'Recent Elite Endgames'

Sun, 2015-07-05 08:39 -- IM Max Illingworth
[Event "3rd Norway Chess 2015"]
[Site "Stavanger NOR"]
[Date "2015.06.25"]
[Round "9.3"]
[White "Hammer, J."]
[Black "Carlsen, M."]
[Result "1-0"]
[ECO "D30"]
[WhiteElo "2677"]
[BlackElo "2876"]
[Annotator "Illingworth,Max"]
[PlyCount "67"]
[EventDate "2015.06.16"]
[EventRounds "9"]
[EventCountry "NOR"]
[EventCategory "22"]
[Source "Mark Crowther"]
[SourceDate "2015.06.29"]

{For this post I'll repeat a previous theme, namely the most instructive
recent endgames. First up is the all-Norway matchup from the Stavanger
super-tournament (won by Topalov), which will give us a good opportunity to
discuss the unique characteristics of opposite-coloured bishop positions
without queens.} 1. d4 d5 2. c4 e6 3. Nf3 a6 4. Bg5 f6 5. Bd2 dxc4 6. e3 Nc6 7.
Bxc4 Bd6 8. e4 Nge7 9. O-O O-O 10. Qb3 Kh8 11. Bxe6 Bxh2+ 12. Kxh2 Qd6+ 13. Kh1
Bxe6 14. d5 Bg8 15. Qa3 Qxa3 16. Nxa3 {Despite a frankly absurd opening choice,
Carlsen has made it to the following endgame where he is certainly worse 
(White dominates the centre and has the better bishop - always important in
opposite-coloured bishop endgames), but can probably defend with correct play.}
Na7 (16... Nb8 $5 {was probably better, intending normal development with ...
Nd7 and then to challenge White's centre with ...c6, such as after} 17. Bb4 Re8
18. Rac1 c6 {. Black's dream would be to provoke d6 as then he can fully
blockade White's pawns on the light squares, but in any case Black's position
is remarkably robust. White's best continuation is probably} 19. Nc4 $1 cxd5
20. Nb6 dxe4 21. Nxa8 Nec6 22. Bc3 exf3 23. Rfe1 Ne5 24. Bxe5 fxe5 25. Rc7 Nc6
26. Nb6 Re6 27. gxf3 Rh6+ 28. Kg2 Nd4 29. Rxb7 Rg6+ 30. Kf1 Nxf3 31. Re3 e4 32.
Ke2 Nd4+ 33. Kd1 Nf5 $14 {and with the pawns starting to be liquidated, Black
should be able to hold the draw with best play, though it won't be a piece of
cake. Obviously there were plenty of alternatives for both sides, but I should
leave some room for your own investigations.}) 17. Bb4 (17. Rac1 $1 {was more
accurate, and only after} c6 18. d6 $1 Nec8 19. Bf4 {is much more effective
than in the other variation as Black's knights are lacking in coordination.})
17... Rfe8 18. Rac1 {This also looks good, but there are some subtle
differences.} Rac8 $2 {In opposite-coloured bishop positions it is of the
utmost importance to keep your bishop active - actually it can be even more
important than the material balance.} (18... Ng6 {is very natural but has a
tactical flaw:} 19. Rxc7 (19. Rfe1 c6 {is playable for Black, who in any case
brings his bishop into play by destroying the d5-pawn.}) 19... Rxe4 20. Bc5 $1
Bxd5 21. Rd1 Bg8 22. Bxa7 Rxa7 23. Rd8 Re6 $1 24. Nd4 Re7 25. Rcc8 Re1+ 26. Kh2
Ne7 27. Rc7 $16 {and despite being a pawn up, Black is all bound up - his
bishop and a7-rook are completely stuck.}) (18... a5 $1 {is the big difference,
but admittedly this isn't such an easy move to find.} 19. Bxa5 f5 {Since the
e7-knight is no longer under attack, Black has time to exchange off White's
centre. After} 20. exf5 b6 (20... Bxd5 21. Rxc7 Nxf5 22. Rd1 Bxa2 23. Rxb7 {
would just be a draw if it were only the opposite-coloured bishops on the
board, but the presence of other pieces makes things a lot more complex and
White has good practical chances to win after} Nc6 24. Bc3 $16) 21. Bd2 Nxf5
22. Rxc7 Bxd5 $14 {and while there's some suffering to go, Black's bishop is
now no worse than its counterpart and with White's knight on a3 a bit out of
play, Black has decent chances to draw the game. Furthermore White has some
