 Bxf4 17.Nh5 Bh6 18.Nh2 Threatening Ng4, is annoying for black, and leaves white somewhat better. I suspect that there may be a couple of improvements for black in there somewhere and that the move is therefore playable, perhaps he would do better not to try to get the pawn back straight away.

13.Be2 g6

The idea behind 12. ... h5 and 13. ... g6 is of course to delay white's natural pawn break of f4-f5. However, the moves do leave black sensitive on the dark squares.

14.0-0 Na4 15.c3 bxc3 16.bxc3 Qa5

16...Bxh4!? was possible as after 17.Nxh4 Qxh4 18.Rf3 Qd8! 19.Rb1 Qa5!= black's extra pawn compensates him for his dark square weaknesses and bad bishop.

17.Rac1 Bd7 18.Rc2 Na7?

This is a bit too slow. Better would have been 18...Rb8.

19.Ng5! Nb5 20.f5! gxf5 21.Nxf7!

21.Nxf5 exf5 22.Qxd5 Bxg5 23.Qxa8+ Bd8 is only equal as black has two pieces for the rook and pawn and white's attack doesn't look very dangerous.

21...d4?!

Not 21...Kxf7? because of 22.Nxf5! Raf8 23.Nxe7+ Kxe7 24.Bg5+ (24.Bc5+ is another way to win as 24...Nxc5 loses to 25.Qg5+ Ke8 26.Bxh5+ Rxh5 27.Rxf8+ Kxf8 28.Rf2+ Ke8 29.Qg8+ Ke7 30.Qf8#) 24...Ke8 25.Rxf8+ Kxf8 26.Qf4+ Kg8 27.Be7! Qb6+ 28.Kh1 Qg1+ 29.Kxg1 Nd6 30.Qg5+ Kh7 31.Qxh5+ Kg7 32.Bf6+ Kg8 33.Qxh8+ Kf7 34.Bh5#

Black might have considered 21...0-0 which is the sort of move you have nightmares about!

a) 22.Ng5 Na3 23.Rcc1 Nb6 threatening to come in to c4 looks almost plausible for black.

b) 22.Nh6+ Kh7 23.Nxh5 and white's attack continues. It's difficult to say whether black will survive, but he might.

c) 22.Nxf5 exf5 23.Qxd5 Nb6 24.Qb3 Qa4 (24...Rxf7? 25.Bxh5 Raf8 26.Bxf7+ Rxf7 27.e6+-) 25.Bxb6 Qxb3 26.axb3 Kxf7 (26...Rxf7? 27.Bc4) 27.c4 is about equal.

22.Nxf5! exf5

The rook on h8 is going nowhere, especially not to f8: 22...Rf8 23.Ng7#

23.Nxh8 dxe3 24.Qd5

24.Bxh5+?! Kd8 25.Qd3 Kc7=

24...0-0-0 25.c4 Kb8

Necessary, in order to avoid the discovered check after cb e.g. 25...Nb6 26.cxb5+ or; 25...Nac3 26.cxb5 pinning the c3 knight.

26.e6

26.cxb5 Was the most accurate continuation; 26...Bxb5 27.Qe6 Bc5 (27...Bxh4 28.Nf7 Bf2+ 29.Rxf2 exf2+ 30.Kxf2 Qb4 31.Nxd8 Qh4+ 32.g3 Qd4+ 33.Kg2 Qe4+ 34.Kg1 Qxc2 35.Bxb5 Qc5+ 36.Kh2 Qxb5 37.Kh3! should still be winning for white.) 28.Nf7 Rd2 29.Nd6 Qd8 30.Nxb5 is very complicated but probably winning for white.

26...Nac3 27.Rxc3?!

27.Qxf5!? is better, but very complicated. 27...Bxe6 28.Qxe6 Nd4 and now;

a) 29.Qxe7 Ncxe2+ 30.Kh2 (30.Rxe2 Nxe2+ 31.Kh2 Qc7+=) 30...Qc7+=;

b) 29.Rb2+ Bb4 30.Qxe3 Ncxe2+ 31.Kh1 Rxh8 32.Rxb4+! Qxb4 33.Qe5+ picks up the rook on h8.

c) 29.Qxe3 Nxc2 30.Qxe7 Nxe2+ (30...Rxh8 31.Rf7! Nxe2+ 32.Kh2 Qb6 33.c5! Qc6 34.Qe5+ Ka8 35.Qxh8+ Qc8 36.Qxc8#) 31.Qxe2 Qc5+ 32.Kh1 Qd4 33.Ng6+-

27...Bc8

27...Bxe6 may be even better as after 28.Qxe6 Nxc3 29.Qxe7 Nxe2+ 30.Kh2 Qc7+ 31.Qxc7+ Kxc7 black's e-pawn looks very dangerous.

