GOTW: So - Le Siège, 42nd Olympiad, 2016

[Event "42nd Olympiad"]
[Site "Baku"]
[Date "2016.09.14"]
[Round "11"]
[White "So, Wesley"]
[Black "Le Siège, Alexandre"]
[Result "1-0"]
[ECO "B13"]
[WhiteElo "2782"]
[BlackElo "2512"]
[Annotator "Alexandre Le Siège"]
[PlyCount "71"]
[EventDate "2016.09.??"]
[EventType "team-swiss"]
[EventRounds "11"]
[EventCountry "AZE"]

{This was the 11th and final round of the Olympiad and Canada was playing
board 1 against the United States. The stakes were pretty high: a win by the
United States would mean a gold medal for them, and a win for Canada would
give us 4th place. As it turned out a draw would also have secured 4th place, 
because of other teams' results, since our tie-break was through the roof. The
second tie-break (after matches won) is calculated by multiplying the number
of points scored against a country by the number of games they won. Since we
beat a lot of teams 4-0 and lost by the smallest margins in our defeats, we
would be ahead of most of the teams at the finish line.     My task wasn't
easy: I was playing Black against the in-form Wesley So, rated 7th in the
world who just achieved a smooth victory in the Sinquefield cup 2016, arguably
one of the strongest tournaments in the world.     We didn't have much time to
prepare since the last round was being played at 11 am instead of the usual
3pm. I just did a quick preparation trying to find any obvious holes in my
repertoire. The task was complicated by the fact that So has a very balanced
repertoire, alternating between 1.e4, 1.d4, and 1.c4/Nf3. I was going to face
the strongest player I ever played. Prior to this, I held Topalov to a draw
some 15 years ago at the Olympiad. Obviously I was a little nervous and was
realistically afraid of getting smashed out in the opening. I was joking with
Tomas Krnan the night before that I would be live on stream in time trouble by
move 12th in a completely hopeless position. Fortunately, things turned out
much better for me than that, as you will soon see.} 1. c4 {A pleasant
surprise, I feel it's easier to solve ones problem vs a 2700 players against
this move than 1. d4 or 1.e4. Everybody has been playing 1. c4 against me in
this olympiad presumably because I've had some terrible theoretical gap in my
games 15 years ago; in fact I did! (see for example my game against Ftacnik).
On the other hand, it's a little presumptuous to think I wouldn't have worked
on this part of my repertoire. Actually, I equalize pretty comfortably against
1. c4 in all my games and whatever defeat I got were due to my poor play in
the middlegame.} c5 {I think it was Bologan who said: against 1.e4 the best
move is 1.e5, against 1.d4 d5 and against 1.c4 c5. There is some truth to that
statement, the classical openings gives you the best chance to neutralize the
advantage of the first move.} 2. Nf3 Nc6 {You have to blame Anton Kovalyov for
this choice.... at least indirectly. I'd decided to have a close look at this
system one week before the Olympiad when I saw one of his White games in Abu
Dhabi where he was lost against a weaker opponent after only 20 moves.} 3. Nc3
g6 {Before playing ...g6, on board 2 Kovalyov vs Nakamura had the following
position:} (3... Nf6 4. g3 d5 5. d4 e6 6. cxd5 Nxd5 7. Bg2 cxd4 8. Nxd4 Nxc3 9.
bxc3 Nxd4 10. Qxd4 Qxd4 11. cxd4 Bd6 {I thought about playing this line too
which is the preferred way to defend the black side of this variation by the
likes of Caruanna. It would have been funny to see the exact position on board
2-3 and see who would dare to deviate first.}) 4. e3 {This is the so-called
"refutation" of this system that has been recommended for many years. But
things are not so clear, and computers have rehabilitated the whole variation.
I've seen commentators on this game say 3... g6 is dubious, but nobody has yet
to come out with a refutation to back up their claim.} Nf6 5. d4 cxd4 {
Otherwise White plays d4-d5 with a space advantage.} 6. exd4 d5 7. Bg5 {
The other main line goes} (7. cxd5 Nxd5 8. Qb3 Nxc3 9. Bc4 Nd5 10. Bxd5 e6 11.
Bxc6+ bxc6 {where Black has the bishop pair and a nice d5 square for his queen.
