[Event "FIDE Candidates Tournament"]
[Site "chess24.com"]
[Date "2020.03.19"]
[Round "3.1"]
[White "Ding, Liren"]
[Black "Caruana, Fabiano"]
[Result "1-0"]
[ECO "D17"]
[WhiteElo "2805"]
[BlackElo "2842"]
[Annotator "Aagaard"]
[PlyCount "117"]
[EventDate "2020.??.??"]
[WhiteTeam "China"]
[BlackTeam "United States"]
[WhiteTeamCountry "CHN"]
[BlackTeamCountry "USA"]
{[%evp 0,117,19,31,19,0,23,15,19,8,43,-1,-2,10,13,6,-13,-13,-8,32,50,69,79,75,
61,55,93,81,125,90,211,115,210,161,191,107,166,151,142,142,142,130,141,131,130,
112,138,134,74,129,150,151,130,128,117,110,96,100,100,87,114,86,127,112,128,
132,125,169,241,233,217,230,236,236,240,243,243,286,283,308,333,263,284,287,
323,356,384,350,326,326,326,362,369,359,530,509,482,482,492,521,521,579,559,
580,580,522,541,541,570,606,605,581,891,1188,1188,564,728,785,785,631] The big
game of the third round was obviously the first clash between the two
pre-tournament favourites. After Ding Liren had lost yesterday, the Chinese
rumour mill was that Ding had withdrawn from the tournament. But Ding was at
the board for the start of the third round. If he managed to win the game, he
would overtake Caurana on tie-break substantially and only be half a point
behind him. For this reason the opening choice was quite surprising from
Caruana. He chose to come with a brand new idea on move nine, forcing his
opponent to think, while feeling the pressure of not knowing when the computer
he was obviously playing with, would find his choices wanting. To me it just
seems overly risky. My best guess (and by no means the only scenario I can
think of) is that Caruana had decided on this risky decision before the
tournament and felt it was easier to go with his preparation than to change
course on this very surprising scenario for round 3...} 1. d4 d5 2. c4 c6 3.
Nf3 Nf6 4. Nc3 {Ding Liren said that he considered playing} (4. cxd5 {and make
a quick draw, but he decided that it was not his style. He cited the
encouragement of family and his coach as inspiration to go on.}) 4... dxc4 5.
a4 Bf5 6. Ne5 e6 7. f3 Bb4 8. Nxc4 (8. e4 Bxe4 9. fxe4 Nxe4 {was the mainline
from the 1980s onwards, but nowaways this is what they do.}) (8. Kf2 Nbd7 9.
Nxc4 O-O 10. e4 {1/2-1/2 (28) Zhao,Z (2537)-Smirnov,A (2194) Canberra 2013})
8... O-O 9. Kf2 $1 {This move looks inherently weird. But when you see the
options White has for e4 and g4 once he is out of the in, you can understand
it easily.} (9. e4 Nxe4 10. fxe4 Qh4+ 11. Kd2 Qxe4 12. Qf3 Qxd4+ 13. Ke1 {
has been played in some games at a lower level. Here Black has a serious
improvement with:} b5 $1 $146 {and it is clear that Black has the initiative.
The following computer line has little to do with what would actually be
played in a game.} 14. axb5 cxb5 15. Na5 Rc8 $15 {Black has the initiative.
This was a rather entertaining line with some forced properties:} 16. Bd2 Nd7
17. Nc6 Rxc6 18. Qxc6 Rd8 19. Qf3 Nc5 20. Qf4 Nb3 21. Qxd4 Rxd4 22. Be3 (22.
Rd1 Bc2) 22... Nxa1 23. Bxd4 Nc2+ 24. Kf2 Nxd4 25. Nxb5 Bc5 26. b4 Bxb4 27.
Nxd4 Bc5 28. Kf3 Bxd4 29. Bd3 Bxd3 30. Rd1 e5 31. Rxd3 f5 32. Rb3 e4+ 33. Kf4
g6 34. g4 {and White will make a draw eventually. But yes, Black could have
tried a lot of different things on the way.}) {[#] Ding was certainly
wondering what was coming for him here, as White is considered to be better in
these lines. Caruana did not disappoint.} 9... e5 $5 $146 {This is Caruana's
idea. A pawn sacrifice.} ({The main line seems to revolve around the following,
although it appears to be entirely unappealing to me.} 9... c5 10. e4 Bg6 (
10... Bxc3 11. bxc3 Bxe4 12. fxe4 Nxe4+ 13. Kg1 Nxc3 14. Qe1 cxd4 {Black has
four pawns for the piece. But this is not a position that suggests to me that
an endgame will happen. Best play may very well go like this:} 15. h4 Nc6 16.
