In addition, the opponent can always be alert to give one exchange back and kill the counterplay, while still keeping some extra material at hand.

In our example we have a setup that resembles pawn structure resulting from Nimzo-Indian or Ragozin defence. White’s position looks very appealing at the first glance. He has pair of Bishops and prepares to roll the central pawns with e3-e4.

Black h5-pawn also seems to be running ahead of the time and can become target for White pieces.

White’s next move looks natural, the Knight presses h5 and supports the advance e3-e4.

27. Ng3 Rxe3!
Black is not interested in passive defence and instead sacrifices an exchange to abruptly change the nature of the position. He doesn’t have to worry about e4 anymore, White c3-pawn suddenly becomes a target and White Queen is sidelined on h4.

28. Bxe3 Rxe3 29. Nxh5 Nxh5 30. Qxh5 Bc6

At the end of the forced variation Black is protecting d5-pawn. White’s positional advantage has disappeared and now he has to search for tactical possibilities to use the extra exchange.

But Black Knight provides excellent cover for the King and c3-pawn is also hanging. Allowing two connected passed pawns would be a bit too much.

However, it is now easy to protect the pawn. The natural 31. Ra3 weakens the back-rank and Black wins with 31…Re1+ 32. Kh2 Qc7+ 33. g3 (33. Qg3 Rh1+) Qe7! when Ra3 is hanging and Black threatens Qe2+.

White is trying to “trap” the Black Rook.

31. Qg5 Rxc3!
But Black doesn’t care, he wants to eliminate the obstacle on the way of his pawns.

32. Qd2 Rxc2 33. Rxc2 Ne6
The Rook was trapped on c3 and Black sacrificed another exchange. The benefits are obvious, Black has two connected and quickly advancing pawns, while d4 is also hanging.

White heavy pieces do not have space and files to attack the enemy King. Still, Black should always be wary not to allow this possibility.

34. Rd1 b4

35. Rb2 b3 36. Qc3 Nc7
Now Black is preparing to offer Queens exchange. As mentioned above, this would effectively deny any possibility of counterplay against Black King. With only two Rooks on the board White cannot stop the charging pawns because the Rooks are actually not suitable for defending tasks.

37. Re2 Qa7 38. Qb4
38. Ra1 is also met with 38…Nb5, Black is happy to trade the Queens.

38…Nb5! 39. Re7 Qa3 40. Qe1 c3 41. Re8+ Bxe8
Remember the introduction? We were warned of the option of giving one exchange back. White is hoping to finally come closer to the Black King.

42. Qxe8+ Kh7 43. Qxf7 Qa8!
Denies perpetual check. 43. Qxb5 was breaking against 43…c2.

44. Re1 Nd6
The last prophylactic measure, White was threatening Qh5+ and Re8+.

45. Qc7 c2 46. Qxd6 b2 47. Qf4 Qc6 to cover against Qf5+. White finally gave up 0-1

He defeated reigning French champion GM Maxim Vachier-Lagrave.

Excerpt from the Round 10 live commentary with GM Simon Williams, where GM Berg presented his victory. Bellow are the moves.

Emanuel Berg SWE 2550 – Maxime Vachier-Lagrave FRA 2699

1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.Bg5 e6 7.f4 h6 8.Bh4 Qb6 9.Qd3 Qxb2 10.Rb1 Qa3 11.f5 Be7 12.fxe6 fxe6 13.Be2 O-O 14.O-O Kh8 15.Kh1 Nbd7 16.Nxe6 Ne5 17.Nxf8 Nxd3 18.Ng6+ Kh7 19.Bxd3 Kxg6 20.e5+ Kh5 21.exf6 Bxf6 22.Bxf6 gxf6 23.Rb3 Qa5 24.Rxf6 Bd7 25.Ne4 Bg4 26.h3 Rg8 27.hxg4+ Rxg4 28.Be2 Qe1+ 29.Kh2 1-0

Obviously, the e3-pawn is located deep inside the enemy’s territory and the first instinct is to try and do something with it. But I kept looking at the diagram and couldn’t find a convenient way to continue, it appeared like white will cover up everything.

