World Chess’s columnist explains how a top correspondence player took a tip from Magnus Carlsen, with spectacular results.

The 29th World Correspondence Championship is ongoing, and with almost 80 percent of the games finished, the current leader is Jacek Oskulski of Poland. Wins with Black in serious correspondence events are almost as rare as unicorns, but Oskulski managed the feat in his game with Ángel-Jerónimo Manso Gil of Spain. How did he do it? By taking some advice from Magnus Carlsen of Norway, the current classical World Champion.

Carlsen defeated Levon Aronian of Armenia on the white side of a Ragozin in January 2015 in an impressive and one-sided game. In video commentary afterward he suggested that Black’s bet would have been to aim for kingside counterplay  by playing g5. This idea wasn’t unknown before the game, but with the champion’s endorsement the move and plan became de rigeur.

Oskulski is among those who used it, though the exact point when he used the move was a novelty. He, like Carlsen, won impressively, and the impression he left is similar. After Carlsen’s game the line looked almost unplayable for Black. After this game, players may view White’s chances with a similarly pessimistic attitude. The ball is back in White’s court.

Manso Gil, Angel-Jernimo vs. Oskulski, Jacek
WC29/final | ICCF | Round ? | 20 Jun 2015 | ECO: D38 | 0-1
1. d4 Nf6 2. c4 e6 3. Nf3 d5 4. Nc3 Bb4 5. cxd5 exd5 6. Bg5 h6 7. Bxf6 Qxf6 8. Qa4+ Nc6 9. e3 O-O 10. Be2 Be6 11. O-O a6 12. Rfc1 Bd6 13. Qd1 Ne7 14. a3 This position first occurred back in 2004, but it really got on the map two years ago when Magnus Carlsen won a very impressive game with White against Levon Aronian in Wijk aan Zee. In his video commentary to the game he suggested that Black ought to have tried ...g5 at some point, and this general suggestion has been tested more than once since then. (In fact it had been tried in a couple of earlier games too, but there's nothing like the World Champion's recommendation to make the move a priority.)
14... c6 Carlsen-Aronian went as follows:
14... Rfd8 15. b4 Nc8 16. Na4 b6 17. Nb2 Ne7 18. Nd3 Ng6 19. a4 a5 20. b5 Re8 21. Rc3 Bf5 22. Rac1 Rad8 23. Nd2 Rd7 24. g3 Nf8 25. Bg4 Nh7 26. Bxf5 Qxf5 27. Qf3 Qg5 28. h4 Qe7 29. Rc6 Nf6 30. Nf4 g6 31. h5 Kg7 32. hxg6 fxg6 33. Nxd5 Nxd5 34. Qxd5 Bxg3 35. Qg2 Bd6 36. Nc4 Rf8 37. Ne5 Bxe5 38. Qxg6+ Kh8 39. Qxh6+ Kg8 40. dxe5 Qxe5 41. Rg6+ Kf7 42. Rc4 Qa1+ 43. Kg2 Rh8 44. Rf4+ Ke8 45. Re6+ Re7 46. Rxe7+ Kxe7 47. Re4+ Here's an example with the immediate push of the g-pawn:  )
14... g5 15. g3 Qg7 16. e4 dxe4 17. Nxe4 Bd5 18. Nxd6 cxd6 19. Bc4 Bc6 20. Bf1 Rac8 21. Bg2 Bd5 22. Rxc8 Rxc8 23. Rc1 Rxc1 24. Qxc1 Qf6 25. Qc3 Nc6 26. h3 Kg7 27. Qd3 b5 28. Nd2 Bxg2 29. Kxg2 Nxd4 30. Ne4 Qe5 31. f4 gxf4 32. gxf4 Qd5 33. Kf2 f5 34. Nxd6 Qxd6 35. Ke3 Qe6+ 36. Kxd4 Qf6+ 37. Kc5 Qxb2 38. Qd7+ Kg6 39. Qe6+ Kg7 40. Qe7+ Kg6 41. Qe8+ Kg7 42. Qe7+ 1/2-1/2 (42) Kunin,V (2579)-Gonzalez Vidal,Y (2550) Havana 2015  )
15. Na4 g5 The immediate
15... Rae8 makes sense, so that if White plays
16. Nc5 Black has the option of retreating the bishop without shutting in the rook.
...  Bc8  )
15... a5 16. Nc5 Bxc5 17. Rxc5 Nf5 18. b4 axb4 19. axb4 Rxa1! 20. Qxa1 Nd6 21. Ra5 Rc8 22. Ne5 Qd8  )
16. Nd2 The immediate
16. Nc5!? makes sense, as every way of meeting the threat to the b-pawn necessitates a concession: ...Bc8 locks in the a-rook, ...Rb8 or ...Ra7 ties the rook down, ...b6 softens up the queenside pawns, and ...Bxc5 surrenders the bishop pair and greatly undermines his kingside attacking chances.  )
16. Ne1 makes some sense, and so does  )
16. b4 , intending a possible minority attack (after Nc5) with a4 and b5. But timing is everything:
16... g4 17. Ne1 h5 18. Nc5 Bxh2+! 19. Kxh2 Qxf2 Black's pawns and kingside initiative are worth more than the piece.  )
16... Rae8 17. Nc5 Bc8 18. Nf1
18. Bg4!? looks sensible, threatening to take on c8 and then on b7. The good news is that Black's best option is to swap off both his bishops:
18... Bxg4 19. Qxg4 Bxc5 20. Rxc5 The bad news is that even after this Black is slightly better after