Rh3 (16. h5 $5 $16) 16... Qd5 17. Bb2 Ne4 18. Ba3 Rfe8 19. Rb1 Nc3 {1-0 (40)
Olivares,F (2329)-Hernandez,C (2258) Bogota 2015} 20. Rxb7 Rab8 21. Rc7 $16)
11. Na2 (11. Be3 $2 cxd4 12. Qxd4 {was played in Georgiadis – Shankland,
Biel 2019. Here Black could have improved a lot on the game with:} Qe7 $1 $146
13. Ne5 Rd8 14. Qc4 Nbd7 15. Nxd7 Nxd7 $17) 11... Nc6 12. Nxb4 Nxb4 {has been
played a number of times. But here I think we are talking about a refutation.}
13. Bd2 $3 Qxd4+ (13... Bxe4 14. fxe4 Nxe4+ 15. Kg1 Qxd4+ 16. Be3 Qd5 17. h4
$16) (13... Qc7 14. Bxb4 cxb4 15. g3 $16) 14. Be3 Qxd1 (14... Nxe4+ 15. fxe4
Qxe4 16. Qf3 Qc2+ 17. Kg3 Be4 18. Qe2 Rad8 19. Qxc2 Nxc2 20. Rc1 Nxe3 21. Nxe3
f5 22. Nc4 $14) 15. Rxd1 b6 {Black has an extra pawn, but beyond this there is
nothing appealing about his position.} 16. a5 (16. h4 $5 $146 h5 17. Ne5 Bh7
18. Bb5 $16) 16... Rfb8 17. Bf4 Rb7 {Kuljasevic – Vocaturo, Le Castella 2018.
} 18. h4 $146 h5 19. a6 Rd7 20. Be2 {White is close to winning.}) (9... Bg6 10.
e4 Nbd7 11. Be3 $14 {has been played in about a dozen games, but it just
inherently sad for Black. Still it could be the objectively best choice:} c5
12. dxc5 Bxc5 13. Bxc5 Nxc5 {was Zhao – Smirnov, Canberra 2013. The obvious
improvement is} 14. Qxd8 $1 $146 Rfxd8 15. Be2 Nb3 16. Rad1 $16 {The bishop on
g6 looks very sad.}) 10. Nxe5 $1 {Ding decided to call Caruana's bluff. You
can understand his mindset easily. With 0/2 he has to go for it and he knows
that White is probably better, as he has analysed this position with an engine
at some point. But of course the engine is short sighted and at times wrong.
Especially know when we have three strong engines that disagree all the time!}
(10. dxe5 Nfd7 11. Bf4 Qh4+ 12. g3 Qe7 13. Kg2 Be6 14. e4 Bxc4 15. Bxc4 Nxe5 {
Black is very close to equalising.}) 10... Bc2 11. Qd2 c5 12. d5 $1 {Ding
plays very principled.} ({Safer was} 12. e3 Bg6 13. d5 {, where Black has a
slightly worse, but overall reasonable position after} Nbd7 14. Nc4 Nb6 {
White is still struggling to get fully developed. Still I would prefer to have
the extra pawn here. Black is lacking in pawn breaks.}) 12... Bb3 (12... Be4 {
is refuted by a few outlandish moves.} 13. g4 Re8 14. Nxf7 Kxf7 15. g5 Ng4+ 16.
Kg3 $1 Bxf3 17. Qf4+ Kg8 18. exf3 Ne5 19. Bb5 {and Black does not have any
compensation for the pawn.}) 13. e4 Re8 (13... Nbd7 14. Nxd7 Nxd7 15. Qf4 {
looks easily better for White. Black needs counterplay.}) 14. Qf4 c4 $1 {
Opening up the position.} 15. Nxc4 $1 ({We can certainly think that Caruana
were hoping for something like} 15. Bxc4 Bd6 16. Qe3 Nbd7 17. Nxd7 (17. Bxb3 $6
Bc5 18. Nc4 {is adventurous, but after} Rc8 $1 {Black is going to be better.})
17... Nxd7 18. Qe2 Bxc4 19. Qxc4 Qh4+ 20. Kf1 Rac8 {, where White is perhaps
not really worse, but has to defend with about an hour less on the clock.})
15... Nbd7 16. Be3 Nf8 $5 ({Caruana was still in preparation here. An
alternative existed and was probably very hard to refute.} 16... Bxc4 17. Bxc4
Ne5 18. Be2 (18. Bb5 a6 {gives Black a useful tempo.}) 18... Ng6 19. Qf5 Rc8
20. Rac1 Ba5 {This is the difficult move to refute and maybe what Caruana
would have tried in practice. There are a lot of things for White to be
dubious about.} ({The endgame with two rooks for the queen after} 20... Bc5 21.