Grandmasters do look deeper into the subtleties of a chess game. Bosko Abramovic realised that Ba2 is a very unfortunate piece, that white king is in danger and discovered a hidden motive – back rank weakness!

24…e2! 25. Qxf3
It was necessary to deflect the white queen away from the d-rank, otherwise white would use it to deflect black’s attacking piece! The authors warn us about the correct move order, erroneous was immediate 24…Nb4? 25. axb4 e2 26. Rd8+! Rxd8 27. Qxf3, and white gets away with it since there is no more Rc1+, 27…Bd2 with unclear play (Volokitin/Grabinsky).

25…Nd4!
Another tempo earned by attacking the queen.

26. Rxd4 e1Q+

27. Qd1
The point of the combination is revealed with the variation 27. Rxe1 Qxe1+ 28. Rd1 (28. Bxe1 Rc1 checkmate) 28…Rc1+ 29. Rxc1 Qxc1 checkmate again. White is forced to concede material with the game continuation…

27…Qxd1+ 28. Rhd1 Bf6 29. Rc4 Rxc4 30. Bxc4 Qe4+ 31. Bd3 Qxg4 32. Re1 g5 33. Be4 Qd7 34. Bf2 Qd2 0-1

The Bishop and the King will create a mating net around the Kh4.

1. Bf6+ g5 2. Kh2!
Taking away h3 and threatening g3+

2…Qxe2
The alternative was unsatisfactory 2…Qxd6+ 3. g3+ Qxg3+ 4. Nxg3, and there is no stalemate because Black d-pawn was freed to move, 4…d5 5. Nf5 checkmate.

3. Bc3!
A beautiful zugzwang! g-pawn cannot move because of Bf6, while Black has to defend e1 and 2nd rank (to prevent g3+) at the same time.

3…Qf2 4. Be5! Qe1 5. g3+ Qxg3+ 6. Bxg3 checkmate

Now back to the theme of domination. Lots of Rinck’s studies featured a trapped (dominated) piece with minimal setups on an almost empty chessboard. The following position found its place in Jonathan Levitt’s “Secrets of Spectacular Chess, 2nd ed”. The Knight is dominating Black Bishop which cannot find a safe heaven on the long diagonal.

1. Nf7! Ba1 2. Kb1
And the Bishop will end up caught, if 2…Bc3 (or Bd4), then 3. Nd6+ Kxc7 4. Nb5+ wins. Alternatively, 2…Bf6 (Bg7) 3. Nd6+ Kxc7 4. Ne8+.

This cute forking reminded me of Peter Leko’s missed win in the game against none other than Garry Kasparov, played at the 2002 Bled Chess Olympiad.

Taught by the previous example, we should find 53. Nf8+ Kd6 54. c7 Kxc7 55. Ne6+ Kd6 56. Nxg5 Kd5 57. Kd3, comfortably winning. Giving the f-pawn away on earlier occasion doesn’t help, 54…f3+ 55. Kxf3 Kxc7 56. Ne6+ Kd6 57. Nxg5 Kc5 58. Ke3 Kc4 59. Kd2, and White arrives just on time to shoulder-off the Black King.

Peter Leko threw the win away by actually playing 53. Ne5 Bd8 54. Nc4 Kd5 and a draw was agreed.

The game of our interest was played in the first round between Anand and Hammer. Anand is famous for his fantastic endgame technique in the positions where he has Bishop(s) and space advantage.

After a long struggle from the Italian opening, the players have reached the diagrammed position.

Anand continues to remove the barriers by trading Black b7-pawn.

48. a6! bxa6+ 49. Kxa6?

But here he missed to create a zugzwang with the brilliant 49. Kc6! The point is that Black Knight is totally out of squares. Black could try 49…a5, but White ignores the pawn and moves 50. Ba3!, keeping f8 and c5 under control, then 50…Nb8+ 51. Kb7 Nd7 52. Kc7! a4 53. Kc6 (now Black is running out of moves) Nb8+ 54. Kb7 Nd7 55. Kc7 and the Knight is lost.