28.Qxf5 Qxc3 29.cxb5 Rxh8 30.bxa6 Rf8?

30...Ka7 would have been better.

31.Rb1+?

31.Qb1+! Ka7 32.Rc1 Qb4 33.Qc2 Bxa6 34.Bxa6 Kxa6 35.Rb1!+- as black's queen has no safe square form which to be able to defend against the threats of 36.Qa4# and 36.Qc6 followed by 37.Qb7#

31...Ka7 32.Qe4 Qc7

32...Bxa6?? 33.Qa4 Qc8 34.Bxa6 Qxa6 35.Qd7+ Ka8 36.Qxe7 Rc8 37.Qb4! e2 38.e7+- with the threat of 39. e4+ and 40.e8=Q+-

33.Qd4+ Bc5 34.Rc1

34.Qd5 might have been better.

34...Rf5 35.Qc4 Re5 36.Qb3 Rxe6 37.a4 Re5 38.Qb5 Qb6 39.Rb1 Qxb5! 40.axb5 Bg4 41.Kf1 Rf5+ 42.Bf3

42.Ke1 Bxe2 43.Kxe2 Rf2+ 44.Kd3 Rd2+ 45.Kc4 e2 46.Kxc5 Rd1 and black will queen on e1.

42...Bxf3 43.gxf3 Rxf3+ 44.Ke2 Rf2+

0-1

I thought I'd wait 'till the end to tell you that this was a blindfold game!

Bologan, V 2585 - Lputian, S 2590
1994
[Alexander]

1.e4 e6 2.d4 d5 3.Nc3 Nf6 4.e5 Nfd7 5.Nce2

This is a very sneaky system. This second move with the queen's knight inaugurates the Shirov/Anand variation. White's idea is to bolster his centre with his c pawn achiveing a position with similarities to the advance variation of the French (1. e4 e6, 2. d4. d5, 3. e5). The differences are that white's queens knight is on e2 instead of b1 and that black's king's knight is on d7 instead of g8. Arguably, these differences favour neither player. The knight on e2 prevents the natural development of the f1 bishop while the knight on d7 lacks active scope.

5. ... c5 6.c3 b6

This is unusual. 6. ... Nc6 is more normal. Interesting is 6. ... cd, 7. cd f6. Also possible is 6. ... Qa5. With 6. ... b6 Lputian tries to find some immediate scope for his light squared bishop.

7.f4 Ba6 8.Nf3 Nc6 9.g4

This is an ambitious try. The machine offers 9.f5 exf5 10.Nf4 Bxf1 11.Rxf1 cxd4 12.cxd4 f6 13.Nxd5 Nb4 14.Nxb4 Bxb4+ 15.Kf2 fxe5 16.Nxe5 Nxe5 17.dxe5 Bc5+ 18.Kg3 0-0= as an alternative. This is a very interesting piece of analysis from the machine but unfortunately it's assessment is incorrect. White's e-pawn is a weakness, and white's king is exposed: 19.Qxd8 Raxd8 20.Bf4 Kf7 (20...h6 21.h4 Kf7 22.Rad1 Ke6=+) 21.Rad1 Ke6=+

9...h5

After a more routine looking move like 9...Rc8 white might try the immediate 10.f5 or prepare this break with 10.Ng3. the immediate 10. f5 does not seem to be too dangerous for black but the prepared break might be a good practical try.
A) 10.f5 Be7 11.fxe6 fxe6 12.g5 (12.Nf4 Bh4+ 13.Kd2 Bxf1 14.Nxe6 Qe7 15.Nxh4 Qxh4 16.Rxf1 Qxh2+-+) 12...Bxe2 13.Bxe2 0-0=
B) 10.Ng3 Bxf1 11.Kxf1 cxd4 12.cxd4 Be7 13.Kg2 0-0 14.f5 looks good for white though there is no obvious way to break through immediatly. After 14. ... Nb4 black has some chances on the queenside but white's progress on the kingside is more impressive.

10.gxh5 Rxh5 11.Ng3 Bxf1

Obviously, a rook move would have left the bishop on a6 en prize.

