Against an up-and-coming young player, one of the world’s best players loses his queen in an elementary combination. Was it an error, or a desperate attempt to change the course of the game?

In the Reykajavik Open, Dmitry Andreikin of Russia, ranked No. 21 in the world, was pushed to the limits by Lars Oskar Hauge, a young Norwegian kid. What happened reminded me of my own game against the same young player a few years ago, though I had an even tougher time than Andreikin.

Hauge, Lars Oskar vs. Andreikin, D.
Reykjavik Open 2016 | Reykjavik ISL | Round 5.11 | 11 Mar 2016 | ECO: B47 | 0-1
1. e4 c5 2. Nf3 Nc6 3. d4 cxd4 4. Nxd4 e6 5. Nc3 Qc7 6. g3 It is surprising to see a young player choose this line, but it is actually not harmless at all. The next few moves weren't particularly interesting.
6... a6 7. Bg2 Nf6 8. O-O Be7 9. Re1 O-O 10. Nxc6 dxc6 11. e5 Rd8! Black's idea is that he wants to force White to move the queen somewhere besides g4 - since the knight is still on f6. Eventually, White probably wants to move to g4, but this should ensure that Black has played an extra move.
12. Bd2 Nd5 13. Qg4 b5 14. Rad1 b4 15. Nb1! Incidentally, this seems to be the novelty, although I have a feeling that neither player had prepared this far. Compared to the move played before, Na4, there is almost no scenario in which Na4 will be better placed. From b1, White can still plan c4 - although it is much nicer to transfer the knight to the king side of course!
15. Na4 was played in an engine game in this position, a striking, but admittedly rare example in which the computers are not able to find the right move.  )
15... a5 16. Bh6 Bf8
16... g6 17. Nd2 would just create a lot of dark-squared weaknesses on the king side.  )
17. Bg5 Rd7 A slightly awkward move to make.
17... Be7 exchanging these bishops would leave a gaping hole on d6, which can be quickly targeted by White.
18. Bxe7 Qxe7 19. Nd2 Ba6 20. Ne4 and Nd6 will just throttle Black.  )
18. Nd2 Kh8 19. Be4!? An interesting regrouping. The bishop wasn't doing a whole lot on g2, but now it is a major player in the eventual kingside attack. Also note that White isn't in any hurry - next he is ready to play Nf3 and soon maybe even create threats like Bxh7. While Black can play h6, it is certainly not desirable as there is no clear counterplay for Black yet, while it gives White a target to aim at on the kingside (maybe even by g4 and g5!??). At the least, it is a very tough psychologically to be Black at this point in the game. It's quite possible that both players were even nearing time pressure as the game hasn't followed known paths. Andreikin now made a move that is instantly declared as a blunder by the chess engine, but is it really?
19... Qxe5!? Loses after
20. Bxh7 according to every chess engine. Could Andreikin have missed this simple tactic, or hoped for some miraculous defense? Tempting as it is to consider this to be just a human sort of blunder, Andreikin almost certainly had anticipated Bxh7, and planned to give up his queen for a rook and a minor piece. Objectively this is almost certainly not the best decision, but it drastically altered the game. Now at least, Black does not have to worry about an immediate mate, and his young opponent has to navigate a new situation. Hauge definitely felt comfortable in the role of the attacker. On the hand, trying to convert a position up queen for rook and piece may seem deceptively simple, but it is not so easy to adjust, as the computers might make you think. Most importantly, Andreikin could play simple moves quickly, in his usual style, something that was definitely not possible when he had to worry about a killer blow on the king side.
20... Qxe1+ 21. Rxe1 Kxh7 22. Nf3 The computer doesn't notice anything wrong with this, but practically, in such a situation, rather than try to consolidate, the best approach would have been to try to destroy Black's position without giving him a chance to organize his own pieces. In fact, on closer examination, Black's pieces are not bad, particularly his undeveloped bishop on c8, which has the potential to be a real monster once it does develop. Still it was hard to notice those issues at this point in the game.
22. h4!? going for h5 and h6 would have been more annoying, but I think Black is certainly not out of the game yet.  )
22... Kg8! 23. h4
23. Ne5 Rc7 and f6 next. It isn't clear how Ne5 helps White.  )
23... f6 24. Bd2
24. Qxe6+ Rf7! 25. Qxc6 Bb7 26. Qb5 fxg5 with three pieces for a queen, I would probably prefer Black in a practical game despite the silicon monsters doomsday warnings.  )
24... e5 Notice any change in the situation? Where is White's initiative? It is probably about this point that Hauge started to feel a little bit uncomfortable about his position, and it was certainly tough to decide whether he should be just trying to crush Black or consolidate. In the end, the inexperienced Norwegian remained caught between the two extremes and couldn't decide on the right plan.
25. h5 Rd6 26. Qh4 Bf5 27. c4
27. h6 should probably have been played. At least White gains something for all that effort put into pushing the h-pawn, and the c2 pawn doesn't seem very important. As I mentioned before, Hauge seemed a little confused between whether he should go all in or keep his pieces organized consolidated, and in this position that confusion is most evident.  )
27... bxc3 28. bxc3 Rb8 29. Qa4 Again, h6 was necessary, but Hauge was probably starting to see ghosts, i.e. threats like Rb2. To be fair, h6 also doesn't quite achieve anything immediately.
29... Nb6!? 30. Qxa5 Bg4! The position has become more and more messy. And Black's pieces are improving their coordination.
31. Be3 Nc4 32. Qa7 Rb5 Maybe not the best move objectively, but I really like how the rook feels on b5. Andreikin was already playing more by an aesthetic feel than calculation.
33. Qa4 Nxe3 34. Rxe3 Bxh5 the bishops are powerful, and the rooks unchallenged. White certainly is no longer better in a practical game, though objectively he probably still is. Importantly, it has become harder for White to make decisions than for Black:
35. Qc2 Bf7 36. Re1 Rd8 A slow regrouping, realizing that it is not necessary to hurry.
37. a4 Rbd5 38. Ra1 Bc5 39. Qe4? The big slip up.
39. a5 Rd3 would have remained extremely messy.  )
39... Rd3! 40. c4 Rc3! Certainly the move missed by White. It is surprisingly impossible for White to defend against Bxc4 and Bd5.
41. a5
41. Qxc6 Rxc4 42. Qa6 Rc3 and the knight is stuck again.  )
41... Bxc4 Bd5 just can't be dealt with. In my game against Hauge, I also somehow managed to turn my desperation into a completely unexpected win, but I feel it won't be long before Hauge stops letting experienced grandmasters cheat death.


Parimarjan Negi is an Indian grandmaster who is the second-youngest ever to earn the title (at 13 years 4 months and 22 days). Ranked No. 88 in the world, he is currently a sophomore at Stanford University.