But there is also a real positional danger to end up worse with the weak pawn
on c6, and especially if White succeeds in trading the dark-square bishop.})
7... Be6 {There is no other decent way to protect the d5 pawn. Incidentally,
this opening can also come from a Panov Caro-Kann.} 8. Bxf6 exf6 {So far we've
been following my game against Gawain Jones from England from a previous round.
I did lose very badly in that game, but it wasn't the opening's fault. I was
curious to see what kind of preparation Wesley had in store for me. I hadn't
found anything clearly promising for White, so I wasn't that worried.} 9. h3 $6
{Jones prefered:} (9. c5 Bg7 10. Bb5 O-O 11. h3 {with a balanced position.
Honestly, I don't understand So's preparation, Black can obtain a comfortable
position with the correct sequence of moves.}) 9... Bb4 $1 {Stronger than ...
Bg7.} 10. c5 O-O $6 {I played this obvious move pretty quickly, but Black has
a clear way to reach a nice position with:} (10... Bxc3+ $1 11. bxc3 O-O 12.
Bd3 b6 13. cxb6 axb6 14. O-O Na5 {follow by ...Qc7, ...Rfc8 and ...Nc4;
Black's position plays itself.}) 11. Rc1 {As mentioned by Anton Kovalyov after
the game, even though Black was OK, in practice the position is not so simple
to play for Black. The situation is strategically unbalanced and the queenside
pawn majority of White could prove a factor in the future. Obviously the
computer gives many ways for Black to play, but over the board after having
missed the strategically simple ...Bxc3+ I was left in an uncomfortable
situation.     I thought for a long time here and should probably just have
played the obvious ...b6 and just hope for the best. There is no point in
trying to work out the complications before they happened. My indecision on
obvious moves in this game will result in an uncomfortable zeitnot later in
the game.} b6 ({I was attracted by:} 11... Qe7 12. Be2 Rfe8 13. a3 (13. O-O $2
Bxc3 14. Rxc3 Bxh3 $1 $15) 13... Bxc3+ 14. Rxc3 Bf5 {but wasn't sure how good
it was. I also took some time deciding if I should throw out ...Qe7 or ...Re8
before playing ...b6. Anyway, like I said, it's pretty pointless spending time
on such subtleties, since the complications are to hard to work for a human.})
12. Bb5 Na5 13. a3 Bxc3+ 14. Rxc3 Nc4 15. b4 {We've reached a position
more-or-less forced after ...b6 that is strategically balanced. White has a
nice trump with his passed c-pawn, but Black has a superb knight on c4, plus
he is slightly ahead in development. I still needed to make something happen
quickly because in the long term the c5-pawn will guarantee White a positional
advantage.} a5 $6 {Kovalyov was critical of this move after the game. For him
it appears that opening up the queenside made no sense. He was actually right,
exchanging a couple of pawns doesn't really help my cause. It happens that I
am still OK, but a simpler solution existed:} (15... a6 $1 16. Bxc4 dxc4 17.
O-O b5 {gives Black a confortable position. The plan is simple: blockade the
d-pawn with ...Bd5, then double my rooks on the d-file to exert pressure on
the d4 pawn. I obviously saw this line and thought it would be fine, but
didn't like giving my opponent two connected passed pawns.}) 16. O-O axb4 17.
