Great Wall of China - Game 7

 "Ding is still standing and is difficult to push him over" - Anish Giri

Game 7 was going to mark the halfway mark of the World Championship, and it proved to be one of the best and hard-fought games with twists and turns almost until the very end! First up Gukesh came up with a fundamentally new concept and started to put pressure on Ding on the clock. Then with the pressure of a looming time pressure, Ding went for a forcing continuation and was slowly getting outplayed and Gukesh was increasing his advantage methodically. But Ding is not a World champion for nothing, he fought valiantly giving his heart out! And started throwing obstacles along Gukesh's path. Eventually Gukesh also landed in time trouble and by move 39 perhaps had lost lion's share of his advantage. But Gukesh made one last tricky move and uncharacteristically Ding erred on the move 40m thereby giving Gukesh the advantage back in the second time control phase. This part of the game, Gukesh is being criticized for missing the win. I feel this is unfair, while it is very painful to miss a winning chance to secure the lead for the first time in this World championship, it was nevertheless not easy from practical terms. 

And, in this world championship there is only half an hour left after move 40 unlike in the earlier events with 3-time controls. There are also three games on continuous days before a rest day. As a lover of Classical Chess, am not happy to see this, development which tires out the players and lets them commit errors just to make the game a tad bit more entertaining for the (large part) oblivious spectators! While things are as they stand, the time trouble phase started soon in the second time control too, and both sides made some errors, but it perhaps hurt Gukesh most that he could not convert. Gone are the days of several adjournments and lot of time to ponder over endgames. Those times clearly showed players playing endgames at a much higher level than now, for understandable reasons! Personally, i wish the players had more time to think about this fascinating endgame. End of the day Ding put up an incredibly brave fight and showed his class and deservedly at the end drew the game. Nakamura made an interesting comment, that whenever Ding thinks for twenty minutes he goes wrong, but if he plays very quickly, he makes the right moves! This maybe a little bit in jest, but it also shows how strong Ding's intuition is!

1. Nf3 d5 2. g3 g6 3. d4 Bg7 4. c4 c6 5. Bg2 Nf6 6. O-O O-O 7. Re1! 

A very interesting new idea and concept. White tries to remain flexible as regards the development of his knight. White challenges Black to find a useful other than the critical dxc4 for which Re1 is well suited to protect e4. If Black is going to take c4, then White's knight is better placed on b1 than c3 as potentially it will not be attacked by black's pawn pushes b5-b4.

7. Nc3 dxc4 8. h3 Na6 9. e4 b5 10. Qe2 Bb7 11. Rd1 Nc7 12. Bf4 Ne6 13. Be5 Qb6 14. d5 Nc5 15. Bd4 Nfd7 16. b4 cxb3 17. axb3 a5 18. Rac1 Rac8 19. Be3 Bxc3 20. Rxc3 cxd5 21. exd5 b4 22. Rcc1 Ba6 23. Qa2 a4 24. bxa4 b3 25. Qd2 Qd6 26. Bf4 Qf6 27. Bg5 Qd6 28. Bf4 Qf6 29. Bg5 ½-½ Romanishin,O (2585)-Anand,V (2715) PCA-Wch Candidates m2 New York 1994 (1) 

This game of Romanishin was mentioned by Leko in his commentary. It was interesting to see the gambit ideas being experimented historically.

 7... dxc4!? 8. e4 Bg4 9. Nbd2 c5!? 

A very nice idea, where Black is transforming the structure into a Benoni where he has exchanged his problem piece- light squared bishop.

 9... b5 10. h3 Bc8 is an idea, which Ding would have considered but of course would not have felt fully comfortable.

10. d5 e6 11. h3 Bxf3 12. Bxf3 exd5 13. exd5 Nbd7

 Black transformed a Neo Grunfeld into a Benoni

 14. Nxc4 b5 15. Na3! 

outwardly not a great square. But White holds back black's queenside majority and later on the knight gets activated via the c2-square. Anish Giri suggested that all this could have very well been part of Gukesh's preparation!

15....Qb6 16. Bf4 

 White's Bishop Pair make a formidable impression.

16... Rfe8 17. Qd2 Rad8 18. Nc2 Nf8 

Usually with a pawn on d5 for the opponent, this is a not a good square for the Knight was the observation of top players.

 18... Rxe1+ 19. Rxe1 Ne8 was suggested by Anish and seems more natural.

 19. b4! 

This was a typical idea in the Benoni, but as Magnus pointed out, allowing a protected passed pawn in the centre is a committal decision and was not easy at all.

 19...c4 20. Be3 Qa6 21. Bd4 Rxe1+ 

 21... N8d7 is apparently Black's best move, but it is a very hard decision to admit one's mistake and come back to the earlier square! 22. Qf4 Qa4! as pointed out by Anish gives Black counterplay

 22. Rxe1 Qxa2!?

For the World Champion a big-time trouble was looming, and he decided to force things through trusting his intuition and grab the pawn on a2!

