Hello Paul et al,
Paul, I think it was very brave of you to choose this game to display and
annotate. For those of you who weren't aware this is one of the great chess
games of the 21st century, and it racked up quite a few silly statistics to
boot, the main one being the Black Rook standing en prise for 24 moves without
being taken. The complications are immense and credit is due to both players
for exhibiting such creativity.
I've been busy with a computer and here's my, and Deep Rybka's, thoughts on
some of the critical points.
First of all it is universally agreed that Morozevich missed a win with 26
Rxf8+.
Secondly, Paul, I can't fault your analysis with the suggested 33 Qb3. Deep
Rybka finds a couple of slightly quicker ways here and there. Black is badly
tied up, so much so that he can try for stalemates in some lines, so once the
inital threat from the outside passed a-Pawn is neutralised, variations
crystalise easily. Good calculation by that organic whatnot.
Third, finally and longest: the apparent mate in 21 with 22 Be6+. A warning to
readers first - some of this long analysis may not appeal, so if you want to
switch off your PC and go and do something else instead I can well understand.
It's that question of marvelling or studying again. For those of you who like
Queen and Knight attacks there's a treat coming up!
Anyway, here we go. The first thing I'd like to point out is that if you look
at the position after Black's 21st move (21...Kf7) you will see that both Black
Rooks and both Black Bishops are still on their home squares. Hence neither
Rook contributes to the defence and the Bishops hardly at all. Indeed the c8
Bishop is blocked by the d7 Knight. The Black Queen does control some important
squares but because it is right on the edge of the board it's influence is very
limited. The Knights, particularly the one on e3, are the awkward pieces to get
around and trying to calculate any line accurately here, let alone the one I'm
about to give, is a very high tariff of difficulty for a human.
However, it's not so difficult for a decent chess engine provided it doesn't
have too short a horizon, that is the point at which it normally stops
analysing, be it ten moves (five per side or ten ply as it is technically
called) or twelve or whatever.
The basic point of 22 Be6+ is that it forks the Black King and Queen so it must
be taken by one or the other. If 22...Qxe6 23 Nd8+ is another fork that
definitely wins the Queen. Black does have three pieces for her, but with his
King caught in the centre combined with the undeveloped state of most of his
force he will be hard-pushed not to lose further material. I have a game of my
own that I will post later that will demonstrate this point.
That brings us to the main line 22.Be6+ Kxe6.
What I find intriguing from here on is that there are actually very few
constructive Black alternatives. I recall Aleksander Kotov in his well-known
book 'Think like a Grandmaster' discussing the 'tree of analysis where one
given move in a position leads to branches and then thickets of possibble moves
to analyse. Written down on paper some of these look like drawings of oak trees
in full blomm. Here there are several clear branches and the effect is more
like seeing a tall pine with short, stubby branches occasionally sticking out
from a very long trunk.
We start with 22.Be6+ Kxe6
23.Qg6+ Nf6
24.gxf6
White gains a lot of potential here by threatening 25 fxg7+ and 26 gxh8=Q
because that completely reverses the material imbalance. So Black must deal
with that threat at some point.
If now
24...Qa1+
25.Kd2 Qxd1+
26.Rxd1 gxf6
27.Nd8+ Kd6
28.Kxe3+!
A nice touch
28... Kc7
29.Qf7+
looks convincing enough to me but Deep Rybka ploughed on with
29...Kb6
30.Qxf6+ Ka7
31.Nc6+ Kb7
32.Na5+ Ka7
33.Qf7+ Kb6
34.Nc4+ Kb5
35.Rd5+ Kc6
36.Nxe5+ Kb6
37.Qf6+ Ka7
38.Nc6+ Kb6
39.Na5+
- what a Knight! -
39...Ka7
40.Qd4+ Bc5
41.Qxc5+ Kb8
42.Qb6+ Bb7
43.Qxb7 mate
... and that is your mate in 22 moves. Note where the Black Rooks and Bishops
stand at move 42. Divergences by Black lead to shorter mates.
Alternatively after 22.Be6+ Kxe6
23.Qg6+ Nf6
24.gxf6
there is the immediate 24...gxf6. Now we have another Queen and Knight tango:
25.Qe8+ Kf5
26.Nd4+ Kf4
27.Ne2+ Kf5
28.Ng3+ Kf4
29.Nh5+
From a5 in the previous line to h5 in this ...
29...Kf5
(Or 29...Kxf3 30.Qc6+ Nd5 31.Rxd5 Qa1+ 32.Kd2 Kg4 33.Rxe5 fxe5 [if 33...Qxh1
34.Qc4+ Kh3 35.Qd3+ Qf3 36 Qxf3+ Kh2 37 Qg3+ Kh1 38 Rd1 mate] 34.Qg6+ Kf3
35.Qg3+ Ke4 36.Qd3 mate)
30.Rd4 Qa1+
31 Kd2 Qd1+
32 Rxd1 Nf1+
(33...Nc4+ 34 Rxc4 makes no difference.)