The game was with the rapid time control and probably both players fought with the time trouble.
The play continued with…

49…Kf7 50. Kb7

50…f5?

And here Hammer panicked and simply threw away the draw. Anand’s maneuver with the King to a6 removed his key piece just that bit far to allow Black to save the game. 50…Nf8! was the correct move, even if Black is losing opposition, which is probably what Hammer saw.

But calculating a little deeper, 50…Nf8 51. Bxf8 (otherwise White would be in trouble if the Knight jumps to e6 and f4) Kf8 52. Kc6 (Preparing to take the opposition) Ke7 (The only move, 52…Kf7? loses to 53. Kd7 and White breaks through) 53. Kc7 Ke6!! (White has the opposition and Black King will be kicked out, but at least now Black can choose on which side to go – and he goes after the e4-pawn!) 54. Kd8 (54. Kc6 Ke7) Kd6 55. Ke8 Kc5 56. Ke7 Kd4 57. Kxf6 Kxe4 58. Kxg5 Kf3 (And in the resulting pawn race both players get a new Queen, with Black to check first and hold the draw) 59. h4 e4 60. h5 e3 61. h6 e2 62. h7 e1=Q 63. h8=Q Qd2+ etc

51. exf5 (Now Black is losing) e4 52. Kc6 Ne5+ 53. Kd5 Nd3 54. Bd2 Nf2 55. Bxg5 Nxh3 56. Be3 – the Knight is trapped 1-0

Author of six chess books, out of which “Winning Chess Middlegames” was one of the best selling chess books in the world of 2009, and the most recent “Ruy Lopez Revisited” was published about six months ago.

Regular commentator since 2007 in Tata Steel Chess, Wijk aan Zee, Super Tournament.

His insights on the 3rd round game between Paco Vallejo and Magnus Carlsen are posted bellow, please note that this was a real-time commentary.

Paco Vallejo – Magnus Carlsen
Sao Paulo Masters Final, round 3

Hello everyone, I am GM Ivan Sokolov, member of the Chess Evolution team, and will be commenting for you the games from Sao Paulo today! Vallejo lost first two games and he is likely eager to get back in the tournament, will Carlsen allow this or he will go himself for the highest spot in the table?

1. Nf3 g6
Interesting start! Pirc or KID?

2. e4 Bg7
So Magnus wants Pirc or fianchetto. It is important that Paco is also 1.e4 player, so he is not confused with this stuff.

3. d4 d6 4. Bc4 Nf6 5. Qe2
Usually Paco wants on 4…Nf6 to employ the e5 line and he goes for it. After 5…O-O a good option is, 6. e5 something that Vallejo has in his arsenal.

5… O-O
6. e5 is a logical consequence of 5.Qe2 and would be strange for white now to change his mind and go for another set-up.

6. O-O
Still keeping the e5 possibility, probably just a transposition. Bg4 and c6 are on the list for Magnus. Anyhow black has to be ready for white’s 7.e5 here.

6… Bg4
To mention as an alternative to e5 is 7.Nbd2 with 8.h3 to take a bishop pair is logical for white. Will Paco go for it?

7. Rd1
Keeping both options open. Paco is trying to get back in the tournament, but on the other hand 0/3 would have been a disaster, so he would likely try to play a balanced, controlled game. 7…c6 is a first hand option for black. With 7…c6 black is ready to play 8…d5, while at the same time gets this square defended as an outpost for his knight.

7… Nc6
Carlsen is preparing on 8.e5, to play 8…Nd7. White can no longer play 8.Nbd2 (d4 hangs) hence white might likely play c3, a good option. Black is with his strategy fighting for d4 square, so white c3 might be logical, though black is still likely to continue with e5. Paco is thinking, not an easy decision, he understands that 8.c3 is logical, but he might not like 8…e5, since than after 9.d5 black has more or less a regular KID position, while it would take white quite some time to start a regular KID queenside pawn advance. Maybe 8.h3 taking the bishop pair, though black looks ok after 8…Bxf3 9.Qxf3 e5.