 23. Ra1 Qb3 

Apparently, the complex 23... c3! retains a sort of balance as various commentators pointed out. But with limited time on the clock, this was extremely hard to come up with.

 24. Ra3 Qb1+ 25. Kg2 Rd7 26. Ra5!

Gukesh took 20 minutes for this wonderful move, which apparently the computer was screaming from the word go! The position was very complex and Gukesh came up with objectively the best move, even though it was far from obvious for various commentators.

I was thinking about 26. Bc3 threatening Ra1 and Nd4 trapping the Queen, but Black has 26... Ne6 and after which it is not easy to make progress.

 26... Qb3 27. Ra3 Qb1 28. Ra5 

Gukesh goes back to Ra5 after repeating once, but Nakamura criticized this decision. He was not gaining time as there was no increment, but he let Ding come one move closer to move 40

28... Qb3 29. Rxb5 Qd3  

30. Qf4!?

The most natural move and a good one for sure! But White has a much stronger even a decisive continuation at his disposal it seems with

 30. Be3! Qxd2 31. Bxd2 White is not even ahead in material. But his pieces are dominating. He has the Bishop Pair and more importantly than that, Black's pieces, especially the Knight on f8 is totally passive and lacking any prospects. Black's pawn on a7 particularly and to a certain extent the one on c4 also can be a weakness, but White's pawn on d5 and b4 are not weaknesses at all. But playing such non forcing moves, when one side is having a big initiative and looking for a killer blow, is extremely difficult in human chess!

30... Qxc2 31. Bxf6 Qf5 32. Qxf5?! 

A very important and instructive moment pointed out by Anish. Here, more than the doubled pawns, the mobility of the pieces is the priority. The knight on f8 needs a chance to come into the game, and this is helped by doubling the f-pawns and freeing up the g6-square for the knight.

 32. Bg5! Qxf4 33. Bxf4 c3 34. Rc5 with advantage

 32... gxf5 33. Bxg7?! 

33. Bg5! 

Even here White had to maintain the bishop pair, and the automatic capture on g7 was criticized by all commentators. But, in the heat of the battle and in time trouble, some moves are done automatically, and this might be one such moment. 

33... Kxg7 34. Rc5 Ng6 35. Rxc4 Ne5 36. Rd4 Nc6 

The knight has become totally activated and even though Black is down a pawn, there is a definite improvement from his earlier situation.

37. Rf4

 37. Rd2! Nxb4 38. d6 Kf6 39. Be2! was a suggestion of Maurice Ashley at the press conference. Clearly this was also far from obvious and missed by players too. Black is not able to come closer with his King to e6 or e5 and attack the pawn on d6. And White's Bishop aims for b5/a4 squares from where it supports its own advanced passed pawn and block's black's a-passed pawn on the same diagonal.

Perhaps Black should still be looking for a blockade on d6 with 37... Ne7!?   

37... Ne7 38. b5 Kf6 39. Rd4 h6 

Black should go for a blockade on d6-square. This is usually effective when dealing with a light square bishop of the opponent. This concept is seen in various games, one that comes to my mind is from the Kramnik-Grischuk match in 2011.

39... Nc8! 40. Be2 Nd6 

Here the Knight does a very effective job of blocking the d5-passed pawn, protecting the f5- weakness and attacking the b5-pawn as well!

 41. Rh4 Kg7 42. Ra4 Kf6 43. Ra6 Ke5=

 40. Kf1!

 White forces Ding to come up with a decision on the 40th move. And it did work, Gukesh was putting pressure on Ding, and he seemed to have cracked(almost!) here. 

40....Ke5? 

Clearly DIng had considered} 40... Nc8! but decided against in trying to be more active intuitively. 41. Rh4 Kg7! Anish pointed out the importance of stepping back to g7 with the King.42. Rc4 Nd6 43. Rc5 Kf6 and Black is faster with his king and holds the balance.

If Black instead plays 41... Kg6 42. Rc4 Nd6 43. Rc6! This pin gains the requisite time to improve White's king Kg7 44. Ke2 Nxb5 45. Kd3 Once the King joins the party as active fighting unit things get hard for Black.

 41. Rh4 Nxd5 42. Rxh6 Nc3 43. Rc6! Ne4 

 44. Ke1?!

 This move was criticized by Magnus, who even went on to call this move an eyesore! But Ding felt this was a very strong move on the part of Gukesh during the game. Just goes to show how complex this endgame is!

 The computer's recommendation to win was 44. h4! when White clearly does not fear the exchange of knight for a bishop. Nd2+ 45. Ke2 Nxf3 46. Kxf3 And White is winning this rook endgame. Well, once again one has to understand this during the game- not easy. 