12.Kxf1

Taking the proffered rook would also have been interesting: 12.Nxh5!? Bg2 13.Rg1 Bxf3 14.Qxf3 Qh4+
A) 15.Kd1 sets a trap
A1) 15...Qxh2? 16.Rh1! winning black's queen.
Therefore better is
A2) 15...g6 16.Ng3 cxd4= after which black probably has enough compensation for the exchange.
B) 15.Ng3 cxd4 16.c4! Nb4 17.cxd5
B1) 17...Nc2+ is inadequate because of 18.Kf2 Nxa1 19.dxe6 Rc8 20.exd7+ Kxd7 21.Qb7+ Rc7 22.Qd5++=
B2) 17. ... Nxd5 is better: 18.Bd2 g5=

12...Rh8 13.f5 Be7 14.fxe6 fxe6 15.Qd3 Nf8 16.Bd2 Qc7 17.Kg2 0-0-0 18.b4 cxb4

18...c4 looks safer. Lputian however wants to use the c4 square for his knight.

19.Rhc1 Na5 20.cxb4 Nc4

This knight is exceedinlgy well placed here. It is unlikely that it will ever be worth black's while to exchange it for the bishop on d2.

21.Nf1 Kb8

It makes sense to take the king off the open c-file.

22.b5

22.Ne3 looks like a more logical follow up to 21. Nf1.
A) 22...b5? 23.a4 a6 24.axb5 axb5 25.Ra6 Qb7 26.Rca1+- for example: 26...g5 27.Nxc4 dxc4 28.Qa3 g4 29.Qa5! gxf3+ 30.Kf1
A1) 30...Rxh2 31.Ra8+ Qxa8 32.Qxa8+ Kc7 33.Ra7+ Kb6 34.Qb7#
A2)30. ... Kc8 31.Ra8+
A21) 31...Kd7 32.Ra7 Kc8 (32...Kc6 33.Rxb7 Kxb7 34.Qxb5+ Kc8 (34...Kc7 35.Ra7+ Kc8 36.Qb7#) 35.Qc6+ Kb8 36.Ra8#;
A22) 31. ... Qxa8 32.Qxa8+ Kd7 33.Ra7+ Ke8 34.Rxe7+ Kxe7 35.Bg5+ Kf7 36.Qxd8+-
B) 22...Nxe3+ 23.Bxe3 Qd7 24.b5+=

22...Qd7 23.a4 Qe8 24.Ng3

This looks like a loss of time but white understandably wants to avoid an exchange of queens after 24. ... Qg6+ in order to continue with his attack on the queenside. The exchange of queens would have left the position looking fairly equal: 24.a5 Qg6+ 25.Qxg6 Nxg6 26.axb6 axb6 27.Ra2 Kb7 28.Rca1 Ra8 29.Rxa8 Rxa8 30.Rxa8 Kxa8=

24...Qg6 25.Qe2

If white was going to exchange queens he should/would have done so before playing Ng3.

25...Rc8 26.a5 bxa5

26...Qg4 was also possible but I think black was wise to keep the a-file half closed: 27.axb6 axb6 28.Ra6 Ng6 29.Rca1
A) 29...Nxd2? 30.Qxd2 Nf4+ 31.Kf2 Nh3+ 32.Ke1 Qxf3 33.Ra8+ Kc7 34.Qc2+ Bc5 (34...Kd7 35.R1a7+ Ke8 36.Rxc8+ Kf7 37.Rxe7+ Kxe7 38.Qc7#) 35.dxc5 Qe3+ 36.Ne2 Qxc5 37.R1a7+ Kd8 38.Qxc5 bxc5 39.b6 Rxa8 40.Rxa8+ Kd7 41.Rxh8 Kc6 42.Rxh3+-
B) 29. ... Kc7 30.Ra7+ must be slightly better for white.

27.Bxa5 Qg4 28.Nd2

28.Qa2 might have been a better try for white, but he would still be slightly worse after 28...Ng6 29.Bd2 Na3!=+ using this wonderful knight to temporarily block the a-file.

28...Qxd4 29.Nxc4 dxc4

29...Ng6! might be even stronger.

