12

In game 1 of the World Championship 2014 between Carlsen and Anand, Anand chose 12.gxf3. Why not play 12.Qc3+ instead? It takes control over the a1-h8 diagonal and prevents the black Nc6 to jump to d4 or e5. In addition, it gives white the option to grab on f3 with the queen, keeping the pawn structure on the kingside intact.

[FEN ""]
[Event "WCh 2014"]
[Site "Sochi RUS"]
[Date "2014.11.08"]
[Round "1"]
[White "Anand, Viswanathan"]
[Black "Carlsen, Magnus"]
[Result "1/2-1/2"]
[Eventdate "2014.11.08"]

1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5 5.Bd2 Bg7 6.e4 Nxc3 7.Bxc3 O-O 8.Qd2 Nc6 9.Nf3 Bg4 10.d5 Bxf3 11.Bxg7 Kxg7 12.Qc3+
Stephen
  • 1,293
  • 1
  • 9
  • 18
Rauan Sagit
  • 6,686
  • 5
  • 36
  • 56
  • 1
    I assumed it was because of **12...Ne5**. That forces **13.Qxe5+ f6 14.Qsomewhere**, after which the bishop can retreat, and Anand probably wanted the bishop over the knight. But I'm not sure enough of this to make it an answer. – RemcoGerlich Nov 10 '14 at 20:34
  • @RemcoGerlich the answer is positional (involving some tactics). – Rauan Sagit Nov 10 '14 at 20:38
  • doubled pawns are not as bad as you think – David Jan 04 '21 at 16:23
  • @David I think doubled pawns can cause serious problems and should be avoided when possible. It is a long term weakness and I prefer to avoid it. Having said that, every position should be judged individually. Still, in this example, gxf3 isolates the h2-pawn and creates weaknesses on f3 and f4 and I am not a big fan of that pawn structure. – Rauan Sagit Jan 06 '21 at 16:20
  • @RemcoGerlich unfortunately that's not how the thinking process of a chess player works. As the answer has pointed out, there are concrete reasons why Qxf3 should be avoided. There is no way Black can attack the h2 pawn, which White can use as a "weapon" by playing h4-h5. There is also no clear way for Black to exploit the f3 and f4 weaknesses (...g5 is out of the question for now) – David Jan 07 '21 at 07:13
  • @David: As for your first sentence, I am a chess player and it is how my thinking process works, whereas the answer used an engine. Of course the chess content of your comment and the answer is better than my four moves... – RemcoGerlich Jan 07 '21 at 08:57
  • @RemcoGerlich then you should definitely change how you think about evaluating chess positions. Concrete calculation should always prevail over abstract "principle". – David Jan 07 '21 at 10:28
  • @David while you have valid points, I think you can be more friendly in your comments. I think we are all exploring the game of chess and being friendly in our feedback is important to keep this community healthy and growing. Chess is not a tough and competitive quest for glory for everyone. For many people, I imagine it is a beautiful game that they enjoy. Dominating on the chess board is one thing, but dominating in the comment section is a different game in my view. – Rauan Sagit Jan 07 '21 at 21:28
  • @David as for the evaluation of the position after gxf3, perhaps you can post a new question and it can be explored in the answer section of that question? I think this discussion is better suited for a new question, since my question focuses on Qc3+. I think doubled pawns is an interesting topic. It is always hard to interpret written text. Saying something like "definitely change" is more tough than "in my opinion" or "from my experience". English StackExchange next stop? Cheers. – Rauan Sagit Jan 07 '21 at 21:35

1 Answers1

10
           [FEN ""]
                [Event "WCh 2014"]
                [Site "Sochi RUS"]
                [Date "2014.11.08"]
                [Round "1"]
                [White "Anand, Viswanathan"]
                [Black "Carlsen, Magnus"]
                [Result "1/2-1/2"]
                [Eventdate "2014.11.08"]
            
                1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5 5.Bd2 Bg7 6.e4 Nxc3 7.Bxc3 O-O 8.Qd2 Nc6 9.Nf3 Bg4 10.d5 Bxf3 11.Bxg7 Kxg7 12.Qc3+ e5 (13.dxc6 Be4 14. Qe5 f6 15. Qc3 Re8 16. Be2 Bxc6)
    (13.gxf3 Nd4 )     
   (13.Qxf3 Nd4 ) 
   (13.dxe6 Qf6 14. Qxf6 Kxf6 15. gxf3 fxe6) 

Computer Houdini 1.5a w32 suggests 13... e5

Now four options for white 13.dc 13.de ep, 13.gf or 13.Qf3 anything else loses a piece.

  • After 13. dc black replies 13... Be4 and white cannot take pawn on e5. If 14. Qe5 then 14... f6 15. Qc3 (15 Qe4 loses a queen). 15... Re8 this forces white reply 16. Be2 and black just takes on c6 16.... Bc6 , this is a win for black.

  • 13.gxf6 black moves its knight to the centre 13... Nd4 Knight on d4 cannot be moved. Pawn on f3 is weak. Black is better.

  • 13.Qf3 Again same idea – black’s knight occupies the centre Nd4 and whatever white answer is black threatens to crush the white centre 14... c6. Again black is better (maybe not as good as after 13. gf)

  • 13.dxe6 e.p. black needs to cover the diagonal – 13... Qf6 or 13... Qd4 anything else consolidates white’s position in the centre.
    13...Qf6 14 Qf6 (otherwise black takes on c3) 14... Kf6 (forced) 15 gf (forced) 15... fe again black is better (weakness on f3)
    13...Qd4 white cannot take on d4, so 14. e7 (if 14. gf black takes on c3 and again weakness on f3 is giving black an advantage) 14... Re8 15. gf again cannot take on d4. And now either 15...Re7 or 15... Qc3 and black has better pawn structure.

Pavel Nefyodov
  • 592
  • 6
  • 11
  • 2
    Good answer. I'd suggest upgrading from Houdini 1.5 to Stockfish 5 (https://stockfishchess.org/download/). Houdini 1.5 is the most advanced free version of Houdini but is quite old for a chess computer (released in 2010). Stockfish 5 is the latest version of Stockfish, a very small amount stronger than the most recent Houdini versions and is also free. – Cleveland Nov 11 '14 at 01:04
  • @Cleveland: Good point. Thank you for the link. – Pavel Nefyodov Nov 11 '14 at 18:48