33 Rxf1 Rg8
34 Rf4+ exf4
35 Qe4 mate.
Once again note where the Black Rooks and Bishops stand at move 33. Yes, the
engine has the calculating power, but on such an open board with the majority
of the Black forces stuck on the edges barely involved is it any wonder that it
can go so deeply with such apparently little effort? Any decent GM could
calculate to the mate from certain points in these lines but they won't need to
because they already know that they are winning.
I should qualify the remark "such apparently little effort". Depending on both
your PC and the strength of your chess engine you may find that this analysis
takes up anywhere between 40-80% of your CPU capacity. That's a lot of thinking.
My final words here, other than once again thanking Paul for his original post,
are just to remind you that two humans played this game and it has gone down in
history.
Happy analysing,
Tyson
On 16 September 2019 at 14:08 Paul Benson <paul.benson@xxxxxxxxxxxxxxx> wrote:
Skype coaching session, 25 August 2019.
Forthcoming coaching sessions in September will be on Sundays: 22, 29, at
14:00 UTC, which is 15:00 BST.
The game discussed (76 moves) is given below with annotations and without
annotations.
Re-structuring of the sessions now offers the option for guests to bring
their own games for group discussion.
Recordings of some previous coaching sessions can be found at:
http://www.open-aurec.com/Skype/PaulBenson/PaulBenson.htm
Paul Benson.
* * *
Annotated Game.
White: A. Morozevich, 2751.
Black: M. Vachier-Lagrave, 2703.
Event: Biel 2009.
Result: 0-1 in 76 moves.
Opening: Sicilian Defence, Scheveningen Variation, English Attack, B80.
| Editorial.
Well known chess fact: "Long Games Are Boring!"
Here is a 76-mover to further reinforce this, "Fact".
Reckon you would do better to sort your sock drawer than play through this
stodge.
||
1. e4 c5
2. Nf3 d6
3. d4 cxd4
4. Nxd4 Nf6
5. Nc3 a6
6. f3
| A subtlety.
If white chooses the far more popular, 6. Be3, black might choose to chase
around with, 6. ... Ng4, forcing white into a 3-way decision.
(A). If, 6. Be3 Ng4 7. Bc1 Nf6 8. Be3 Ng4 9. Bc1 Nf6 10. Be3, indicating a
willingness to draw.
(B). Or if, 6. Be3 Ng4 7. Bc1 Nf6, and choose a different system other than
those entered with, 6. Be3, but this must be seen as a psychological victory
for black.
(C). Bite the bullet with, 6. Be3 Ng4 7. Bg5 h6 8. Bh4 g5 9. Bg3, a
completely different game to those arising from the usual English Attack, the
white kingside pawn-storm is no longer an option.
||
6. ... e6
| The most popular choice here is, 6. ... e5, where the most popular outcome
is opposite-wing castling, complex pawn storms available to both players, a
mutual joyous headache.
||
7. Be3 b5
8. Qd2 Nbd7
9. g4 h6
| If white is to deflect the f6 knight with a pawn there more preparation is
demanded of white.
||
10. O-O-O b4
| Black is fighting for the central break of, pawn d5, this requires the
white c3 knight be pushed around.
||
11. Nce2
| White is not worried about having the c3 knight dislodged.
A common defensive white plan is to shuffle, Kb1, follow with, Nc1, and the
traditional weakish point of a2 is given good cover.
||
11. ... Qc7
12. h4 d5
| Black achieves the aim, a central, pawn d5, break.
White now produces a move to make every beginner smile with recognition.
||
13. Nf4
| No Grand Master has previously tried this move, possibly due to the black
option of forking the white knights with, 13. ... e5.
The more popular choice for white here is, 13. Bf4, when should black again
pawn fork with, 13. ... e5 14. Bg3, the black e5 pawn is pinned, white will
have, Nf5, available should the black queen escape the pin.
||
13. ... e5
| In Grand Master terminology, black is saying: "Show me!"
Instead continuing with the plan of creating a white pawn weakness with, 13.
... dxe4, was more consistent.
||
14. Nfe6
| There are a variety of ways to enter the realm of material loss for
positional gain:
(A). If, 14. Nxd5 Nxd5 15. exd5 exd4 16. Bxd4, white is pressuring the black
g7 pawn, the e-file is fully open, black is a piece for 2 pawns up but
lagging in development, each player probably believes they have the better of
it.
(B). Or if, 14. Nf5 exf4 15. Bxf4 Qa5 16. exd5, same thoughts as the line
above.
||
14. ... fxe6
| Declining gains nothing, sometimes there is little choice but to take the
offered material and suffer the short-term consequences, hoping about 8 moves
down the line something previously unforeseen becomes visible.
||
15. Nxe6 Qa5
16. exd5
| So white actually gets 2 pawns for the piece plus a wonderful,
"Mega-Knight", ruling the black position from the e6 square.
||
16. ... Qxa2
| So the compensation drops to a single pawn for the piece, but nothing is
being done regarding the white monster on e6.
And yes, there is a cheeky tempo-gaining threat of mate with, Qa1, in need of
neutralising which avoids an exchange loss to white, Nc7+.
||
17. Qd3
| A doubler.
Firstly, for the defence, the d2 square is vacated, the black threat of, Qa1+
mate, is eliminated.
