[Event "World Championship"]
[Site "Dubai"]
[Date "2021.12.10"]
[Round "11"]
[White "Nepomniachtchi, Ian"]
[Black "Carlsen, Magnus"]
[Result "0-1"]
[ECO "C54"]
[WhiteElo "2782"]
[BlackElo "2856"]
[Annotator "Monokroussos,Dennis"]
[PlyCount "98"]
[EventDate "2021.11.27"]
[EventCountry "UAE"]
1. e4 {After an English in game 9, normal service is restored.} e5 2. Nf3 Nc6
3. Bc4 {A match novelty, though this is one of the most popular openings in
all of chess over the past ten years, especially at the super-GM level.} Nf6 ({
I guess Carlsen's 3...Nf6 was played to sidestep the trendy} 3... Bc5 4. c3 (4.
d3 Nf6 {transposes to the game.}) 4... Nf6 5. d4 exd4 6. e5 d5 7. Bb5 Ne4 8.
cxd4 Bb6 9. Nc3 O-O 10. Be3 {, which Carlsen has faced a bunch of times in the
last year or two - mainly against So, but also against Nepo himself 2-3 months
ago in the Norway Chess tournament in Stavanger.} Bg4 11. h3 Bh5 12. Qc2 Nxc3
13. bxc3 f6 14. exf6 Qxf6 15. Be2 Na5 16. O-O Bxf3 17. Bxf3 c6 18. Rae1 Nc4 19.
Bc1 Qg6 20. Qd1 Rae8 21. Rxe8 Qxe8 22. Be2 Qe6 23. Bxc4 dxc4 24. Re1 Qf7 25.
Qe2 Ba5 26. Qe3 h6 27. Ba3 Rd8 28. h4 Kh7 29. Re2 Qf5 30. Qc1 Rd5 31. f3 b5 32.
Re4 Bb6 33. Be7 {1/2-1/2 (33) Nepomniachtchi,I (2792)-Carlsen,M (2855)
Stavanger 2021}) 4. d3 ({Nepo keeps the game in Giuoco channels rather than
playing the "duffer's move", as Tarrasch called it many years ago.} 4. Ng5 d5
5. exd5 Na5 6. Bb5+ c6 7. dxc6 bxc6 8. Bd3 {is more popular than 8.Be2, and
was seen in another Carlsen game from earlier this year.} Nd5 9. Nf3 Bd6 10.
O-O O-O 11. Re1 f5 12. Nxe5 Qf6 13. Nf3 g5 14. c4 Nf4 15. Bf1 g4 16. d4 gxf3
17. Qxf3 Ne6 18. Qc3 Nb7 19. c5 Bc7 20. b4 a5 21. b5 {This was an internet
game between So and Carlsen (and several other games as well). Carlsen played
21...Qxd4, a mistake that would have left him in serious trouble had White
played 22.Bc4. Instead, Black should play} cxb5 22. Bxb5 a4 23. Bc4 Na5 24. Bd5
Rb8 25. Qe3 Re8 $11 {1/2-1/2 (113) Janik,I (2484)-Nasuta,G (2531) Warsaw 2021})
4... Bc5 5. c3 d6 6. O-O a5 {Preventing White's queenside expansion while
giving the Bc5 a retreat square.} ({There are lethal quantities of theory that
can be explored after} 6... O-O {,}) (6... a6 {,}) ({and} 6... h6 {. If you
start studying it now, you'll only be several thousand games behind by the
time you catch up with today's theory.}) 7. Re1 Ba7 {Relatively rare, but
often the lines in the Giuoco can transpose into one another. That's part of
its masochistic charm.} (7... O-O 8. Nbd2 Ba7 9. Nf1 Be6 10. Bb5 Ne7 {occurred
in a couple of Carlsen's games from earlier this year, against Duda in Wijk
aan Zee and yet another online battle with So.}) (7... h6 {is a possibility
here too.}) 8. Na3 {This makes sense with the bishop on a7, both to play Nb5
but also to go Nc2, Be3, and recapture on e3 with the knight. It's also
conceivable that the knight will go to c4 in some positions after either Bb5
or Bb3.} (8. h3 O-O 9. Na3 (9. a4 h6 10. Nbd2 Be6 11. Bxe6 fxe6 12. Nc4 Nh5 {
was fine for Black in Firouzja,A (2754)-Nepomniachtchi,I (2792) from 2-3
months ago in Stavanger 2021, even if the youngster eventually prevailed,
winning on move 57.}) 9... h6 10. Nb5 Bb6 11. Be3 Bxe3 12. Rxe3 Re8 {was
played in a blitz game last December between Nepomniachtchi and Grischuk.} 13.
d4 $6 {was an unwise pawn sac that didn't work out.} exd4 14. cxd4 Rxe4 $15 {
0-1 (59) Nepomniachtchi,I (2784)-Grischuk,A (2777) chess24.com INT 2020}) (8.