30.b6 Qd5+

30...axb6? 31.Bxb6! Qd5+ (31...Qxb6 32.Rab1) 32.Kg1 Bc5+ 33.Bxc5 Rxc5 34.Rd1+- Qxe5 35.Qa2! Qe3+ 36.Kg2 Rd5 37.Rab1+ Kc7 38.Rxd5 exd5 39.Qa5+
A) 39...Kd7 40.Qxd5+ Ke8 41.Rb8+ Ke7 42.Rb7+ Ke8 43.Qf7+ Kd8 44.Rb8#
B) 39...Kd6 40.Nf5+ wins black's queen.
C) 39...Kc8 40.Qa6+ Kd8 (40...Kd7 41.Rb7+ Ke8 42.Qa8#) 41.Rb8+ Kd7 (41...Ke7 42.Nf5+ not only wins black's queen but mates in a few moves: 42...Kf7 43.Rb7+ Nd7 44.Rxd7+ Kg8 45.Rxg7+ Kf8 46.Qd6+ Ke8 47.Qb8#) 42.Rb7+ Ke8 43.Qa8#

31.Qe4 a6 32.Bc3 Kb7 33.Qxd5+

33.Rf1 looks like a reasonable alternative though black is still clearly better after 33...Qxe4+ 34.Nxe4 Ng6 35.Rf7 Kxb6 36.Rxg7 (36.Bd4+ Kc6) 36...Nh4+ 37.Kh1 Nf5 38.Rg6 Rc6 39.Rb1+ Ka7 as he is simply a pawn up.

33...exd5-+ 34.Rd1

34.Rf1 again looks plausible but of course, black is still clearly better after 34...Bc5 35.Rf7+ Kxb6 36.Rxg7 (36.Ba5+ doesn't really achieve anything 36...Kc6 37.Nf5 d4 38.Nxg7 Kd5 39.Re1 Ne6 40.Nxe6 Kxe6 41.Rf6+ Kd5 42.Rxa6 Rcg8+ 43.Kh1 Ra8 44.Rxa8 Rxa8 45.Bd2 Ra2 46.Bf4 d3 47.e6 Be7-+ when black's pawns look a lot more dangerous and likely to queen than white's.) 36...d4-+ when black's connected passed pawns look formidable.

34...Ne6 35.Ne2

35.Nf5 Bc5 36.Nd6+ Bxd6 37.exd6 Rc6 and the d-pawn is taboo because after 38.Rxd5 Nf4+ picks up the rook on d5.

35...Rcd8 36.Ra5

It's too late for 36.Rf1 as it can now be met by Rdf8-+

36...Bc5 37.Nd4 Bxd4 38.Bxd4 Rh4 39.Bf2 Rg4+ 40.Kh3?

40.Kf1 was better but still losing as after 40...d4 the black c and d-pawns look very dangerous.

40...Rg5 41.Be3

41.Rd4 holds out for a few more moves but still white is mated after 41...Rh8+ 42.Bh4 Nxd4 43.Rxd5 Nf3 44.Rd7+ Kxb6 45.Rd6+ Kc5 46.Rd5+ Kxd5 47.e6 Rxh4#

41...Rh8#

0-1


Evans-Quek, S 2257 - Alexander 2099
Oxford, 2004
[Alexander]

1. d4 Nf6 2. c4 e6 3. Nf3 Bb4+ 4. Nbd2 b6 5. a3 Bxd2+ 6. Bxd2 Bb7

It looks like White's ambitious approach has been successful as he has bagged the Bishop-pair and appears not to have had to make any concessions in return. Although Black's position is compact, and his pieces can find useful squares easily, White must be slightly better here.

7. Bg5

Developing the bishop outside the pawn chain makes sense but it can become a target here.

7... d6 8. e3 Nbd7 9. Bd3 h6 10. Bh4 g5 11. Bg3 h5

A very aggressive move.

12. h3

This is a bit too cautious, but is an understandable reaction to Black's pawn advances. 12 Nxg5 would be dubious on account of 12 ... h4, 13 Bf4 e5, trapping the Bishop. After 14 0-0 Qe7, I would be able to get my King out of the centre before I captured the Bishop and I would be clearly better. 12 h4 is the main line but I would be happy with my extra pawn after 12... g4 13 Ng5 Bxg2 14 Rg1 Bb7.

12... Ne4

Taking away White's Bishop-pair as 13 Bh2 g4 is unappetising for White.

13. Bxe4 Bxe4 14. Kf1

This artificial looking move is actually quite good. White doesn't want to castle as removing the rook from the h-file allows ...g4!

