GOTM; January, 2013.    

Note that the diagrams are fairly large, if you save them to your computer, you will be able to see and study the positions better.  
(The size that is used here is less than 1/2 the actual size of the "*.jpg" images.)  

This system is known as ... "The Kalishnikov Sicilian."  [more]  
(I have seen Nak play this in blitz on the "Play-Chess" server, yet I could find no tournament examples of him using this system over the last few years.)   

     6.g3!?,  (Square control, possible fianchetto of White's LSB.)   
According to a book on the Kalishnikov, this is a relatively harmless line, with no real teeth. I think that the main reason that Carlsen chose this particular branch of this opening was that it was unlikely that Nakamura could have thoroughly prepared this exact variation for tournament play. If this is true, it also means that that Carlsen probably has prepared it himself - in some depth.  

   r1bqkbnr/pp3ppp/2np4/1N2p3/4P3/6P1/PPP2P1P/RNBQKB1R b KQkq - 0 6   

6.g2-g3 is like the 6th choice of the computer's opening book.  

                         [ According to the "Powerbook" the main line here would be:   
                            6.c4 Be7;  7.N1c3 a6;  8.Na3 Be6;  9.Be2,    
                            This seems to be the most commonly played line for White, at this particular fork in the road.   

                                           ( Or Black could try: 9.Bd3 Bg5;  10.0-0 Bxc1;  11.Rxc1 Nf6;  12.Nc2 0-0;  "~"   with an unclear position.  

                                             [ See MCO-14, page # 344; col. #01, and also note (e.). ] )   

                           Returning to the main PB line, we have the following moves:   
                           9...Bg5;  10.0-0 Bxc1;  11.Rxc1 Nf6;  12.Nc2, "+/="  and White is just marginally better.   

                           See the contest: 
                           GM David Navara (2702) - GM Alexander Moiseenko (2679);  [B32] / 
                           FRA-chT; Top 12 / Mulhouse (Rnd. #6.2) / 31,05,2011.   
                           {White won a nice game 1-0 in around thirty total moves.} [replay] ]   

     6...h5!?;   (Too aggressive?)    
True to his style of playing chess, Nakamura chooses an extremely belligerent line, yet this might be going a bit overboard, at least in my opinion.   
(6...h5!? is not new, although it is seldom used. See the famous encounter:  GM J. Nunn - GM N. Short; / Hoogovens / Wijk ann Zee, 1990.  
 {White won a sharp game, 1-0 in just 33 total moves.} [replay] )   

   r1bqkbnr/pp3pp1/2np4/1N2p2p/4P3/6P1/PPP2P1P/RNBQKB1R w KQkq h6 0 7   

6...Be7;  looked to be a much safer path for the second player. (See below.)  

                         [ A more conventional way of playing would have been:   
                           (>/=) 6...Be7;  7.Bg2 Be6;  8.N1c3 a6;  9.Na3 Nf6;  10.0-0, "+/="  10...0-0;  "~"  (unclear)   
                           when White is slightly better, yet the second player has fairly good play in this position.  

                           (I searched the database, there were only a handful of games - with this position.   
                            However, none of these were between GM's or even players with a FIDE rating of over 2400.) ]  

The next few moves are normal, first Black kicks back the dangerous White Knight on b5; and then both sides simply develop their pieces. 
(After 8...b7-b5; I cannot find any matches to this game in any DB.)  
     7.N1c3 a6;  8.Na3 b5N;  (The first new move of this game.)   9.Nd5 Nge7;  10.Bg2 Bg4!?;    
Again, this is very aggressive ... 
and there may not be anything really wrong with it. Yet I must point out that Fritz found a much simpler continuation that probably yields Black easy equality. 

                         [ Black gets a quick and fairly simple path to equality by playing the following continuation:   
                            >/=  RR 10...Nxd5!;  11.exd5 Ne7;  12.c3 Bg4;  "="  (equal)  when all the engines agree   
                            that Black has a solid game ... with no real problems here. ]   

   11.f3 Be6;  12.c3 h4;  13.Nc2 Bxd5!?;  (Dubious?)   
Black rids himself of his backward Pawn and also gets rid of the monster of a WN on the most important d5-square ... ... ...   

   r2qkb1r/4npp1/p1np4/1p1bp3/4P2p/2P2PP1/PPN3BP/R1BQK2R w KQkq - 0 14   

Yet it seems to me to be a very bad idea for Black to give up his important light-squared Bishop at this stage of the game. 
(This point seems to be the beginning of most of Black's problems in this particular stress-filled encounter ... 
watching the live feed from the official website, Nakamura {at times} seemed to be in real, physical discomfort.)  

                         [ Another possibility was:  RR  13...h3!?;  14.Bf1 Bd7;  15.Nce3 Nxd5;  16.Nxd5, "+/="  (A small edge for White.)   
                           when White may be a bit better, yet this is still a substantial improvement over the course of the actual game.   

                                         ******************************************************************************************    

                           Clearly - the best variation was the following continuation:    
                            >/= 13...Na5!;  14.f4 Rc8;  15.Ncb4 exf4;  16.Nxf4, "+/="    
                            when Fritz, Rybka and Houdini agree that White has a small edge, yet it is less than 25/100. ]   

     14.exd5 Na5;  15.f4!,  (center / initiative)    
Clearly this was the most active continuation here for White. 
(White gains space, attacks the center, unleashes his KB, and also plans to use the f-file after he castles Kingside.)  

