Some key takeaways from the game, for Caro-Kann players and in general:
- The benefit to Black of exchanging light-square bishops
- Conditions for being able to defend against White's ideas for pressure on the h-file
- The key role of queenside counterplay (see moves 23-25) so as to not give White a free hand on the kingside
- The recurrence of tactical ideas such as Ne4-d2, which eventually becomes decisive for Black
- How either player could have chosen to go for a drawing line (perpetual) at different times
[Event "81st Tata Steel Masters 2019"]
[Site "Wijk aan Zee"]
[Date "2019.01.13"]
[Round "2"]
[White "Duda, Jan-Krzysztof"]
[Black "Van Foreest, Jorden"]
[Result "0-1"]
[ECO "B13"]
[WhiteElo "2738"]
[BlackElo "2612"]
[Annotator "ChessAdmin/Komodo 11.2"]
[PlyCount "100"]
[EventDate "2019.??.??"]
1. e4 c6 2. d4 d5 3. exd5 cxd5 4. Bf4 {an alternative try to the usual Bd3.
White accelerates his seizure of the h2-b8 diagonal and does not let Black's
queen get to c7.} Nc6 5. c3 Nf6 (5... Bf5 {is in alternative way to play,
anticipating the Bd3 development by White and looking to exchange off bishops
early. In the game, this happens much later and also results in a change of
Black's kingside pawn structure.}) 6. Nd2 {the second most popular move in the
database. White prepares to support Ngf3 after Black's next.} Bg4 7. Qb3 {
a normal reaction by White, once the bishop ceases protecting the b7 pawn.} Qc8
{this is less committal than the alternative ...Na5.} 8. Ngf3 e6 9. Bd3 (9. Ne5
$5 {is an interesting idea here, but simply retreating the bishop to f5 seems
to take any sting out of it.} Bf5 $11) 9... Bh5 {a relatively rare option,
with ...Be7 being standard. The idea here is to retreat to g6 and exchange off
the Bd3, which otherwise is well-positioned to target Black's kingside.} 10.
O-O Bg6 11. Bxg6 {Komodo, in contrast to some other engines, assesses that it
is better for White to not exchange on g6. For White, often the idea behind
this exchange is to create a target for an h-pawn thrust, which is what in
fact occurs later on.} hxg6 12. Rae1 {committing to a central/kingside
strategy.} Be7 $11 {Taking stock of the position, Black has full equality. The
light-squared bishop exchange has left Black solid on the kingside and White
has no obvious advantage, although he can try for play on the h-file, as in
the game. Duda's next move is a novelty in the database and was likely an
attempt to introduce some uncertainty into the position, while again looking
for chances on the h-file.} 13. g3 $146 {as we will see later, the idea of
this move is to prepare the pawn thrust h2-h4.} (13. Ne5 {is a typical try by
White that scores well in the database (67 percent), but should not
objectively be a concern to Black. The database figures also seem to be skewed
by lower-rated games.} Nxe5 14. dxe5 Nd7 {similar to the game continuation
looks fine for Black.}) 13... O-O {no reason to postpone castling.} 14. Ne5
Nxe5 {Black chooses to remove the well-placed central knight immediately.
Although not forced, this is an excellent defensive idea, as otherwise White
typically starts developing tactical ideas on the e-file to target the e6 pawn,
with a knight sacrifice a possibility later on g6 or f7.} 15. dxe5 Nd7 16. h4 {
White follows up on his earlier idea of g3. Should Black be worried? As the
defender in this type of position, the usual assessment is that after two
minor piece exchanges, Black should be all right, since the sacrificial
possibilities by White are limited. White will also need time to bring his
other pieces to bear on the kingside.} Nc5 {puts the knight on its best square,
kicking the queen and eyeing e4 and d3.} 17. Qc2 Qc6 {putting the queen on the
long diagonal and improving its mobility. Now White's lack of a light-squared
bishop to oppose it is highlighted.} (17... b5 {is what the engines like in
this position, following up with ...Qb7 to put the queen on the long diagonal
and give black the option of a minority attack on the queenside.}) 18. Re3 {
this just ends up being awkward for White and wasting time. Presumably the
idea was to eventually transfer the rook along the third rank.} (18. Nf3 $5)
18... Qa6 {pressuring the a-pawn and also placing itself on another useful
diagonal.} 19. Qb1 {with this move, it's clear that White no longer has the
initiative and must start responding to Black's threats.} Rac8 20. Kg2 {
needed to clear the first rank for his rook to shift to h1.} b5 {now Van
Foreest plays the pawn advance, gaining space and with the eventual idea of a
minority attack along with the a-pawn.} 21. Bg5 Qb7 {although Black has been
moving his queen often, each time it has been with a purpose and has improved
his relative position. This time is no different, as the queen will still be
well-placed on e7 after the exchange of bishops, and the a-pawn is now free to
advance.} 22. Bxe7 Qxe7 23. Rh1 a5 24. Qd1 b4 {Black's counterplay on the
queenside balances White's play on the kingside.} 25. Qg4 (25. h5 {would
amount to the same thing after} bxc3) 25... bxc3 {active defense.} (25... Qb7
$5 {is another interesting way to defend.} 26. h5 d4+ 27. Rf3 bxc3 28. bxc3
dxc3 29. hxg6 fxg6 30. Qxg6 Rf5 $11 {and White has nothing better than a
perpetual.}) 26. bxc3 Rb8 {Black again correctly emphasizes counterplay,
threatening to go to b2 with his rook.} 27. h5 g5 {Van Foreest goes for the
option that is equal, but allows him to keep going in hopes of a win. The
correct decision, as it turns out.} (27... Rb2 28. hxg6 fxg6 29. Qxg6 Rxd2 30.