14... Qf6

I wanted to play g4 and this move (Qf6) seemed to me to prepare it quite well. Putting my Queen on f6 protects the Rook on h8, puts my Queen on the same file as my opponent's King, and prepares Queenside castling so that my other Rook can come over to point at my opponent's King too. 14... Qe7 and 14...f5 were the other two moves I considered but the former didn't seem to fulfil so many useful purposes and the latter looked like it was blocking rather than opening lines for attack.

15. Nd2 Bb7

I was very tempted to play 15. ... Bd3+ "lodging my Bishop in the heart of my opponent's position" but after 16 Kg1 O-O-O 17 Rc1 Rdg8 18. Rc3 I felt that whichever square I put the Bishop on it would only be getting in the way. Also, I was worried about discoveries from the Knight as well as the threat of blocking my Bishop in with some sequence such as 17 f3, 18 Bf2, and 19 e4. I decided that on b7 my Bishop would still be pointing in the right direction while fulfilling light square defensive duties on the Queenside in readiness for when I put my king there.

16. h4

This forces me to make a decision about my pawns.

16... gxh4

16. ... g4 seemed to me to be blocking all the lines I wanted to make use of. Perhaps I should have played 16... 0-0-0, as, if 17 hxg5 Qxg5, White cannot play 18 Bh4 because of the threat to his g2 pawn. However, I felt that at some point I would have to take the pawn on h4 to open up the g file for my pieces, so why not?

17. Rxh4

17 Bxh4 is probably good too, although I would still have been happy with the way things were going after 17. .. Qg6 18 f3 Nf6, after which I could castle Queenside and play ...Rdg8 with some pressure.

17... O-O-O 18. Nf3

Of course not 18 Rxh5 Rxh5 19 Qxh5 Rh8, as after White moves his Queen, 20...Rh1+ will pick up the Rook on a1. 20. Qg4 doesn't help as the check on g8 is meaningless; Black can simply play ...Qd8.

18... Rdg8 19. b4 Nf8

With White having begun his operations on the Queenside I felt I needed to get things rolling on the other side of the board. The only way I could see of making progress there was to push my h-pawn and so I needed to dislodge the Rook from h4. This knight manoeuvre Nd7-f8-g6 was the only way I could see of doing this.

20. d5

This is intended to block my Bishop on b7 in and prevent it from taking part in the attack. I assumed that White also wanted to swing his rook to the queenside via d4 and d2 to c2 to be quickly followed by c5. However, this is quite a slow plan and I was confident that my threats on the Kingside would make an impact first. As we shall see White has different plans for this Rook.

20... Ng6 21. Rd4

It might be better for White to keep his Rook on the h-file to help to waylay Black's attack.

21... e5 22. Re4?

I hadn't even considered this move. I am very sure that this is the wrong square for the Rook! The fact that White felt this was better than 22 Rd2 is another good reason for leaving the rook on the h-file.

22... h4

I spent a long time on this move. My immediate thought was 22... Kb8 to be quickly followed by 23... Bc8 and 24... Bf5, but of course by then White will have played c5 so that he can escape to c4 with his Rook. Then I hit on the idea of ...Qg7 followed by f5, which also traps the Rook. I calculated 22.. . Qg7 23 c5 f5 24 Rc4 Ba6 25 Qb3 (threatening b5) 25...Bxc4+ 26 Qxc4 and in this position I saw that White was threatening 26... Qa6+ and if 26... Kb8 then 27 c6 wins or if 26... Kd8 then 27 Qxa7, threatens 28 Qb8+ Kd7, 29 c6+ Ke7 30 Qxc7+ which I evaluated as either unclear or bad for me. I was unsure whether I could find ideas on moves 26 and 29 to hold off White's Queen or to generate threats of my own on the kingside. I decided that things would be different if I had an extra tempo on the kingside and so I turned to analysing the possibility of playing h4 first. I calculated 22... h4 23 Bh2 Qg7 24 c5 f5 25 Rc4 Ba6 26 Qb3 Bxc4 27 Qxc4 and felt that here, the fact that moving the Knight somewhere opens an attack on White's g-pawn meant that I would have enough time to defend on the Queenside. Although I didn't see it, it is clear that after 27...Ne7 28 Qa6+ Kd8 29 Qxa7, the move 29...Nc8 would cover everything on the Queenside while I could continue my attack on the Kingside. I evaluated the position after moving my Knight off the g-file as good for Black. At this point I noticed that after 22... h4 23 Bh2 Qg7, White could play 24 Rg4 and after 24... f5, 25 Rg5. I decided however that after 25...Qf6, White had no good way of meeting the threat of e4. Objectively, I think that it should not be too difficult to find sufficient resources on moves 26 and 29 in the variation without 22. ... h4. The way I played was safer, and therefore, against a stronger opponent, was arguably a good practical decision.