                         [ RR 15.Ne3!?, "+/="  (White is slightly better.) ]   

The next few moves all look to be be forced and/or best ... for both sides, here. 
(Some programs like 16.Qf3, {instead} in this line for White. And </= 16...Nh6?; 17.0-0! was just about winning for White.) 
     15...Nf5;  16.g4! h3;  17.Be4 Nh4?!;   (An error?)   
Clearly, this was already a bad play for Nakamura, who was consuming a lot of time on his chess clock.   

                         [ A big improvement would have been the following line here:    
                           >/= 17...Qh4+; ('!')  18.Kf1 Ne7;  19.Qe2, "+/="  (A solid edge for White.)   
                           and while Carlsen is somewhat better here, it is not inconceivable that    
                           Black could hang on for an eventual draw. ]   

     18.0-0, "+/"  (White is clearly better.)   
This has got to be one of those cases where Black has been left to scratch his head and wonder exactly where he went wrong ...   

   r2qkb1r/5pp1/p2p4/np1Pp3/4BPPn/2P4p/PPN4P/R1BQ1RK1 b kq - 0 18   

Not only does Black have a bad game, it is not easy to find a plan of logical development where Black does not end up shedding a decisive amount of material.   

                         [ RR 18.fxe5 dxe5;  19.Qe2, "+/=" ]   

     18...g6!?;   (hmmm)    
This makes even more holes in Black's position, maybe better was the box's suggestion of playing the BQ to the c8-square.    

Now 20.g5!, was practically winning here for White. (At least, according to the strong chess engine, Fritz 13.) 

Note that - in a wide open board - Carlsen decides to safely shelter his King in the corner. 
     19.Kh1!? Bg7?!;   (A bad play.)    
It's easy to understand that Black wants to get a little development going ... 
and also cover his gaping dark-squared weaknesses. 

However, this was - for all intents and purposes - a virtual blunder here for Nakamura, who had to play 19...f5![]; which may have even saved his game. 
(Both Fritz and Houdini agree that 19...f7-f5!; was the best move for Black.)  

   *** *** *** *** *** *** *** ***     *** *** *** *** *** *** *** ***      *** *** *** *** *** *** *** ***     

     20.f5!?,   (Too routine?)    
While this certainly gives White a big edge, it is far from being the best continuation here. 
(See the line given, just below.)   

                         [ A definite improvement was:  >/=  20.fxe5! Bxe5;  

                                          ( Or </= 20...dxe5?!; 21.d6!,  "+/-")   
                           

                           21.Nd4!, '±'  (+/)   (White is clearly better.)    
                           Analysis - Fritz 13 & also the very powerful engines, Rybka and Houdini. ]  

     20...gxf5;  21.gxf5!?,   (Sub-optimal?)    
This is OK, and White even gets a nice solid edge with this move. But I feel compelled to point out that the machine chooses a completely different line.   

                         [ Rybka, Houdini, Fritz and also Shredder all prefer the following variation over what was chosen in the actual game:   
                           (>/=)  RR  21.Bxf5 Bh6;  22.Bxh6 Rxh6;  23.Qe2 Qg5;  24.Ne3,  "+/="  (White has a small advantage here.)   

                            and White is solidly better. (The main difference between this line and the actual game continuation is that the first    
                            player rids himself of his LSB, which was potentially a bad Bishop.) ]   

     21...Ng2?;   (Simply terrible!)    
This looks risky - and it was. 
(In fact, this could be the losing move for Black. It causes the Fritz engine's red light to come on the the evaluations to go crazy. After things settle down ... 
 after a few minutes of machine time ... the box shows that White should have a won game.)  

The other aspect of this move is that Carlsen now gets a grip on the game, and from here (until resignation), he never relinquishes his iron control over the board.   

     22.f6! Bf8;  (hmmm)   
To capture on f6 ... or not to capture? (That is the question!)   

               [ Black is also losing after the following continuation here:    
                 RR  22...Bxf6;  23.Qf3 Rh4;  24.Bg5 Nf4;  25.Bxh4 Bxh4;  26.Rg1 Kd7;  27.Nd4! Qf6;  28.Ne6!,  "+/-"  (White has a won game.)  
                 and Black can delay the end of the game for a while, but it is certain that he will lose. ]   

Now Fritz likes 23.Bf5!, while Houdini prefers 23.Qe2.  (Just about everything is winning here for White.)    
     23.Qf3 Qc7;  24.Nb4 Nb7;  25.Nc6 Nc5;  26.Bf5!,   (Best.)   
After missing it a couple of times, Carlsen finally discovers the same path that the metal monster likes.   

                         [ RR  26.Bd2!? Nxe4;  27.Qxe4,  "+/="  ]  

Now Black is lost, he cannot even castle here. (The box chooses 26...a5; but it also shows that Black's position is pretty much hopeless.)   
     26...Nd7;  27.Bg5 Rg8!?;    
Black attempts to gain a little activity for his pieces.   

Objectively, 27...Nb8; and/or 27...Nc5; were both to be preferred over this move. 
(It would be useless to attach a full question mark here. Black must perish, all that remains is for Nakamura to chose the method of his execution.)  

Now White finds the winning idea of Qh5, with the simple threat of Qh7, winning material.   
     28.Qh5! Nb6;  29.Be6 Rxg5[];  30.Qxg5 fxe6;  31.dxe6,  "+/-"  (White has a decisive edge.)  
Black has had all that he can stand - and wastes no time in giving up his sinking ship.   

   r3kb2/2q5/pnNpPP2/1p2p1Q1/8/2P4p/PP4nP/R4R1K b q - 0 31   

Relentless play by Carlsen who swamps America's best player ... and even had the gall to make it look rather easy, as well. 
(BTW - Magnus tied Garry's record for the best score in Wijk aan Zee, with 10/13.)   

   Copyright (c) A.J. Goldsby, 2013. All rights reserved.