axb4 Qb8 $6 {Esthetically this move looks wrong, but it also introduces some
concreted tactical threats against the bishop on c4, and I couldn't see a
refutation so I went for it. As it turned out, like in most cases, the
intuitive and harmonious move ...Rb8 was best.} 18. Bxc4 $1 {After a prolonged
thought So finds the right solution. It appears at first sight that Black gets
a lot of pressure on the white pawn center, but the displacement of my queen
to b8 has created a nice tactical opportunity on the f6-pawn.} dxc4 19. Nd2 {
The logical follow up: the knight is ready to jump on e4 at the first
opportunity.} bxc5 20. bxc5 Qb4 (20... Rd8 {is simply answered by} 21. Nxc4 {
and White keep both of his pawns alive.}) 21. Qf3 $1 {As soon as So played
this move I realised I was in trouble. I was counting on the simple ...Bd5 to
meet Qf3, but unfortunately hadn't analysed any further. With the confidence
So exuded when playing Qf3 I immediately saw Ne4 look deadly once my queen
take on c3.} Bd5 $1 {Unwillingly I played this move realising the alternatives
are even worse. A good decision from my part, since the unbalanced nature of
the resulting position will give me more chances than simply getting ground
down in a simply worse position.} 22. Qxd5 $6 {Qg3 was stronger, but I think
most grandmasters would take on d5. It's really hard to see the computer-like
defense that Black has on move 26.} Qxc3 23. Ne4 Qd3 24. Nxf6+ Kh8 (24... Kg7
25. Qe5 $1 {is deadly}) 25. Nd7 Rfe8 {Preventing Qe5+ and praying that my
opponent may miss the treat of ...Qxf1+ followed by ...Ra1 mate!} 26. Ne5 Qf5
$2 {The silicion beasts are quick to point out that ...Qe2! would save Black.}
(26... Qe2 $1 27. Nxf7+ Kg7 28. Ne5 c3 {and the c-pawn gives enough
counterplay according to the computers. But honestly, it looks incredibly
scary to play this way in a game with such an exposed king.}) 27. Qxc4 $2 (27.
Nxf7+ $1 Kg7 28. Ne5 $18 {and Black can't really hold to his c-pawn because
its king is too weak.}) 27... Kg7 {Now I am back in the game, White is
obviously better, but it's not so easy to tell what's the best way to proceed..
.} 28. Qc3 $2 {Here, So began playing very fast -- and very badly -- trying to
exploit my time trouble. This was really not warranted, since he has a clearly
better position. Maybe he was feeling the pressure of the occasion and was
hoping for a mistake and a quick victory!?} (28. Ng4 $1 $16 {follow by Ne3 is
a nice way to reorganize the pieces and stop any counterplay, after which
White can start pushing his pawns.}) 28... Ra2 $1 {Suddenly Black has serious
counterplay!} 29. Nd3 $2 {Diagram [#] A historic position: I now had the
chance to help Canada change the fate of the whole Olympiad. This is how the
situation looked on the other board: Barrev was dead lost, Kovalyov was
slightly worse, but a draw was the likely result, and Hansen was close to
winning. I just needed a draw to stop the Americans from winning the gold
medal and propel Canada to 4th place. I was down to 3 minutes (+30 second
increment) to make 12 moves. At the time I didn't realise that the position
had changed dramatically from much worse to winning for Black. I guess I was
too pessimistic, just one move away and I couldn't reassess correctly the
position in time trouble. I unfortunately played a practical, fast move to
give me more time for the rest of the game and forgot to calculate lines
properly. There was quite an obvious move that simply wins for Black:} Qf6 $2 (
29... Ree2 $1 {Simple and strong. I would probably play this way even in
bullet chess. The rooks suddenly acquire incredible power by targeting f2 and
by stopping the passed pawns from behind. The threat is simply ...Red2 to kick
the knight and then take on f2. White has no choice but to go on a suicidal
mission just to stay alive:} 30. d5+ f6 31. c6 $1 (31. d6 $2 Rad2 $19) 31...
Red2 32. c7 Rac2 $1 33. c8=Q $1 Qxc8 {After this forced sequence, Black will
eventually win the d-pawn and get an easily winning position. This miss is
really heartbreaking, since everything is more or less forced after ...Ree2
and the extra 30s would have given me ample time to make the time control
confortably.}) 30. Nb4 Rae2 31. Nd5 Qg5 32. Nc7 R8e3 33. Qc1 h6 34. d5 Qe5 $4 {
Very low on time I missed a second chance to change the fate of the Olympiad.
This mistake is really bad since White has three different ways to win after
it.} (34... Rd3 $1 {which I saw was a clear draw, but I wasn't completely sure
of that in the game. For example:} 35. Qxg5 hxg5 36. Rc1 Rdd2 $1 {and the
classic repetition on the second rank saves Black. Every line goes more or
less like this. If White tries to prevent the perpetual he risks losing his
pawns.}) 35. fxe3 $18 Qg3 36. Qa1+ {I saw this move but missed the simple Qa8+
on ...Kg8. Nevertheless 36.Ne6+ and 36.Rf7+ also win.   Unfortunately, I
didn't make use of my chance handed to me on a gold platter by So. I guess
it's just the learning curve of adapting to 2700+ opponent and also a reminder.
. . to play faster!} 1-0