Nbd2 O-O {is (relatively speaking) a return to theory, including the two
Carlsen games (vs. So and Duda) given above.}) 8... h6 (8... O-O {allows the
pin on g5. It's not fatal, but it is annoying and sensible to prevent.} 9. Bg5
h6 10. Bh4 g5 11. Bg3 Nh7 12. Nd2 (12. Nc2 {may improve.}) 12... g4 13. Nb5 Bb6
14. Bb3 a4 15. Bc2 h5 16. h4 $11 {1-0 (58) Vachier Lagrave,M (2778)-Ding,L
(2816) Shenzhen 2018}) 9. Nc2 {After almost 18 minutes.} (9. Nb5 $146 Bb6 10.
Be3 Bxe3 11. Rxe3 {was another attempt to make sense of 9.Na3.}) 9... O-O 10.
Be3 $146 Bxe3 11. Nxe3 {Transposing to one very high-level game.} Re8 $146 (
11... Be6 {looks fine. Caruana lost a blitz game to MVL from here, but not
because of this move.} 12. Bb5 (12. Qc2) (12. b3) (12. Bb3) 12... Ne7 13. Ba4
Ng6 14. Bc2 {1-0 (54) Vachier Lagrave,M (2789)-Caruana,F (2816) Paris 2018
(blitz)} c6 $11 (14... b5 $11)) 12. a4 {Transposing to two earlier games. It's
hard to escape theory for good in the Giuoco.} (12. Qb3 a4 $1 13. Bxf7+ $4 (13.
Qc2 Bd7 $1 $11) 13... Kf8 14. Qc4 Na5 $19) 12... Be6 13. Bxe6 $146 {Another
novelty...and another transposition. Nepo played this immediately.} (13. Bb5 {
is the only way to play for an advantage. It isn't much, but the text gives
White nothing.} Bd7 14. Qb3 Ne7 $146 (14... Qc8 {was played in a couple of
earlier games, and now} 15. Nd2 $14 {retains a small edge. White's pieces are
slightly more active, and he'll be able to achieve d4 soon while Black will
have a harder time achieving ...d5.}) 15. Bc4 Rf8 16. Qc2 {followed by d4
keeps a mild initiative.} (16. Qxb7 $4 Rb8 17. Qa6 Rb6 18. Qxa5 Nc6 $19)) 13...
Rxe6 14. Qb3 b6 15. Rad1 $146 {The final novelty, at last.} (15. Nd5 Ne7 16. c4
Nd7 $1 {followed by ...Nc6 and ...Nc5 is at least equal for Black.} (16... Ng6
$6 17. d4 $14 {1-0 (41) Ostrovskiy,A (2439)-Nagarajan,P (2206) Burlingame 2019}
)) 15... Ne7 16. h3 Qd7 {Building for ...d5.} ({The immediate} 16... d5 $4 {
is a blunder.} 17. exd5 Nexd5 18. Nxd5 Nxd5 19. Rxe5 $18 {Not every "Marshall"
gives Black good compensation.}) 17. Nh2 Rd8 $11 18. Nhg4 Nxg4 19. hxg4 d5 {
Now everything is ready for this advance.} 20. d4 (20. exd5 Nxd5 21. Nxd5 Qxd5
22. Qxd5 Rxd5 $11) 20... exd4 ({Were it not for Carlsen's resource on move 23,
he'd have had to choose} 20... dxe4 {, which is also good enough to retain
equality.} 21. dxe5 Qe8 22. Qc4 Rxe5 23. Qxc7 Nd5 24. Nxd5 Rexd5 25. Rxd5 Rxd5
26. Qxb6 Qxa4 $11) 21. exd5 Re4 $1 {The only move.} (21... Nxd5 $4 22. Rxd4 c6
(22... c5 23. Rd2 Rd6 24. Red1 $18) 23. c4 $18) 22. Qc2 $1 (22. Rxd4 Rxd4 23.
cxd4 Nxd5 24. Nf5 c6 25. Re5 Re8 26. Qg3 f6 27. Rxe8+ Qxe8 $11) (22. cxd4 Nxd5
23. Nxd5 Rxe1+ 24. Rxe1 Qxd5 25. Re8+ Kh7 26. Qc2+ g6 $11) 22... Rf4 $1 (22...
d3 $2 23. Qxd3 Rxa4 24. c4 $18 {is terrible for Black thanks to White's
beautiful center and especially Black's trapped rook on a4.}) (22... Re5 $2 23.