Bxc5 Rxc5 22. Nb5 Re5 23. Rxc5 Rxf5 24. exf5 Nf8 25. Rd1 {, looks very
promising for White and entirely risk free.} a6 26. Nc7 (26. Nc3 $6 {I am less
happy with. After} Qb6 27. b4 N8d7 28. a5 Qa7 29. Na4 Nxc5 30. bxc5 Qb8 31. d6
{, which seems to happen almost by force, Black can give up the knight with:}
Qc8 32. Kf1 h5 33. Nb6 Qxc5 34. d7 Nxd7 35. Rxd7 Qxa5 36. Nc4 Qb5 {. This
endgame is of course better for White, but it is also a draw.}) 26... Qd6 27.
Rc4 {White is better, but a lot of play remains.}) 21. Nb5 Rxc1 22. Rxc1 Ne7
23. Qf4 Nexd5 24. exd5 Nxd5 25. Rd1 $1 Rxe3 26. Qc4 Re5 27. f4 Rf5 28. Bf3 Qb6+
29. Rd4 $1 Ne7 30. b4 $18 Rxb5 31. axb5 Bxb4 32. Kf1 Nf5 33. Qc8+ Bf8 34. Rd8
Qxb5+ 35. Be2 Qb1+ 36. Bd1 Qb5+ 37. Kg1 Qb6+ 38. Kh1 Qb4 39. Rxf8+ Qxf8 40.
Qxf5 {and White has a technical win.}) 17. Bd4 Ng6 $2 {This looks like a basic
mistake. Caruana played all the previous moves after move 8 on less than 40
seconds. I honestly think that he messed up his preparation here. It happens
frequently for top players. In the last candidates, Karjakin mixed up two
lines in his white game against Aronian and lost more or less right out of the
opening, even if the game dragged on for a bit. And in the 2018 match in New
York, Carlsen forgot to insert an intermediate move in a game, leaving him
quickly in a lost position, as he continued playing his "analysis". At this
point I believe that Caruana had planned to play} (17... Rc8 18. Ne3 (18. Bxf6
gxf6 19. Ne3 Ng6 20. Qh6 Bf8 21. Qh3 Bc5 22. Re1 Qb6 {is objectively equal,
but who would want to be White in such a position?}) 18... Ng6 {After the
inclusion, this makes a lot of sense.} 19. Qg3 Bxc3 20. bxc3 Rxe4 $1 {, when
White seems to be better, but not that much and with plenty of difficulties
available for White.} 21. Kg1 $1 (21. Rb1 {leads to very double edged play.}
Rxe3 (21... Rxd4 22. cxd4 Rc3 23. Re1 Nh5 24. Qg4 Nf6) 22. Bxe3 Rxc3 23. Re1
Nxd5 24. Bxa7 b6 {with very sharp play.}) 21... Rh4 22. d6 $5 (22. h3 Nxd5 {
should be about equal.}) 22... Rf4 $1 23. Rb1 Be6 {Is White better? Stockfish
thinks so, but the other engines do not really. And he still has to get here
and make a lot of moves on the way.}) 18. Qf5 $2 {Ding rewards Caruana for his
enterprise – by not punishing him for his mistake.} (18. Qc1 $1 {would have
refuted Black's play. We can easily see why the knight should not be allowed
to stay at c4.} Bxc4 (18... Rc8 19. Nd2 {wins a piece. After something like}
Bxd5 20. exd5 Nxd5 21. Nb3 $1 {there is no meaningful compensation.}) 19. Bxc4
Rc8 {White hs a wide choice here. The simplest is not the critical looking 20.