8. Bb5
Attacking Nc6 and hence fighting for the d4-square.

8… Nd7
Black is renewing pressure on the d4-square. And again on 9.c3 black is likely to play 9…e5. I think that Carlsen is having a comfortable game. Maybe after a 0/2 start it is difficult for Vallejo psychologically. It might well be that Vallejo would have to go something like 9.c3 e5 10.Bxc6 bxc5 11.dxe5 Nxe5 12.Nbd2 with 13.h3 to follow. Immediate 9.Bxc6 bxc6 10.h3 is not good for white, since black gets an excellent game after 10…Bxf3 11.Qxf3 c5. Carlsen is ahead in development and Vallejo would have to stabilize the position. At some stage he is very likely to take Bxc6 in order to minimize the dynamic element in a position.

9. c3 e5 10. Bxc6 bxc6 11. h3 Bxf3
As expected Paco has exchanged black knight on c6 in order to release a central tension. I still think that 11.dxe5 with Nbd2 and h3 to follow was better option for white. Namely black can now get a good game with 12…exd4 13.cxd4 c5

12. Qxf3
Option for Carlsen to consider here is also 12…f5!? as after 13.exf5 he can proceed with 13….d5, later collecting f5 pawn and having a great game. White’s main problem is that he is behind in development. Black is very comfortable. Carlsen understands perfectly well that he is very fine. He is now looking for the ways to take over.

12… exd4
So Magnus decides to open the long diagonal. Now after 13.cxd4 he would likely proceed with 13…c5

13. cxd4 c5
14.Be3 now looks like almost forced for Vallejo

14. Be3
14…Qb8 is a standard plan. 14…Rb8 is also good. Vallejo’s problem after 14…Rb8 is that its difficult to defend b2.

14… Rb8
If 15.Qe2, black would hit on e4-square. 15.b3 Rb4! might go terribly wrong for white. White cannot sac b2-pawn in a clever way. 15.b3 is not an option (due to 15…Rb4), so 15.Qe2 looks forced. Not easy for Paco, when you start 0/2 this is not kind of a game you are looking for!

15. Qe2
Now lets see how Magnus develops his initiative, white needs just one move 16.Nc3 and has completed development and is ok. Can Carlsen play 15…Rb4 16.a3 Rxd4! with a great compensation for the exchange.

15… Rb4
Yes he does!

16. a3 Rb3
No Rxd4…. a pity… I am convinced that the compensation for exchange would have been great. 16…Rb3 is btw positionally very sound, white b2 pawn is now weak. I think Paco has to play 17. Nc3. 17.Nc3 Qb8 and now if 18. Rab1 cxd4 and after swapping everything on d4 white Nc3 would hang, therefore 17.Ra2

17. Ra2
This looks very unnatural for white and cannot be good.

17… Qb8 18. Qc2
Rf8 is black only unemployed piece and is likely to now get in action. Vallejo threatens to consolidate, so Magnus has to be smart and has to be fast.

18… a5
Carlsen plays 18…a5 in order to protect his rook position on b3. If white now plays move like 19.Nd2 black calmly responds with 19…a4. So Vallejo would have to develop his knight to c3. Can Vallejo take dxc5?

19. dxc5 Nxc5
20.Bxc5 is to be expected. The drawback of early take of c5 for white is that black now threatens to place his bishop to d4.

20. Bxc5 dxc5 21. Qxc5
Paco does not allow Bd4 plan. I am under the impression that Magnus did not take the most out of the position. Perhaps sacrificing an exchange on d4 was to be considered after all. Black Rb3 is dominant to white passive Ra2, so Rd8 may look logical.

21… Rd8
Since black rook is dominating white, Magnus would like to exchange some pieces, so this domination would be more felt.

22. Nc3
Deep thought for Vallejo and not much time left. This was, however, probably the best move. Perhaps 22…Rd1 23.Nd1 Rd3. Carlsen is thinking, though difficult to come up with something else rather than 22…Rxd1.

22… Rxd1+ 23. Nxd1 Qd8 24. Ra1
24.Qd5 was an alternative.