Secondly, the white queen threatens a powerful, Qg6+, invasion, this must be
prevented immediately.
||
17. ... Kf7
18. g5
| A tripler.
Firstly, the tactics insist The black f6 knight must move.
Secondly, the advancing white g-pawn has more ambitions than might appear
from this move.
Thirdly, the h3 - c8 diagonal is opened up for the f1 bishop, a recurrent
theme in many games in the English Attack.
||
18. ... Nxd5
| And now there is zero material compensation for the sacrificed white
knight.
Furthermore the white e6 knight now has no support.
Instead, 18. ... hxg5 19. Nxg5+, gives black serious problems, some ideas
run:
(A). If, 18. ... hxg5 19. Nxg5+ Kg8 20. Qg6 Rh6 21. Qf7+ Kh8 22. Ne6,
threatening both, Nxf8, and, Bxh6, winning material.
(B). Or if, 18. ... hxg5 19. Nxg5+ Ke8 20. Qg6+ Kd8 21. Ne6+ Ke7 22. d6+ Kxe6
23. Bh3+ mate.
(C). Or if, 18. ... hxg5 19. Nxg5+ Ke8 20. Qg6+ Ke7 21. d6+ Kd8 22. Nf7+ Qxf7
23. Qxf7, black has lost the queen for 2 white minor pieces.
(D). Or if, 18. ... hxg5 19. Nxg5+ Ke7 20. d6+ Kd8 21. Qg6, black must lose
either a whole rook to, 22. Nf7+, or give up the queen on the white f7
knight, both miserable.
||
19. Bh3
| A tripler.
Firstly, the white e6 knight is again supported.
Secondly, tactical-antennae should now be twitching over possible
combinations on the ex-rayed black d7 knight.
Thirdly, the black d5 knight is now a genuine target.
Instead, 18. Qxd5 Qxd5 19. Rxd5 Kxe6, leaves white a piece down in a
queenless middlegame, black is easily winning despite the lack of
development.
||
19. ... Nxe3
| Following the principle: When ahead in material, trade pieces, but keep
pawns on.
And the reverse is generally true: When behind in material, trade pawns, but
keep pieces on.
White to play is technically 2 pieces down, so the, "Obvious", move is to
recapture with, Qxe3, right?
||
20. Nd8+
| White is prodding and probing on the black king, there are several replies
to this check, black must be accurate to avoid a speedy crush.
||
20. ... Ke7
| The best of a bad lot, the less favourable options run:
(A). If, 20. ... Kg8 21. Be6+ Qxe6 22. Nxe6 Nxd1 23. Rxd1, black might have
white rook plus 2 minor pieces for the queen, but the lack of development
should allow the white pieces to find something.
(B). Or if, 20. ... Ke8 21. Bxd7+ Bxd7 22. Qxd7+ mate, the unleashed h3
bishop demonstrates that x-ray attacks contain venom.
(C). Or if, 20. ... Ke8 21. Bxd7+ Kxd8 22. Be6+ Kc7 23. Bxa2 Nxd1 24. Rxd1,
black has rook plus bishop for the lost queen, but the king is wide open to
the remaining white forces, expect more black material to fall off soon.
||
21. Nc6+ Kf7
| Retreating is painful:
(A). If, 21. ... Ke8 22. Bxd7+ Bxd7 23. Qxd7+ mate.
(B). Or if, 21. ... Ke8 22. Bxd7 Kf7 23. Qxe3, material equality is
re-established, but the superior development strongly favours white.
||
22. g6+
| Post-game Fritzing apparently reports that, 22. Be6+, leads to a forced
mate in 21 moves, "Obviously".
Grand Rabbit cannot find anything concrete, the best chances seem to come
from, 22. Be6+ Kxe6 23. Qg6+ Nf6 24. gxf6 gxf6, 25. Rd8 Bb7 26. Qe8+ Kf5 27.
Qd7+ Kf4 28. ???, yes, what has white got for move 28 here?
Fine stuff Fritz, but the, "Human-Thinking", game choice of confining the
black king on g8 with the h8 rook imprisoned, for who-knows-how-long, comes
naturally in five seconds of casual strategic reflection.
||
22. ... Kg8
| The black king voluntarily enters a prison.
Instead, 22. Ke8 23. Bxd7+ Bxd7 24. Qxd7+ mate.
Alternatively, 22. ... Kf6, seems to be leading with the chin, white has so
many good ideas to win material back while keeping the strong position.
The general idea is, 22. ... Kf6 23, Qxe3, when black must do something about
the potential loss of a piece with, 24. Bxd7, but if that knight either moves
or is eliminated, then white has, Qxe5+.
Also after, 22. ... Kf6 23. Qxe3, the line-opening advance of, pawn f4, is in
the air.
||
23. Qxe3 Bc5
| A doubler.
Firstly, the black f8 bishop develops with a tempo-attack on the white e3
queen.
Secondly, when a unit moves it vacates a square for someone else, but who
would like to go to f8?