problems turning his queenside majority into a passed pawn.}) 19. Nd4 $1 {Now
White has full control of the position; all of Black's minor pieces look silly
and it's hard to suggest a good post for either knight.} f5 {Carlsen realises
he is in trouble and quite prudently decides to complicate.} (19... Ng6 20. f3
Ne5 21. Rfd1 $16 {is given by the engine, but very few would be willing to
defend so passively and I have little faith in Black's survival odds here.})
20. f3 fxe4 21. fxe4 {White has preserved his strong centre, but Black is
counting on it for counterplay.} Ng6 22. Nf5 $2 {White hopes to transform his
centre into an attack on the king. It's true, the initiative can be very
valuable in opposite-coloured bishop positions as the defender is unable to
properly challenge on the colour complex of the attacker's bishop, but here it
was better to preserve White's long-term advantages with} (22. Rfe1 Ne5 23. Bc3
$1 (23. Rcd1 $2 c5 $1 24. dxc6 Naxc6 $15 {is probably what put White off this
line.}) 23... Nd3 24. Nf5 {and the threat to g7 places Black in hot water.})
22... Rxe4 23. Bc3 Bxd5 $2 {If you had to pick someone to save a difficult
position for you, Carlsen would be the obvious choice, but June 25 wasn't his
day.} (23... h6 $1 24. Bxg7+ Kh7 {is fine for Black - there's no way to mate
his king and the d5-pawn will soon drop off. After} 25. Bd4 (25. Rcd1 $6 Rd8)
25... b6 26. d6 c5 $5 27. Bf6 Nc6 {Black can be fairly happy as the d-pawn
isn't going to queen safely, e.g.} 28. d7 Rf8 29. Nd6 Rf4 30. Ne8 Bxa2 31. Rxf4
Nxf4 {and the d7-pawn is tied up to defending the e8-knight, which in turn
doesn't have a brilliant square.}) 24. Bxg7+ Kg8 25. Bd4 $1 {The threats of
Nh6 and Bxa7 force Black to sacrifice the exchange, after which his position
is simply losing. There you have it - the opposite-coloured bishop is a mighty
attacking force!} Rxd4 (25... h5 26. Bxa7) (25... Bxa2 26. Nh6#) 26. Nxd4 Bxa2
27. b3 $6 {The obvious move, trapping the bishop, but} (27. Nc4 Bxc4 28. Rxc4 {
was stronger - in such an open position the rook is significantly better than
the knight. This advantage will only increase once Black's rook is exchanged.})
27... Rd8 $2 {After this Black is lost.} (27... c5 {was required, when the win
is by no means automatic. After} 28. Nf5 Bxb3 29. Nd6 Rc7 30. Rb1 c4 31. Rfc1
Rd7 32. Naxc4 Bxc4 33. Nxc4 Nc6 {the limited number of pawns should allow
Black to easily hold a draw. Even some positions with a rook against a knight
with one pawn each on the kingside are drawn.}) 28. Ne6 Rd2 29. Rc3 $1 {White
wants to play Rf3 and Rf8 with mate, and this is quite hard to stop. Of course,
in the endgame we shouldn't forget that two rooks can coordinate to make
mating threats.} Re2 30. Nf4 (30. Rf6 {and Rcf3 was more decisive.}) 30... Rb2
31. Nxg6 hxg6 32. Rxc7 Rxb3 (32... Nc6 33. Rxb7 Rxb3 34. Rxb3 Bxb3 35. Rf6 Ne5
36. Rxa6 {is winning for White, but at least he still has to show some
technique. In general, exchanging the knights will make White's task a lot
easier.}) 33. Rd1 Nc6 34. Rdd7 {Stop. Hammer time!} 1-0 
[Event "43rd GM 2015"]
[Site "Dortmund GER"]
[Date "2015.06.28"]
[Round "2.3"]
[White "Caruana, F."]
[Black "So, W."]
[Result "0-1"]
[ECO "B90"]
[WhiteElo "2805"]
[BlackElo "2778"]
[Annotator "Illingworth,Max"]
[PlyCount "138"]
[EventDate "2015.06.27"]
[EventRounds "7"]
[EventCountry "GER"]
[EventCategory "19"]
[Source "Mark Crowther"]
[SourceDate "2015.06.29"]

{I really like this game for the way it illustrates how to exploit the
advantage of a rook and a couple of pawns for two minor pieces in the endgame.