Bb5, which does win with accurate play, but} 20. Ba2 {, when Black has nothing
meaningful besides sacrificing all the pieces:} Nxe4+ 21. fxe4 Rxe4 22. Nxe4
Rxc1 23. Raxc1 Qh4+ 24. Ke3 Qh6+ 25. Ke2 $1 {The most accurate.} Qh5+ 26. Kd3
Nf4+ 27. Kc4 a5 28. Nc3 {White is completely winning.}) 18... Bxc4 {Caruana
played this so quickly (6 minutes) you could be confused and believe he was
still in preparation. As said, I don't think he was.} 19. Bxc4 Qc7 20. Be2 ({
Another important line goes:} 20. b3 Bc5 21. Rhd1 Re5 $1 22. Qh3 Nf4 23. Qg3
N4h5 {with a repetition of moves after} 24. Qh4 g5 25. Qh3 Nf4 26. Qg3 N4h5 {
with a draw.}) 20... Bc5 21. Bxc5 Qxc5+ {The accurate and most natural way to
play. The point being that after} (21... Re5 22. Qh3 Nf4 23. Qh4 g5 {White has
the extra option} 24. Qxf4 $1 gxf4 25. Bd4 Rg5 (25... a6 $5 26. Rhd1) 26. Rhd1
{With great compensation for the queen.} Qe7 27. Kg1 Nd7 28. Bf1 $14 {is
probably quite a bit better for White. Black will struggle to create active
play, while White will have no such difficulties.}) 22. Kf1 h6 $4 ({As losers
do not seem to have to attend the press conferences at the Candidates (a big
loss), we have no explanation for why Caruana did not play} 22... Re5 23. Qh3
Qb4 $1 {, when he would have plenty of counterplay. The main reasons I can
come up with is a) he missed it, and b) he though he was OK in the game. The
most likely explanation is a combination of the two. Play should continue:} 24.
Rb1 Nf4 25. Qg3 N6h5 26. Qf2 f5 $1 27. Rd1 $1 (27. g3 fxe4 $1 28. gxf4 Nxf4 29.
Rd1 exf3 {is very dangerous for White. After} 30. Bb5 Ng2 31. Qd4 Re1+ 32. Kf2
Qxd4+ 33. Rxd4 Rxh1 34. d6 {White has enough counterplay, but this is not the
way you would want to go.}) 27... Qxb2 28. Qd4 Nxe2 29. Nxe2 Qxd4 30. Rxd4 Nf6
31. exf5 Rae8 32. Rd2 Rxf5 33. d6 Rd8 {The draw is on the horizon.}) 23. Rd1
Qb6 (23... Re5 24. Qh3 Nf4 25. Qg3 {leads nowhere now.}) 24. Rd2 Qe3 25. Rc2 a6
26. Qh3 b5 $1 {Black has to create counterplay. If White is able to
consolidate, it will be a matter of technique.} 27. Qg3 b4 $4 {The third
horrific move of the game.} (27... Rec8 $1 {would have kept the tension. It is
not obvious that Ding would have chosen the best continuation here, as it
would give up a piece:} 28. Qf2 $1 (28. Qe1 b4 29. Nd1 Qb3 30. Rd2 Rc1 {
would give Black some real counterplay. White may be better, but he would have
real practical problems to solve.}) 28... Qxf2+ 29. Kxf2 b4 30. Ke3 Nd7 31.
Rhc1 $16 {White has a big advantage with a bishop against two knights and the
powerful passed pawns.}) 28. Nd1 Qb3 29. Rd2 Qxa4 30. Qf2 $1 {White is on the
verge of being coordinated. Black has no chances from here on and could have
resigned, if he had not seen miracles happen from time to time.} Qd7 (30...
Rac8 {is too late now. After} 31. Qd4 Rc1 32. Kf2 Nf4 33. Re1 {White is fully
coordinated.}) 31. g3 Qh3+ 32. Kg1 a5 33. Qd4 Nh5 34. Nf2 Qd7 35. f4 {[#]}
Nhxf4 {Caruana tried to muddle things up, but with no success.} 36. gxf4 Nxf4
37. Kf1 Qd6 38. Rg1 f6 39. Bb5 Re7 40. b3 Rf8 41. Rc2 Ne6 42. Qe3 Ng5 43. h4
Nf7 44. Rc6 Qb8 45. Qc5 Qd8 46. Rxf6 Kh8 47. Rf5 Rc7 48. Bc6 a4 49. bxa4 b3 50.
Rg3 b2 51. Rb3 Nd6 52. Rxf8+ Qxf8 53. Rxb2 Qf4 54. Rb8+ Kh7 55. Qc2 Re7 56. e5+
Nf5 57. Qe4 Qc1+ 58. Kg2 g6 59. Rb3 1-0