Instead, black dare not snaffle a useless pawn with, 23. ... Qa1+ 24. Kd2
Qxb2, as, 25. Be6+ mate, shows why the black queen is advised to remain on
the a2 - g8 diagonal for quite some time to come.
||
24. Qe4
| Setting up an x-ray attack on the black a8 rook, threat, Ne7+, though
perhaps white might think the c6 knight is presently of greater value than
the undeveloped black rook.
||
24. ... Nf8
25. Rd8
| As black increases the guard on e6 white immediately reduces it by pinning
both the black f8 knight and c8 bishop.
All of the black pieces are now on the edge of the board, and with virtually
no safe moves available, this game must be over in a few moves, right?
||
25. ... Bb7
| Common sense should warn that greed must fail, 25. ... Qa1+ 26. Kd2 Qxh1
27. Qd5+ Be6 28. Qxe6+ mate.
||
26. Rxa8
| White has played with inspiration up to here, but sadly Fritz and friends
again insist in making their opinion known.
White can gain a decisive advantage with, 26. Rxf8+, when any of 3
re-captures, the black g8 king, the black c5 bishop, the black a8 rook, are
all met with 27. Qxe5, and black has severe problems defending against a
white, 28. Be6+, a costly royal skewer.
Still, the black f8 knight, the black g8 king, and the black h8 rook, seem to
be having problems, so it is unfair to criticise white, but only to point out
the opportunity of a quick kill has been missed.
||
26. ... Bxa8
27. h5
| Securing the white g6 pawn, the white e4 queen is again free to roam.
Black in the game to play must do something, but what?
A little process of elimination is required.
Clue #1: The black queen dare not leave the a2 - h8 diagonal just yet.
Clue #2: The black c5 bishop dare not move off the a3 - f8 diagonal,which
would permit, Ne7+ mate.
Clue #3: The black a8 bishop cannot get into the game other than, Bb7, pawn
a5, and Ba6, as if that was anything useful anyway.
Clue #4: The f8 knight will be captured on whichever of the 4 available
squares it chooses to move.
Clue #5: White just having defended the important g6 pawn, now intends, Qxe5,
trying to push the black c5 bishop off the a3 - f8 diagonal, so as to follow
with, Ne7+ mate.
So it is not a question of what black should do, but more a question of what
is left, and however ridiculous it seems, consider it carefully, and a light
might go on.
||
27. ... Rh7
| A most remarkable quadrupler.
Firstly, the black king's rook was trapped and with no obvious means of
escape, it offers itself in order to break the white prison.
Secondly, when a unit moves it vacates a square for someone else, the black
g8 king now has a means of getting off that dangerous a2 - g8 diagonal.
Thirdly, the potential white knight threat of, Ne7+, will not now be mate,
the c5 bishop can think of getting active.
Fourthly, the black a2 queen no longer needs to guard the a2 - g8 diagonal,
she can also thing of getting active.
Instead the similarly altruistic, 27. ... Nh7, does not solve the problems on
the a2 - g8 diagonal, a couple of sample lines run:
(A). If, 27. ... Nh7 28. Rd1 Bxc6 29. Qxc6 Bf8 30. Qe6+ Qxe6 31. Bxe6+ mate.
(B). Or if, 27. ... Nh7 28. Rd1 Nf6 29. Rd8+ Bf8 30. Ne7+ mate.
Now black is no longer scared of the threats on the a2 - g8 diagonal, white
must take care of the h1 rook.
||
28. Re1 Bxc6
| Black chooses to eliminate the white c6 knight, meaning, Ne7+ mate, can
never happen, the c5 bishop is now released for other active duties.
||
29. Qxc6 Bd4
| Isn't chess a strange game?
Black has a couple of pieces completely tied up in knots on the kingside by
the white h3 bishop and a couple of advanced pawns.
Yet in the centre/queenside, despite being an exchange down in that region,
black is beginning to generate play, the white king is about to feel rather
uncomfortable.
||
30. Kd2 Qxb2
31. Qc4+
| White chooses to force the black king to h8, retaining the option of, gxh7,
when appropriate.
||
31. ... Kh8
32. Kd3
| Avoiding, 32. ... Bc3+, costing white an exchange.
||
32. ... a5
33. Qc8
| Grand Master confuses Grand Rabbit.
White chooses to keep the middlegame going, fine, but surely an endgame with
queens off the board would be easier to play, especially as the complexities
must have consumed much time on the clock?
Why not try, 33. Qb3, leaving black to decide how the queens will be traded
off?
Apologies for any inaccurate calculations, all done with aging bio-organic
engine, some ideas run:
(A). If, 33. Qb3 Qa3 34. Qxa3 bxa3 35. c3 Bb6 36. Rxe4, the black bishop
cannot guard both black a-pawns and the a3 pawn cannot be forced through,
surely a white win with a little technique.
(B). Or if, 33. Qb3 Qxb3 34. cxb3 Bc3 35. Rd1, and the white king simply
marches to win the black a5 pawn, then organise, Rxb4, giving up the exchange
to create a distant passed b-pawn, other plans using the advantage of the
exchange surely available.
(C). Or if, 33. Qb3 Qc3+ 34. Qxc3 Bxc3 35. Re4 a4 36. Rc4 a3 37. Rc8 Kg8 38.