Qh7+ Kf7 {is a perpetual for White.}) 28. h6 g6 {this defensive idea should be
well known to Caro-Kann players, to prevent a breakthrough on the h-file.} 29.
Nf3 {targeting the weak g5 pawn, but} Ne4 {holds everything together for Black.
} 30. Re2 {covering the second rank against the threat of ...Rb2. However,
this leaves the c-pawn hanging.} Nxc3 31. Rc2 Ne4 32. Nd4 {although Black is a
pawn up, now White gets to have equivalent play for it, gaining the initiative
in compensation. The main threat here is the knight fork on c6.} Rbc8 33. Nc6
Qa3 {preserving the a-pawn.} 34. Rhc1 Kh7 {another important defensive idea
for Black, blockading the h-pawn to prevent a future sacrifice on h7. The
h-pawn is now a long-term liability for White, especially in an endgame, when
Black opening the h-file by capturing the pawn will no longer be of
consequence. Black does have to be careful to manage potential threats from
any occupation of the f6 square by White, however.} 35. Qe2 {now Black has to
take care of his Ne4, which is out of squares after a White pawn advances to
f3. Previously it could have gone to d2, with the tactical idea of Rxd2
followed by ...Qxc1. The Qe2 instead now covers the d2 square.} g4 {Black
gives back the material so his pieces regain freedom of movement.} 36. Qxg4 Rc7
37. Qf4 {this appears to be a try by White to maintain winning chances.} (37.
f3 {leads to lines where Black can win the a-pawn, but his queen is too
exposed to White's rooks for it to matter, so essentially White can get a
perpetual on the queen.} Nd2 38. Rc3 Qb2 (38... Qxa2 {doesn't gain Black
anything either} 39. R3c2 Qa3 40. Rc3 Nb3 $11) 39. R3c2 Qa3 $11 {with a
repetition of position to follow, whether Black takes the a2 pawn or not.})
37... Rfc8 $15 {obvious and good. Now White has to be careful about the Nc6,
which has only one viable square (d4), as well as his currently
under-protected rooks.} 38. g4 $6 {White likely underestimated the
consequences of Black's next move.} (38. Qg4 Qd3 $15) (38. Nd4 {bailing out
with the knight is likely the best (and most practical) option.} g5 {now has
much less sting, since the g4 square is available for the queen.} 39. Qg4 Rxc2
40. Rxc2 Rxc2 41. Nxc2 Qf8 $15) 38... g5 $1 39. Qh2 (39. Qf3 Qxf3+ 40. Kxf3 f6
41. exf6 Nxf6 {and the h-pawn will fall after} 42. Nd4 Rxc2 43. Rxc2 Rxc2 44.
Nxc2 e5 {first preventing the knight's return to d4} 45. Ne3 Kxh6 $17 {and
Black (at the GM level) should be able to convert the endgame with his extra
passed d-pawn.}) 39... Nd2 $19 {using the same tactical idea with the knight
move to d2 as before, but now threatening ... Qf3+. By this point, White is
essentially lost.} 40. Rc3 (40. Qg3 {is objectively better, but still leads to
a lost endgame, so White tries something else.} Qxg3+ 41. Kxg3 Kxh6 $19) 40...
Qa4 {the only winning move, threatening the g4 pawn. It's not too hard to find,
though.} 41. Qh5 {protecting the pawn, but now Black's queen dominates in the
center.} Qe4+ 42. Kh3 (42. f3 Qe2+ 43. Kg3 Nf1+ $19) 42... d4 {passed pawns
must be pushed! Now White's position collapses all over the board.} 43. Rg3 (
43. Qxg5 $2 Rg8 $19) 43... Qf4 {threatening the f2 pawn as well as ...Ne4 with
a double attack on the rooks.} 44. Rd1 Ne4 {an elegant finish.} (44... Rxc6 {
is straightforward, but perhaps required a bit more calculation. Or Black just
preferred winning with the text move.} 45. Rxd2 Rc3 (45... Qxd2 $2 46. Qxf7+ {
with a perpetual.}) 46. Rb2 Rxg3+ 47. fxg3 Qf1+ 48. Rg2 Rc2 $19) 45. Rxd4 Nxg3
46. fxg3 (46. Rxf4 Nxh5 {and Black is a full rook up, with the Nc6 next to
fall.}) 46... Qf1+ 47. Kh2 Rxc6 48. Qxg5 Rc2+ 49. Rd2 Rxd2+ 50. Qxd2 Rc1 0-1