} 1. Nf3 c5 2. e4 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. h3 e5 7. Nde2 b5 8. g4
b4 9. Nd5 Nxd5 10. exd5 h5 11. gxh5 Rxh5 12. a3 bxa3 13. Rxa3 Nd7 14. Bg2 Nf6
15. O-O Rb8 16. f4 Be7 17. c4 Qb6+ 18. Rf2 Rh4 19. Rc3 Bd7 20. b3 Bxh3 21. Bxh3
Ne4 22. Qe1 Nxc3 23. Qxc3 Qxb3 24. Qxb3 Rxb3 25. Bc8 e4 $2 {Objectively this
is a mistake, although the position is quite chaotic and it's quite natural
that So wanted to keep control.} (25... a5 {however was strongest; I suppose}
26. c5 {was the reason for avoiding this, but the engine gives a wacky
variation with} exf4 27. c6 Kd8 28. Ba6 Rg4+ 29. Kh2 Bf6 $17 {where
surprisingly White has no way to support his protected passed c6-pawn, and
therefore Black is much better with a slew of passed pawns.}) ({Bojkov gives} 
25... exf4 26. Bxf4 Ra3 {as a more pragmatic but still effective continuation.}
) 26. Bxa6 $2 {In such a sharp position, it's more important to keep the
initiative than grab material.} (26. Rg2 g6 27. Rg3 $1 {is totally
counter-intuitive, given that generally you should keep your rook on the board
when you have minor pieces vs. a rook, but the point is that exchanges will
safeguard the White king, and after} Rxg3+ 28. Nxg3 f5 29. Bxa6 {White is OK.})
26... Bd8 $1 {A very strong manoeuvre; ...Bb6 will set all kinds of issues for
White.} 27. c5 {An unfortunate necessity.} (27. Kg2 $2 Bb6 28. Rf1 Rg4+ 29. Kh2
f5 $19 {and White's king is far too exposed.}) 27... dxc5 28. Rg2 g6 29. Rg3 $1
{Now White admits his mistake, but the key difference now is that White had to
give up the c4-pawn and therefore Black has plenty of passers.} Rxg3+ 30. Nxg3
Rg4 31. Kh2 f5 32. Ne2 (32. Bb5+ Kf7 33. d6 {was called for, seeking
counterplay to distract Black from advancing his c- and e-pawns.}) 32... Rh4+ {
It's slightly puzzling that Black allowed White's king off the edge, and I'd
prefer} (32... Bf6) 33. Kg2 Bc7 34. Bc8 (34. Bb2 $1 {has been noted elsewhere
as a better defence; to be honest I don't see a way for Black to make progress
following} Ke7 35. Be5 Bd6 36. Bb5 {- if} Rh7 37. Bc4 Rf7 38. Nc3 {gives White
plenty of counterplay with Nb5 before Black can activate his rooks. But after
other continuations his rook would be unable to move off the h- and g-files
because of the Be5 and Kg2.}) 34... Kd8 35. Be6 Ke7 36. Be3 Bd6 37. Bf2 Rh8 $1
{Now that Black's rook is in the game again, Black is ready to advance his
c-pawn, and White is in big trouble.} 38. Be1 c4 39. Bc3 Ra8 40. Be5 {When
defending for a long time, it's natural to try something active, and yet this
might be the decisive mistake.} (40. Kf2 {brings the king to the defence, and
while the computer greatly prefers Black, it may still be within the drawing
zone if White places his king on d2 and sits on the position.}) 40... Ra2 41.
Kf1 Rd2 42. Bc3 Rd3 {Now that the rook is on d3, it supports both passed pawns
and can also attack the f4-pawn. White is unable to cover everything at once -
a typical technique in the endgame, of stretching the opponent's defences with
multiple weaknesses/passed pawns.} 43. Ba5 Ba3 (43... Ra3 $1 44. Bc3 Rb3 {and .
..Bxf4 or ...Bb4 would just win.}) 44. Bg8 Bd6 45. Be6 Bc5 46. Ke1 Rb3 47. Kd2
Rb2+ 48. Kd1 Bb4 49. Bxb4+ Rxb4 50. Kc2 Rb3 51. Nc3 e3 52. Bg8 Rb6 53. Ne2 Ra6
54. d6+ Rxd6 55. Bxc4 Kf6 56. Bd3 g5 57. fxg5+ Kxg5 58. Ng3 Rc6+ 59. Kd1 f4 60.