Be6+ mate, that light square weakness strikes again.
(D). Or if, 33. Qb3 Qc3+ 34. Qxc3 bxc3 35. f4, black cannot capture, exf4, as
the d4 bishop would fall, white then has, fxe5, and this new white passed
e-pawn will crash through to the 8th rank.
(E). Or if, 33. Qb3 a4 34. Qxb2 Bxb2 35. Re4 a3 36. Rc4 a2 37. Rc8 Kg8 39.
Be6+ mate, the black bishop on b2 failed to protect the f8 knight.
(F). Or if, 33. Qb3 a4 34. Qxb2 Bxb2 35. Re4 Bd4 36. Kc4 a3 37. Kb3 Kg8 38.
f4 Bc3 39. fxe5 Rh8 40. e6 Nxe6 41. Bxe6+ Kf8 42. Bf7 Bd2 43. Re8+ mate.
(G). Or if, 33. Qb3 a4 34. Qxb2 Bxb2 35. Re4 Bd4 36. Kc4 a3 37. Kb3 Kg8 38.
f4 Bc3 39. fxe5 Rh8 40. e6 Nxg6 41. hxg6 Kf8 42. e7+ Ke8 43. Bf1 Kd7 44.
Bb5+, and promotion of the white e-pawn costs black a rook.
(H). Or if, 33. Qb3 a4 34. Qxb2 Bxb2 35. Re4 Bd4 36. Kc4 a3 37. Kb3 Kg8 38.
f4 Bc3 39. fxe5 Rh8 40. e6 Nh7 41. e7 Nf6 42. Be6+ mate.
Lastly a very tricky line, pay careful attention on how white uses the c-pawn
to win the black bishop and the f-pawn to give black a move to break the
stalemate trap.
(I). Or if, 33. Qb3 a4 34. Qxb2 Bxb2 35. Re4 Bd4 36. Kc4 a3 37. Kb3 Bc5 38.
Rxe5 Bd6 39. Re8 Bc5 40. Be6 Bd6 41. Ka4 Bc5 42. Kb5 Bd6 43. Kc6 b3 44. Bxb3
Bb4 45. c4 a2 46. Bxa2 Ba3 47. Bb1 Kg8 48. gxh7+ Kh8 49. Bg6 Bb4 50. f4 Ba3
51. f5 Bb4 52. f6 Ba3 53. c5 Bxc5 54. Kxc5 gxf6 56. Rxf8+ Kg7 57. Rg8+ mate.
Sure, this last line is 25 moves long, A Grand Master would never calculate
this far, they just go to the point where they feel it is decisive and stop,
content in the knowledge they will work out how to win when the position
actually arises.
||
33. ... Qa3+
| A very important check, an x-ray defence is placed on the black f8 knight
with tempo-gain.
||
34. Ke4 b3
| Protecting the stranded black f8 knight from a3 while also creating a
little confusion in the centre/queenside.
||
35. cxb3 a4
| Setting a sneaky trap for white to fall into while the clock is most likely
already rather low.
||
36. Rb1
| White had to avoid capturing on a4, the ideas run:
(A). If, 36. bxa4 Qb4 37. Rc1 Bc5+ 38. Kf5 Qf4+ mate, we all knew the f8
knight would prove useful at some time, here it controls the e6 flight
square.
(B). Or if, 36. bxa4 Qb4 37. Rc1 Bc5+ 38. Kd5 Qd4+ 39. Kc6 Qd6+ 40. Kb7 Qb6+
41. Ka8 Qa7+ mate, but white can avoid this.
(C). Or if, 36. bxa4 Qb4 37. Rc1 Bc5+ 38. Kd5 Qd4+ 39. Kc6 Qd6+ 40. Kb5 Qb6+
41. Ka4 Qb4+ mate, also avoidable.
(D). Or if, 36. bxa4 Qb4 37. Rc1 Bc5+ 38. Kd5 Qd4+ 39. Kc6 Qd6+ 40. Kb5 Qb6+
41. Kc4 Qb4+ 42. Kd3 Qd4+ 43. Ke2 Qe3+ 44. Kf1 Qf2+ mate, yes, also
avoidable.
So how does white avoid all these shock mates where black shows how an
opposite-colour bishop middlegame favours whoever is on the attack?
(E). Or if, 36. bxa4 Qb4 37. Rc1 Bc5+ 38. Kd5 Qd4+ 39. Kc6 Qd6+ 40. Kb5 Qb6+
41. Kc4 Qb4+ 42. Kd5 Qd4+, and the white king circles the black c5 bishop
until a draw is agreed or 3-fold repetition claimed.
There is also, 36. bxa4 Bf2, setting similar mating/perpetual traps to be
crunched through.
||
36. ... Qb4
37. Qc4
| White realises the active black queen and bishop need neutralising, fine,
but along the way black has made some progress on the queenside, take careful
note of the black a-pawn.
||
37. ... Qb7+
38. Qd5
| Forced.