24… Qd2
Black is very dominant, though not that many pieces left on the board. Not easy for Paco to entangle his pieces, perhaps Rc1?

25. Rc1
Yes, it is played, but the computers immediately show it is not a good idea… apparently Qe1 with Rd3 to follow is strong for black.

25… Qe1+ 26. Kh2
Now Rd3 wins material, due to a terrible position of his king, white can after Rd3 not prevent material loses. 26…Bh6 is also maybe winning.

26… Qxe4
Strange, but still lot’s of problems for white. A great game by Magnus until now, amazing to see how he is finding chances to play for a win. The Rd3 missed is probably due to time trouble.

27. Qc2 Be5+ 28. g3
Black’s position is still better and he should try to keep the queens on the board. Maybe now 28…Qf3 with future h5-h4. Actually the position is much more easy to play for Black than for White, Black got a clear target – White’s king! Paco will have here a very hard test to survive the next 12 moves until the time control.

28… Rd3 29. Ne3
Of course White needs to get their knight finally in the game, also next move could be Qc4 in a hope for a Queen exchange. Somehow Black’s 28…Rd3 is not looking to good.

29… h5 30. Nc4 Qd5
Amazing, Magnus blundered a piece?

31. Nxe5 Rd2
Such mistakes we don’t see very often from Carlsen…probably his biggest blunder of the 2011 year…

32. Qc5 Qxc5
White’s position is totally winning.

33. Rxc5 Rxf2+ 34. Kg1 Rxb2 35. Rxc7 Rb3 36. Kg2 Rxa3 37. Nxf7 a4 38. Ra7 Ra1 39. Ne5 g5
What to say…. Black played a great game, overplayed White, been very close to winning, then let White a bit of “fresh air” and blundered a full piece in one move… Very sad story for the current World Number one…

40. g4 hxg4 41. hxg4 a3
White can play now the simple Nf3-g5… seems that Black got no chances at all.

42. Nf3 1-0
And the game concludes with 3 full points for Vallejo, who is now ahead of Carlsen in the table. Thank you for following with me GM Ivan Sokolov and the Chess Evolution team.

What about Botvinnik? The famous strategist made many positional sacrifices, but usually with no escape for his opponents.

We’ll take a look at Knight sacrifice for two pawns. Pawns alone are not enough of compensation for a piece. So player has to seek for positional elements in his favour, but elements which have “permanent” (long-term) character. Some of those positional elements are: supported outpost, strong pawn center, open file, opponent’s strong pieces elimination etc.

White: Mikhail Botvinnik
Black: Ludek Pachman
Oberhausen 1961

This was a rather uneventful King’s Indian game with lots of pieces’ shuffling, but finally Pachman pushed f6-f5 and provoked White’s reaction.

1. Nf5!

Sacrificing knight for two pawns and one tempo (time unit in chess, one tempo=one move). 1. ef5 gives nothing because of 1…Nd4 and 2…Bf5.

1…gf5 2. ef5 Ng7 3. g4!

Only three moves after the diagram, the position has dramatically changed. White has a pawn-wall on the king side, but this is not an attacking formation. The first purpose of this wall is to keep Black pieces on passive squares. Ng7, Bf8 and Bd7 have no good moves. The only active piece is Nc5, but soon White will start to press all over the board.

3…Re5 4. Bd4 Re1

4…Bf5 attempt fails to 5. gf5 Rf5 6. Bg7

5. Re1 Re8

This is a slight mistake, but it’s not easy to give good advice for Black. Now the light bishop has no e8 square. It was better to regroup with Be8-Bf7.

6. Rd1!

Avoiding the trade, pressing over the d-file and keeping the e8-square blocked for light bishop. Botvinnik always knew which pieces to exchange and which to keep on the board.

6…Bc8 7. b4 ab4 8. ab4 Na6

The last Black active piece is pushed back and White owns the e4-square now.

9. Ne4 Re4

Black is forced to give material back as Nf6 was a deadly threat.

10. Be4 Nb4 11. Qd2 d5 12. cd5 cd5 13. Bf3

The game was adjourned here, and Black resigned the same evening.