23. Bh2 Qg7 24. Rg4 f5 25. Rg5 Qf6 26. c5 e4

All according to plan!

27. Qb1

This is a good move in a difficult position. It dissuades ideas involving ...Ba6 because of the threat of b5, and after 27...exf3 White can play 28 Rxf5 escaping. 27 Qd4 was also possible, but after 27...Qd8 both the Knight and the Rook are still threatened.

27... 27... exf3 28. Rxf5 Ba6+?!

I played this move after "seeing the bright lights". I saw 29 Ke1 Qc3+ 30 Kd1 Be2 checkmate, 29 Kg1 Nf4 30 Rxf6 Nh3+, 31 Kh1 fxg2 checkmate, and 29 Kg1 Qg7 30 Rxf3 Ne5 winning. I also saw 29 Kg1 Qc3, threatening 30... Bd3 which looked pretty good. The fact that White's Queen is tied to the defence of both Rooks and that my pieces appear to be ideally placed to deliver a killer blow convinced me that this move was justified. Perhaps this move is not quite so bad but I am sure there is a better move 'out there'.

29. Kg1

Unfortunately it is all an illusion! 29... Nf4 is met by 30 Rxf4, 29... Qg7 is met by 30 Rg5, and 29... Qc3 is met by 30 Rxf3. Instead of 28... Ba6+, 28...fxg2+ would leave Black a clear piece up and although his attack is over he can, after... h3 devote all his resources to defence and exchange.

29... Qc3

In the post-mortem my opponent said he thought that this move was okay but I am not convinced. The move 29. ... Qc3 gives up my f-pawn which might be useful for my attack! I am sure that 29... Qg7 is better as after 30 Rg5, which seems to defend everything, Black can play 30...Bd3! to be followed by ...Qf6 when White's Rook has no place to run to.

30. Rxf3 Bd3?!

Probably not very good but I could still see some dim and distant lights. 30...Qg7 is still the best move with the threat of 31.. .Ne5 being very difficult to meet.

31. e4 Bxb1

A 'kibitzer' at the post-mortem pointed out that the move 31... Nf4! is a significant improvement. I agree. 32 Bxf4 allows 32.. . Qg7, threatening both 33...Bxb1 and 33...Qxg2 checkmate, and it must not be forgotten that White's Queen is still tied to the defence of the a1-Roook. 32 Qd1? also meets with 32... Qg7 when White's position is falling in around him. Whites best is probably 32 Rxd3 but after 32... Nxd3, White is still facing threats on the Kingside and is almost a clear Rook down.

32. Rxc3 Bxe4 33. cxd6 Bxd5 34. Rxc7+ Kd8

Of course not 34... Kb8 35 d7! which is good for White.

35. Rac1 h3 36. Rc8+ Kd7 37. R8c7+ Ke6 38. Re1+ Kf5 39. g4+ Kxg4 40. Rd1 Nf4

I think this is over optimistic. I was still expecting to find a killer blow but I do not think that this is the right path. Saying that however, Black still has a winning position.

41. Rd4 Kf3+

Both 41...Kh5+, which was suggested by Mr Sherwin, and 41...Kf5+ are better than this move, which is again over-optimistic. After 41... Kf5+ 42 Kf1 Bg2+ 43 Ke1, 43...Ne6, holds everything together and Black is still winning.

42. Kf1 Be4 43. Rc3+ Bd3+ 44. Ke1 Kg2?!

Both players are short of time and it shows! 44...Re8+ might be better. The position has now slipped through my fingers to equality.

45. Bxf4 Bf5 46. f3 h2 47. Rd2+ Kg1 48. Be3+

Instead of this Mr. Dickinson pointed out that 48 Ke2 was worth a shot. Fortunately, I had seen that in this position I needed to play 48... Kh1, as 48... h1=Q?? would lead to a win for White after 49 Rc1+! Kg2 50 Ke3+! Kh3 51. Rxh1 checkmate.

48... Kh1 49. Rf2 Rg1+ 50. Ke2 Rhg8 51. Rc1

Here my opponent offered me a draw with his flag hanging, and, being rather worried about the safety of my king, I accepted. Winning on time in this position would be meaningless anyway.

1/2-1/2

All games annotated by Alexander. © 2004.