Instead, 38. Kf5 Qxf3+ mate, or, 38. Kd3 Qxf3+, picks off the white h3
bishop.
||
38. ... Qb4
39. Qc4
| So black can take a repetition draw, but is there another way to keep
pressuring white?
||
39. ... Qd2
| With the same sort of threat involving, Qe3+, and then capturing the white
f3 pawn, to pick off the white h3 bishop.
||
40. Bg4
| Black can do little on the kingside, but now the queenside is a totally
different matter.
||
40. ... a3
| And just how is white intending to stop the advancing black a-pawn?
||
41. Qf7
| A doubler.
Firstly, white again threatens, Qxf8+ mate.
Secondly, the white f3 pawn is backwardly protected, the black idea of, Qe3+,
then, Qxf3+, is neutralised.
Black to play finds a neat switchback manoeuvre to both defend the kingside
and maintain the main queenside threat.
||
41. ... Qc2+
42. Kd5 Qc5+
| The black queen guards the f8 knight again, the black d4 bishop controls
the a-file promotion square, and all done with tempo-attacks on the white
king.
||
43. Ke4 a2
44. Rc1
| Making an offer which cannot be accepted, furthermore the black queen must
think carefully about her future, or perhaps not?
||
44. ... a1=Q
| Making a queen-offer or two, one of which must be accepted.
||
45. Rxc5
| All captures of an opposing queen should be equal, but some are more equal
than others.
White had to take the black c5 queen, the alternate, 45. Rxa1, is a blunder.
If, 45. Rxa1 Qc2+ 46. Kd5 Qxb3+ 47. Ke4 Qxf7, black has a decisive material
advantage, even if white snatches the black h7 rook.
||
45. ... Bxc5
46. Qd5
| Forced, black was threatening, Qd4+, forcing the white king to f5, after
which, Qf4+ mate would occur.
Now that black has managed to convert a queenside pawn advantage into a
rook-snaffle, it is time to get queens off.
||
46. ... Qe1+
| There is no law against pushing your opponent into making forced replies,
though in the interest of keeping chess-friends, try not to smile as you do
so.
Instead, 46. ... Qd4+, similarly achieves the aim, but lacks the joy of
kicking the white king around for a couple of moves.
||
47. Kd3 Qd1+
48. Kc4 Qxd5+
49. Kxd5 Ba3
| Black accepts the e5 pawn should be let go.
Instead, 49. ... Bd4 50. f4, undermining the black d4 bishop, white will gain
a very useful passed e-pawn.
White now has to assess whether taking the black h7 rook and so freeing the
black knight plus king to enter the game is the way to proceed, not easy.
There is also a little black kingside re-arranging plan given in the proposed
queen-trade analysis to consider, white might choose to permit black to
remain a rook and knight up with the apparently impossible task of
unravelling the kingside.
||
50. Bf5
| White solidifies the kingside, the moment of decision has yet to arise.
||
50. ... Kg8
51. Kxe5 Rh8
| Strategies have crystallised.
Had white on the previous move chosen, gxh7+, black would capture, Nxh7,
giving white another choice.
If white trades, Bxh7+, white will eventually win the black bishop by forcing
the b3 pawn onto b8 for a queen, however, black will have shuffled the king
on the light squares to win the white h5 pawn producing a simply-winning
kingside 2 versus 1 pawn ending.
If white declines trading, Bxh7+, the black knight plus king will enter the
centre, the only 2 difficulties black must then avoid are entering an
opposite bishop ending, or being left with only the bishop plus h-pawn when
the white king can occupy the h1 square, both scenarios being drawn.
||
52. Kd5 Nh7
| Making white an offer which cannot be refused, this knight must not be
allowed to escape.
Instead, 52. ... Nxg6 53. Bxg6 Kf8 54. Ke6, believe it or not, white a rook
for a useless b-pawn down, holds a draw.
The black king cannot escape, the white king just shuffles, Kd7 - Ke6 - Kd7 -
Ke6, denying the black king the escape of, Ke7.
||
53. gxh7+ Kf7
54. Bg6+
| Planning a self-immobilisation of the kingside, white is attempting to hold
matters by king-shuffling, this is no longer a game white can draw, the real
question is can black find the winning method?
||
54. ... Kf6
| So how is black to set about winning this?
Do not think of, "Move-Reply", analysis, that is, "I go here, he goes there,
then I go here, etc., etc."
A much broader strategic approach is required, find this, and only then find
the moves which fit the plan.
Remember, white cannot do anything positive other than promote the h7 pawn,
and if that seems impossible, think again.
If black is ultimately going to win this, then the black h8 rook must
centralise.
Fine, but in doing so, black must be aware of the white plan of, Bf7, then,
Bg8, shielding the promotion square, h8=Q, will follow.
However, the white promotion plan takes 3 moves, and on h8 the reincarnated
white queen needs a couple of more moves to make an impact on the game.
So, black can afford permitting white promotion providing the combination of
black king, rook, bishop, can weave a forcing mating pattern.
Once all this is realised we must ask, what is the first step on the black
road to victory?