20... Nf5 followed by ...Nd6. The knight is great there, making it hard for White to get the minority attack rolling, and in the meantime Black can make progress in the center and on the kingside. His queen goes to g6, threatening ...f5, and then he might double rooks on the e-file, looking to play ...Ne4 and/or ...f4. Two valuable games illustrating these plans are Bobotsov-Petrosian, Lugano 1968 and Portisch-Kasparov, Skelleftea 1989. In general, White is not particularly happy in these positions if his light squared bishop is gone and he can't easily achieve b4-b5.  )
18... Qg7 Preparing ...f5. Black is already better.
19. Qb3 Clearly not the piece setup White wants, but otherwise Black will have time to play ...f5 and ...Ng6, defending the b-pawn laterally with the queen.
19... Bxc5 This is a concession, but even so Black is ahead in the race: his kingside play will be more dangerous than White's queenside counterplay.
20. dxc5
20. Rxc5 f5 also favors Black, who will play ...f4 next against almost any sensible White move.  )
20... Ng6 21. Re1 The rook no longer had anything sensible to do on c1, so he shuffles the rooks to the center.
21... Re7 22. Rad1 f5 Here the question is, what can White do? Opening the queenside will take a very long time and there are no levers for central counterplay. Black will simply and quickly push the kingside pawns, and his pieces are perfectly placed to support the pawns and to use the open lines once those pawns are exchanged away.
23. Nd2 g4
23... f4 24. e4 is the point.  )
24. Bd3 h5 25. Qc3 Ne5 Black is also better after trading queens. It isn't clear which decision is objectively better, and in over the board play it might be a matter of taste. In correspondence chess it's more complicated, and maybe the players worked out some very deep drawing scheme for White in the ending. The bottom line for us is that White is suffering in either case.
26. Bc2 h4 27. Qd4 Qg5 28. Qf4 Qh5 29. b4 Ng6 30. Qd4 White's position looks a little healthier than it did a few moves ago, and a4 (and then b5) seem just over the horizon. Black has two important pawn breaks of his own here: ...f4 and ...g3, both of which can be supported by ...Qh6 and ...Rfe8.
30... Rfe8 31. Rf1 Now White can meet ...f4 with exf4.
31... Qh6 32. a4 Nf8
32... f4 was playable, as
33. Bxg6 Qxg6 34. Qxf4 Bf5! gives Black more than enough compensation for the pawn. (Among other ideas, Black can play ...h3, creating a light squared gash around White's king. )  )
33. Qc3 f4! Finally Black is ready.
34. exf4 Ne6! 35. g3 Black is much better here, so White decides it's best to have a pawn to compensate him in his suffering.
35... Ng7! Keeping control of the e-file and preparing to seal White in with ...Nf5. After that, Black can look for a breakthrough, e.g. on the h-file.
36. Nb3 hxg3!
36... Nf5? 37. Bxf5 Bxf5 38. Nd4  )
37. hxg3
37. fxg3?? loses on the spot to
37... Re2  )
37... Nf5 38. Rde1 Now
38. Bxf5 Bxf5 39. Nd4 is worse than useless:
39... Rh7 and mate next move.  )
38... Rh7! Just in time! Against anything else White would be completely fine.
39. Rxe8+ Kf7 40. Re7+! Kxe7 41. Re1+ White needed this to come with tempo so his king could escape.
41... Kf8 42. Kf1 Now White is on the verge of having an extra pawn and the better position to boot. Black has only one move to keep the advantage (and even to avoid being much worse).
42... d4! 43. Nxd4 Nxd4 44. Be4!
44. Bxh7?? Qh1#  )
44. Qxd4? Qh1+ 45. Ke2 Qf3+! The tempting and obvious
...  46. Kd2 Rd7  )
44... Rd7! Black is winning. White's two pawns aren't enough for the piece, especially since they are doubled pawns and immobile too, as the f2-pawn in particular is needed to shield White's vulnerable king.
45. Rd1 Qh8 46. Qe3 Nf5 47. Qe1 Rxd1 48. Qxd1 Qd4 49. Qxd4 Nxd4 50. Kg2 Bf5 51. Bxf5 Nxf5 After
51... Nxf5 52. f3 Black has more than one way to win the ending. For example:
52... a5 53. bxa5 Nd4 54. fxg4 Nb3 and White's queenside pawns all disappear (...Nxa5-b3xc5xa4) before White can create any real trouble on the kingside.  )


Dennis Monokroussos is a FIDE master who has written about chess on his blog “The Chess Mind,” since 2005. He has been teaching chess for almost 20 years and for the last 10 years has been making instructional chess videos, which can be found at Between 1995 and 2006, he taught philosophy, including a four-year stint at the University of Notre Dame.