Nf1 Rb6 61. Nh2 Kh4 62. Ke1 Kg3 63. Nf1+ Kf3 64. Nh2+ Kg2 65. Nf1 Re6 66. Bc4
Re5 67. Ba6 Ra5 68. Bb7+ Kg1 69. Nxe3 Re5 0-1 
[Event "10th Edmonton GM 2015"]
[Site "Edmonton CAN"]
[Date "2015.06.25"]
[Round "6.1"]
[White "Harikrishna, P."]
[Black "Ivanchuk, V."]
[Result "1/2-1/2"]
[ECO "E63"]
[WhiteElo "2733"]
[BlackElo "2733"]
[Annotator "Illingworth,Max"]
[PlyCount "81"]
[EventDate "2015.06.20"]
[EventRounds "9"]
[EventCountry "CAN"]
[EventCategory "10"]
[Source "Mark Crowther"]
[SourceDate "2015.06.29"]

{Of course, we shouldn't neglect those games where one manages to defend a bad
position - it may be less entertaining, but it's a valuable skill to have.} 1.
d4 Nf6 2. Nf3 d6 3. g3 g6 4. Bg2 Bg7 5. O-O O-O 6. c4 Nc6 7. Nc3 a6 8. d5 Na5
9. b3 Bd7 10. Bd2 b5 11. cxb5 axb5 12. Nd4 b4 13. Ncb5 Qb8 14. a4 bxa3 15. Nxa3
Ng4 16. Nac2 Qb6 17. e3 Ne5 18. Nb4 Nb7 19. Nbc6 e6 20. Qc2 Rxa1 21. Rxa1 Re8
22. Rc1 Bxc6 23. Nxc6 exd5 24. Nxe5 Bxe5 25. Qc6 Qxc6 26. Rxc6 Rc8 27. Bxd5 {
This endgame looks fantastic for White, with a bishop pair advantage and a
weak pawn on c7 to target. But the drawing margin in chess is quite high.} Nd8
$1 {A good start, before White can play b4-b5-b6 and win the c7-pawn.} 28. Rc2
(28. Ra6 c6 29. Bg2 Kg7 30. f4 Bf6 {is also defensible for Black.}) 28... c5 $5
{Another strong move, as the d6-pawn is harder to target than the c7-pawn, and
furthermore the b3-pawn could be blockaded if White does not act fast.} 29. b4
({I think it was better to try} 29. f4 Bf6 30. Ra2 Ne6 31. Ra6 Rd8 32. Ra7 $16
{when Black is tied up defending the f7-pawn as well as against attacks on the
d6-pawn, and White has a very big advantage. He has various ways to improve
his position, starting with Kf2-f3, g4-g5, Ba5 and Ra6.}) 29... Ne6 30. f4 Bf6
31. Ra2 (31. Kf2 {was once again more patient and probably better. In general,
when you have such long-term advantages, one can afford a more strategic
approach.}) 31... cxb4 (31... Nc7 $5 32. Bc6 c4 {gives Black counterplay with
the passed c-pawn, but Ivanchuk's choice is simpler.}) 32. Bxb4 Be7 33. Ra7 Kf8
{The position is ugly for Black, but there is no way through his defences.
Especially with all the pawns on one side of the board now, White's bishop
pair aren't clearly better than the knight and bishop, and the one weakness on
d6 can be satisfactorily covered.} 34. e4 Rb8 35. Ba3 Nd4 36. e5 (36. Ra8 Rxa8
37. Bxa8 Nc2 {would also be a pretty easy draw.}) 36... Rb1+ 37. Kg2 Nb5 38.
Ra8+ Kg7 39. exd6 Bxd6 40. Bxd6 Nxd6 41. Ra2 {Draw agreed as the position is a
dead draw.} 1/2-1/2 
[Event "50th Capablanca Mem Elite"]
[Site "Havana CUB"]
[Date "2015.06.23"]
[Round "8.2"]
[White "Yu Yangyi"]
[Black "Dominguez Perez, L."]