Simple, eliminate the potentially dangerous white b3 pawn, this requires the
king rather than the rook, so start getting him queenside, remembering the a3
bishop is happy to provide shuffle moves in order to force the white king to
give up the opposition, allowing the black king to invade.
||
55. f4 Bc1
56. f5 Bd2
| Preventing white, pawn b4, no point in letting white have the slightest
hint of activity.
||
57. Kd6
| White king takes the opposition.
||
57. ... Be1
| Black bishop shuffles, the power over the e1 - a5 diagonal remains in
place, but now white must move, perhaps a non-critical zugzwang, but
nevertheless, it is zugzwang.
||
58. Kd7 Bb4
| Taking control of the a3 - f8 diagonal, in particular the e7 and d6 squares
are denied the white king, irrespective of where the black king wanders.
||
59. Kc7 Ke5
60. Kd7 Ba3
| A refinement is evolving as the game proceeds, the general plan was run the
black king to win the white b3 pawn, yes, but another skittle is in the air.
If the white king is allowed to approach the kingside, say landing on f7,
there might arise complications involving white, pawn f6, forcing, black,
gxf6, when white invades, Kg7, hitting the black h8 rook with subsequent
threats of, h8=Q.
Black might be able to cope with all this by placing the bishop on the a1 -
h8 diagonal, fine, but then it is not preventing the advance of the white
b-pawn.
So black is selecting moves which eliminate the white b-pawn advance which
simultaneously keep the white king out of the kingside.
Once the black king is close enough to the white b3 pawn, if needed, the
black bishop can protect the a1 - h8 diagonal.
When converting a strong position into a clearly winning game, do not give
the opponent any chances of counterplay whatsoever.
||
61. Kc6 Kd4
62. Kc7 Kc3
63. Kd7 Kb4
| Black is keeping the only potential white activity under control.
If white now tries for kingside activity, 64. f6 gxf6 65. Ke6 Bb2 66. Kf7
Kxb3 67. Kg7 f5+ 68. Kxh6 f4 69. Be4 Kc4 70. Kg5 Rf8, the black b2 bishop
covers the h8 promotion square, the black king and rook will force the passed
f-pawn forward to cost white the bishop.
||
64. Kd6
| Attacking gets nowhere, some ideas run:
(A). If, 64. Ke7 Kxb3+ 65. Kf7 Bb2 66. Ke6 Bf6 67. Kf7 Rd8, the white king
has voluntarily caged himself in the kingside, white can only give up
material with, 68. h8=Q Rxh8, and the major white threat, the h7 pawn is
gone, black now wins easily.
With, 64. Kd6, white is keeping the king in the advanced queenside quarter of
the board, hoping there will be no means for black to make full use of the
extra rook.
(B). Or if, 64. Kc6 Rf8 65. Kd7 Kxb3 66. Bf7+ Rxf7 67. Ke8 Ra7 68. h8=Q Ra8+
69. Kf7 Rxh8 70. Kxg7 Bb2+ 71. Kg6 Kc4 72. f6 Kd5 73. f7 Ke6 74. f8=Q Rxf8
75. Kxh6 Kf7, the remaining white h5 pawn has no promotion prospects.
(C). Or if, 64. Kc6 Rf8 65. Kd7 Kxb3 66. Bf7+ Rxf7 67. Ke8 Ra7 68. h8=Q Ra8+
69. Kf7 Rxh8 70. Kxg7 Bb2+ 71. Kf7 Kc4 72. Ke6 Kc5 73. f6 Kc6 74. f7 Bg7 75.
Kf5 Kd6 76. Kg6 Bf8, the white f-pawn cannot progress.
||
64. ... Kxb3+
| Perhaps the rarest type of move in chess, a king makes a capture and gives
check.
||
65. Kd5 Bb2
| The strategies for each player are now clear.
White Keep the kingside grip for as long as appropriate while the king
shuffles around in the advanced queenside quarter.
Black: Re-groups the pieces to limit the mobility of the white king, slowly
but surely forcing him to the edge of the board.
||
66. Kd6
| Trying to hide in the kingside does not help white.
When the king on king chase hits the edge of the board and black needs the
opposition there is a bishop move to pass the onus to move back onto white.
Imagine the chased white king hiding on h3, the black king on f3, the black
bishop on the b8 - h2 diagonal, perhaps c7.
Once the white king is on the edge of the board, as just imagined, the h8
rook swings sideways, moves such as, Rb8, then, Rb1, permit white, h8=Q, but,
Rh1+ mate would occur.
The white king cannot resist being nudged to the edge of the board, once the
black king becomes established on f3, it is time for the h8 rook to act.
||
66. ... Bf6
67. Kc5 Kc3
68. Kd6 Kd4
69. Kc6 Rd8
| Preventing the white king from returning to the d-file, the net is closing
in.
||
70. Kb6 Kd5
| How bizarre!
Black has been taking the opposition for a few moves but now takes up the,
"Knight's Jump", formation.
That is, if either king were replaced with a knight then the remaining king
would be instantly placed in check.
The reason for this formation is should the white king choose to re-take the
opposition then the black rook will give check, forcing the white king onto
the edge of the board.
||
71. Kc7
| The white king declines taking the opposition preferring to keeps the,
"Knight's Jump", formation.