Rxd4 {gives White an extra pawn and the better position, because the attempt
to regain it makes things worse:} Nxd5 $4 24. c4 Nxe3 25. Rxd7 $18) 23. g3 $4 {
After nine minutes, Nepo makes yet another blunder in the match, and this one
finishes it. Did he really think that Carlsen had blundered?} (23. Rxd4 Rxd4
24. cxd4 Nxd5 25. Nxd5 Qxd5 26. Qxc7 Qxd4 $11 {is a draw once the players
reach move 40.}) 23... dxe3 $19 {Forced, but winning - and the win is obvious.}
24. gxf4 Qxg4+ {Apparently, and unbelievably, Nepo overlooked this move. (At
least that was what Anna Muzychuk, who was commenting on the FIDE broadcast
with Viswanathan Anand, thought she overheard Nepo say immediately after the
game.)} 25. Kf1 Qh3+ 26. Kg1 (26. Ke2 exf2 27. Rf1 (27. Kxf2 $4 Qh2+ $19) 27...
Nf5 $19) 26... Nf5 $2 {If Carlsen had failed to win the game, this move would
deserve two question marks. Black is still winning, but he could have forced
resignation in two or three moves.} (26... exf2+ $1 27. Qxf2 (27. Kxf2 $2 Qh2+
$19) 27... Rd6 $1 $19 {and there's no defense to 28...Rg6(+). Black will have
a huge material advantage and an ongoing attack.}) 27. d6 $1 {Stopping ...
Rd6-g6.} Nh4 {Here and on move 29, Black had options that were "better" from
the computer's perspective, but Carlsen understood what he was doing and found
a way to win that made him comfortable.} 28. fxe3 Qg3+ ({To avoid:} 28... Nf3+
$2 29. Kf2 Nxe1 $2 {would give away almost all of the advantage.} (29... Qh2+
$4 {is even worse:} 30. Kxf3 Qxc2 31. dxc7 Rxd1 32. c8=Q+ $15) ({Fortunately
for Black,} 29... Rxd6 $142 $1 {would maintain a winning advantage.} 30. Rxd6
Qh2+ 31. Kxf3 Qxc2 $19) 30. dxc7 $1 Qf3+ 31. Kg1 Re8 32. Rxe1 $17) 29. Kf1 Nf3
{Threatening ...Qh3+/Qg1+ followed by ...Qg2+ and ...Qxc2.} 30. Qf2 ({White
can't avoid the coming endgame with the hyper-greedy} 30. Re2 $2 {because of}
Qg1#) ({or with} 30. dxc7 $2 {thanks to} Qg1+ 31. Ke2 Qg2#) (30. Qg2 Qxg2+ 31.
Kxg2 Nxe1+ {transposes to the game.} 32. Rxe1 Rxd6 $19) 30... Qh3+ 31. Qg2
Qxg2+ 32. Kxg2 Nxe1+ 33. Rxe1 Rxd6 {It looks as if White has made tremendous
progress since his blunder, but appearances can be deceiving. The combination
of Black's extra pawn and the unstoppable penetration of Black's rook mean
that White's situation is hopeless.} 34. Kf3 (34. Re2 Rd1 $19) 34... Rd2 35.
Rb1 g6 36. b4 axb4 37. Rxb4 (37. cxb4 Ra2) 37... Ra2 {If White tries to sit,
Black will bring his king into the game and start nudging his h-pawn forward.
White has no chance to save the game in that scenario, so he goes for
counterplay.} 38. Ke4 $1 h5 39. Kd5 Rc2 40. Rb3 h4 41. Kc6 h3 42. Kxc7 h2 43.
Rb1 Rxc3+ (43... f6 $1 {followed by ...g5 also wins.}) 44. Kxb6 Rb3+ 45. Rxb3
h1=Q 46. a5 {If White gets the pawn to a7, it will be difficult if not
impossible for Black to win. Carlsen makes it look easy from here, but that's
thanks to some precise moves.} Qe4 $1 {Threatening ...Qe6+.} 47. Ka7 Qe7+ 48.
Ka8 Kg7 $1 {Now Black can safely give checks on the 8th rank.} 49. Rb6 (49. a6
Qe6 {wins the a-pawn, as} 50. Ra3 {loses the rook to} Qe8+ 51. Kb7 Qe7+ {
and 52...Qxa3.}) 49... Qc5 $1 {A nice move to finish the match.} (49... Qc5 $1
{This forces} 50. Ra6 {, when White is completely immobilized. Black can go
for ...f6 and ...g5 immediately, or take on e3 first.} Qb5 {is perhaps the
more human way, keeping White immobilized.} 51. Ra7 (51. Ka7 f5 $19) 51... Kf6
52. a6 (52. e4 Qe8+ 53. Kb7 Qxe4+) 52... Qb6 53. e4 Ke6 54. f5+ gxf5 55. exf5+
Kf6 $19) 0-1