[Result "1-0"]
[ECO "B90"]
[WhiteElo "2715"]
[BlackElo "2746"]
[Annotator "Illingworth,Max"]
[PlyCount "97"]
[EventDate "2015.06.15"]
[EventRounds "10"]
[EventCountry "CUB"]
[EventCategory "19"]
[Source "Mark Crowther"]
[SourceDate "2015.06.29"]

{To finish, I'll leave you with two more recent endgames to examine in your
own time. In this first one, White converts a decisive advantage in the
endgame.} 1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Be3 Ng4 7. Bc1
Nf6 8. Be3 Ng4 9. Bc1 Nc6 10. h3 Nf6 11. g4 Qb6 12. Nb3 e6 13. g5 Nd7 14. Bf4
Qc7 15. Qd2 b5 16. O-O-O Nce5 17. Bg3 Rb8 18. f4 Nc4 19. Bxc4 bxc4 20. Nd4 Nc5
21. Rhe1 Qb6 22. b3 cxb3 23. axb3 Bb7 24. f5 Rc8 25. Kb1 Be7 26. f6 gxf6 27.
gxf6 Bf8 28. Qe3 Nd7 29. Nd5 Qc5 30. Nxe6 Qxe3 31. Ng7+ Bxg7 32. fxg7 Qxb3+ 33.
cxb3 Rg8 34. Ne3 Bxe4+ 35. Ka1 Kd8 36. Nc4 d5 37. Nd6 Rxg7 38. Rxd5 Rxg3 39.
Nxe4 Rg6 40. Red1 Rc7 41. Nc5 Rxc5 42. Rxc5 Rb6 43. Kb2 Rb7 44. Rc6 Ke7 45. Ka3
f5 46. Rd4 Ne5 47. Rxa6 Ng6 48. h4 Nf8 49. Rd5 1-0 
[Event "68th ch-RUS HL 2015"]
[Site "Kaliningrad RUS"]
[Date "2015.06.26"]
[Round "5.8"]
[White "Malakhov, V."]
[Black "Landa, K."]
[Result "1-0"]
[ECO "A33"]
[WhiteElo "2699"]
[BlackElo "2627"]
[Annotator "Illingworth,Max"]
[PlyCount "161"]
[EventDate "2015.06.22"]
[EventRounds "9"]
[EventCountry "RUS"]
[Source "Mark Crowther"]
[SourceDate "2015.06.29"]

{This example is a great one for examining how to grind down an opponent from
an equal/marginally better endgame; indeed, I may well analyse it for my next
post!} 1. c4 c5 2. Nf3 Nc6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 e6 6. a3 Be7 7. e4 O-O
8. Nf3 Qa5 9. Bd3 d5 10. exd5 exd5 11. cxd5 Nxd5 12. Bxh7+ Kxh7 13. Qxd5 Qa6
14. Qb5 Qxb5 15. Nxb5 Bg4 16. O-O Bxf3 17. gxf3 Bf6 18. Nd6 Nd4 19. Kg2 Be5 20.
Nc4 Nb3 21. Rb1 Bd4 22. Rd1 Rac8 23. Na5 Nxc1 24. Rxd4 b6 25. Nc4 Ne2 26. Rh4+
Kg8 27. Kf1 g5 28. Rg4 Nf4 29. Rxg5+ Kh7 30. b3 Rc5 31. Rg4 Ng6 32. Rd1 b5 33.
Ne3 Rc3 34. Rb4 Rb8 35. Rd5 a5 36. Rbxb5 Rxb3 37. Rxb8 Rxb8 38. Rxa5 Nf4 39.
Ke1 Rb2 40. Rf5 Nd3+ 41. Kf1 Rxf2+ 42. Kg1 Ra2 43. Rxf7+ Kg6 44. Rf8 Rxa3 45.
Nc2 Ra4 46. Rd8 Ra2 47. Ne3 Nf4 48. Ng4 Nh3+ 49. Kf1 Kg5 50. Rh8 Ra1+ 51. Kg2
Nf4+ 52. Kg3 Nh5+ 53. Kf2 Ra2+ 54. Ke1 Ra1+ 55. Kd2 Ra2+ 56. Kd1 Ra1+ 57. Kc2
Ra3 58. Rg8+ Kf4 59. Rf8+ Kg5 60. Kd2 Ra2+ 61. Kc3 Ra3+ 62. Kb4 Ra2 63. Rh8 Rc2
64. Kb3 Rc7 65. Rf8 Rc1 66. Ne3 Rh1 67. Ng4 Rc1 68. Kb2 Rc5 69. Nf2 Rc7 70.
Ne4+ Kg6 71. h4 Rf7 72. Rxf7 Kxf7 73. Kc3 Nf4 74. Kd4 Ng2 75. h5 Nf4 76. Ng3
Kf6 77. Ke4 Kg5 78. Ke5 Nd3+ 79. Ke6 Nf4+ 80. Kf7 Ng2 81. Kg7 1-0