A fascinating concept pointed out by Anish Giri was 44. Bxe4! fxe4 to go into a Rook endgame and even undoubling Black's pawns in the process. If the Queenside pawns are exchanged it would lead to a three versus two on the same flank and would most likely result in a draw. However, that is not so easy to achieve from Black's viewpoint since White's pawn on b5 is pretty advanced. Apart from that there are various scenarios in which White's active rook could cut off Black's king along the rank which might be a decisive factor in many a rook endgame. In the given case White carries on with a variation which shows lot of subtleties along the way. 45. Rc5+ ! Ke6 46. g4!

From the view of economy this is brilliant. White gets a square for his rook on f5, from where it will protect both the b5 pawn on f2 pawn. The other pawns are protected by pawns, which in turn are protected by the king! Rubinstein would have surely loved this! Apart from the above listed features, this move also disconnects Black's e and f pawns, but having said all this the variation goes on and wins in a very concrete fashion as showed by Anish. 

Rd3 47. Kg2 Rd2 48. Rc6+ Ke5 49. Rc7 Kf6 (49... Ke6 50. Rxa7 e3 51. Ra3 $18) 50. Kg3 Rd3+ 51. Kf4 Rxh3 52. Rxa7 Rf3+ 53. Kxe4 Rxf2 54. Kd5 Rf4 55. b6 Rxg4 56. b7 Rb4 57. Kc5! with the idea Ra6+ and Rb6 A typical idea known from similar Rook endgames.

 44... f6?!

 Objectively a mistake after twenty-two minutes of thought in the final time control! But as Carlsen put it, it's a fantastic practical move, Black moves the pawn away from the seventh rank where it can be a target for opponent's rook, and at the same time increases the attacking scope of his rook. Ding said he thought this position was lost and came up with this as a last resort fighting resource.

 44... Rd5! 45. Be2 Rd4 is suggested by the computer and is apparently fine for Black!

45. h4!? 

45. Rc8 to try and check from behind is a resource and trying to check opponent's activity by our own is a resource. But once again, not at all obvious!

 45... Rd3 46. Bd1? 

This crucial moment was where White gave up the advantage and let black eventually draw the game. Anish pointed out a win for White here with

46. Ke2! Rd5 47. b6 axb6 48. Rxb6 Even though the pawns are on the same flank, White has an outside h-passer and Black's king is not well placed in the center surprisingly! This means that White is perhaps winning this endgame.

46... Rb3 47. Bxe4 fxe4 48. Rc5+ Ke6 49. g4! Once again, the Rook on the fifth rank is super strong for White and this position is winning for White. This is probably due to too many weaknesses in Black's camp, namely a7 and e4 pawns and also White's h-passed pawn is a formidable force.

46... f4! 47. gxf4+ Kxf4 

Black fought ferociously keeping all his pieces including his king very active.

48. Bc2  

The natural 48. Rc4 does not win because of 48....Rh3! 49. f3 Rh1+ 50. Ke2 Rh2+!

is the line that Gukesh had missed earlier, as he pointed out.

48... Rd5 49. Rc4 f5 50. Rb4 

 50. Bd1 Rxb5!?  (50... Rc5 51. Rxc5 Nxc5 is also not easy for White to make progress.) 51. f3 Re5 52. Be2 Kg3 53. fxe4 Kxh4 54. exf5+ Kg5! would lead to Rook Bishop versus Rook position!

50... Kf3 51. Bd1+ Kg2 

In time trouble phase for both sides, Ding played very very actively. The king going deep into opponent's camp and maintaining coordination. Gukesh kept trying his every chance, but it did not prove enough to unsettle Ding.

52. Rb3 Re5 53. f4 Re7 54. Re3 Rh755. h5 Nf6 56. Re5 Nxh5 57. Rxf5 Ng3 58. Rf8 Rb7 59. Ba4 Kf3 60. f5 Kf4 61. f6 Ne4 62. Bc2 Nd6 63. Rd8 Ke5 64. Bb3 Nf7 65. Rd5+ Kxf6 66. Kd2 Rb6 67. Bc4 Rd6 68. Kc3 Rxd5 69. Bxd5 Nd6 70. Kb4 Nxb5 71. Kxb5 a6+ 72. Kxa6 1/2-1/2

A thorough slugfest where at some point Gukesh even ran down to having just 2 seconds on his clock in the final time control, but it ended in a draw with all material exhausted. Ding stands tall as the great wall of China! Will Gukesh be able to breach him or will he wear out trying! The second half is sure to be even more exciting!

Special Thanks to Anish, Magnus and Hikaru and Peter Leko whose comments have been included which helped in better understanding of the game.

Replayable version of the game analysed