Instead, 71. Kb5 Rb8+, forces the white king onto the edge of the board,
black would then place the king on the c-file, white will be forced into
playing, Bf7, after which play would continue with black producing an
important defensive concept, yet to be revealed in game.
||
71. ... Kc5
| Now we are back to taking the opposition, black is hoping for white to
further retreat the king toward the corner of the board.
||
72. Bf7
| White breaks the kingside grip with the idea of, 73. Bg8, shielding the h7
pawn promotion.
This has been available since black centralised with, Rd8, there had better
be a good reply to this white idea, or black will have to humbly retreat
with, Rh8, and accept it is a draw.
||
72. ... g5
| Aha, when a unit moves it vacates the square for someone else.
In this case black actually steps over the vacated g6 square, but the
principle is roughly the same, any movement creates a variety options for
others.
Instead, 72. Kb7 Kd6, black again takes the, "Knight's Jump", formation, any
white king reply leads it into a b-file opposition vulnerable to, Rb8+, or a
retreat onto the a-file, both undesirable.
||
73. fxg6
| Black cannot be left with a racing g-pawn.
So, white now has connected passed pawns, ideally each will promote costing
black a piece, right?
Not really, the black f6 bishop immobilises them, neither can safely cross
over the h8 - a1 diagonal without the help of the white king, who is
imprisoned on the opposite side of the board.
Black to play no longer fears the white h7 pawn, someone who has had a
wretched game so far is about to become the strongest piece on the board.
Instead, 73. hxg6, is the same problem, no white pawn can safely advance.
||
73. ... Rd6
| A neat twist.
The black king and rook, who were entombed for so long over on the kingside,
now rule the queenside, the white king is now reduced to just 6 squares.
||
74. Be8
White tries to keep the king off the edge, instead, 74. Kb7 Rd7+, forces the
white king onto the a-file or 8th rank, the net would be closing.
Black to play finds a means to scare the white king into moving off c7.
||
74. ... Be5
| Ignore x-ray attacks at your peril.
||
75. Kb7 Rb6+
| Only 3 squares available to the white king now.
||
76. Kc8
| Moving to the a-file was equally helpless, a couple of lines showing the
main ideas run:
(A). If, 76. Ka8 Kd6 77. Ka7 Bd4 78. Ka8 Kc7 79. Bf7 Ra6+ mate.
(B). Or if, 76. Ka7 Bd4 77. Bf7 Kd6 78. Ka8 Kc7 79. Ka7 Rxg6+ 80. Ka8 Ra6+
mate.
||
76. ... Kd6
White resigns, 0-1
| The choices were:
(A). If, 77. Kd8 Rb8+ mate.
(B). Or if, 77. Bd7 Ke7 78. Ba4 Rb8+ mate.
||
* * *
## Unannotated Game.
White: A. Morozevich, 2751.
Black: M. Vachier-Lagrave, 2703.
Event: Biel 2009.
Result: 0-1 in 76 moves.
Opening: Sicilian Defence, Scheveningen Variation, English Attack, B80.
1. e4 c5
2. Nf3 d6
3. d4 cxd4
4. Nxd4 Nf6
5. Nc3 a6
6. f3 e6
7. Be3 b5
8. Qd2 Nbd7
9. g4 h6
10. O-O-O b4
11. Nce2 Qc7
12. h4 d5
13. Nf4 e5
14. Nfe6 fxe6
15. Nxe6 Qa5
16. exd5 Qxa2
17. Qd3 Kf7
18. g5 Nxd5
19. Bh3 Nxe3
20. Nd8+ Ke7
21. Nc6+ Kf7
22. g6+ Kg8
23. Qxe3 Bc5
24. Qe4 Nf8
25. Rd8 Bb7
26. Rxa8 Bxa8
27. h5 Rh7
28. Re1 Bxc6
29. Qxc6 Bd4
30. Kd2 Qxb2
31. Qc4+ Kh8
32. Kd3 a5
33. Qc8 Qa3+
34. Ke4 b3
35. cxb3 a4
36. Rb1 Qb4
37. Qc4 Qb7+
38. Qd5 Qb4
39. Qc4 Qd2
40. Bg4 a3
41. Qf7 Qc2+
42. Kd5 Qc5+
43. Ke4 a2
44. Rc1 a1=Q
45. Rxc5 Bxc5
46. Qd5 Qe1+
47. Kd3 Qd1+
48. Kc4 Qxd5+
49. Kxd5 Ba3
50. Bf5 Kg8
51. Kxe5 Rh8
52. Kd5 Nh7
53. gxh7+ Kf7
54. Bg6+ Kf6
55. f4 Bc1
56. f5 Bd2
57. Kd6 Be1
58. Kd7 Bb4
59. Kc7 Ke5
60. Kd7 Ba3
61. Kc6 Kd4
62. Kc7 Kc3
63. Kd7 Kb4
64. Kd6 Kxb3+
65. Kd5 Bb2
66. Kd6 Bf6
67. Kc5 Kc3
68. Kd6 Kd4
69. Kc6 Rd8
70. Kb6 Kd5
71. Kc7 Kc5
72. Bf7 g5
73. fxg6 Rd6
74. Be8 Be5
75. Kb7 Rb6+
76. Kc8 Kd6
White resigns, 0-1
* * *
