Three tempi are worth a pawn, they say

In a few years when everybody has a quantum computer at home correspondence chess will be a pointless exercise. Hopefully this won’t happen until the World Championship cycle that just started on the Lechenicher Schachserver is finished. It’s my first one, and I’d like to make the finals at least once before the machines become strong enough to eliminate the human factor. So far things are going well, but there’s a few years of play ahead, and there will be much stronger opponents.

corr1.jpg
if only e4 was occupied. Almost a perfect rectangle of pieces e1-g1-g4-e4 after 18…Re3.

This is a game from the preliminary stage which will be followed by the quarterfinals, the semifinals and the final. It’s not an immortal game, but there are a few interesting cases of overruling the machine, and the final sequence, intiated by 23…Re3, is nice. It very much reminds me of the beautiful geometry in an earlier game („The Bodensee counter gambit in action“) when 18…Re3 (see diagram on the right) paralyzed the white position and made it collapse quickly.

This game represents the latest addition to our site’s database of annotated games and opening surveys. I’ve also put all the moves and lines below into a Lichess study for you to replay and analyze more conveniently.

Villain (1969) – Hero (2211)
Lechenicher Schachserver 2017, World Championship Cycle 2021

1. d4 Nf6

1…d5 is the safer move. In anything d4/d5 White will have a hard time to get pressure and a even harder time to keep it up. As an outrated player he will also have it easier to be solid and avoid a loss.

In case Black outrates White like in this game things are different. In order to beat a guy and his engine with Black complexity and imbalances are required. One needs to take risks to achieve this. 1…Nf6 and 2…e6 seems like a manageable risk to me.

2. Nf3 e6 3. g3 b5 4. Qd3

corr2.jpg

My initial thought was that attacking b5 with the early queen excursion 4.Qd3 is more a joke than a serious move. But then I saw that it had been played by a bunch of serious guys (Hübner, Svidler, Mamedyarov among others), and I understood that Qd3 not only attacks b5, it also prepares e2-e4.

Still, a crude move like that makes me want to get into close combat immediately, even with Black. Already before checking the database and firing up the engine I had made up my mind that I very much wanted to go 4…c5 and spice things up.

Once I had Stockfish running, the engine came up with joke moves itself: Stockfish wants to go 4…Ba6, 4…c6 or 4…a6 (in that order). While 4…Ba6 looks interesting, I don’t like 4…a6 (chosen by a vast majority according to my Megabase, but it only justifies 4.Qd3) and I hate 4…c6. How can a move like that be the 2nd choice of a 3.400 player? It looks even more ugly than 4.Qd3. 4…c6 has been played by Seirawan, though. I assume even the great Yasser can have a bad day occasionally (or the opposite, play like a superhuman with 3.400 Elo).

4… c5 5. Qxb5 Nc6

corr3.jpg

6. c3

(6. dxc5?! Rb8 7. Qc4 Qa5+ 8. Bd2 Qxc5 9. Qxc5 Bxc5 Even with queens off the
board Black has more than enough compensation.)

6… Rb8 7. Qd3

corr4.jpg

„Three tempi are worth a pawn“, they say. If that’s true, Black should be fine due to Qd1-d3xb5-d3.

7… cxd4 8. Nxd4

(8. cxd4 Qa5+ 9. Nc3 Ba6 and Black is slightly better already.)

corr5.jpg

8… Bb7

White has trouble to develop into a meaningful setup. 9.Bg2?? is desirable, but it loses material due to 9….Ne5 -+.

(8… Ne5 Immdiately jumping at White’s throat is possible, but doesn’t lead to much. 9.Dc2 Bb7 10. f3 Qc7 11. Nd2 and Black has compensation, but nothing more. After the impetuous 11… Neg4?! 12. e4 Ne3 13. Qd3 Nxf1 14. Kxf1 White consolidates and will be just a pawn up soon.)

9. f3

So far Black can be happy with the opening, but I wasn’t sure how to move on from here, especially what I want to do with my e and d pawn. Build a big center, put them on e6/d6 or just go …d5 (and maybe combine it with …Qb6 and …Ba6)? There are plenty of options, each of them seems reasonable. Since I wasn’t sure how to set myself up, I decided to just continue developing, put my guys on better squares while watching out for an opportunity to force things.

9… Be7 10. Bf4?!

corr6.jpg

Probably not the most accurate of moves. Black wants to go …d6 anyway, so why provoke it, and while b2 is undefended, I don’t see what the bishop on f4 accomplishes except presenting Black a piece to drive back with tempo. If I were White I wouldn’t touch the Bc1, go 10.Bg2 instead, make it a priority to castle, then go e2-e4 and only then play Be3.

10… d6 11. Nxc6 Bxc6 12. b3 O-O

corr7.jpg

White has wasted a tempo for 10.Bf4, had to invest another one for 12.b3. Due to this additional loss of time he can’t continue development with Bg2 and 0-0 anymore. Black has the upper hand already.

13. Be3

Admits that 10.Bf4 wasn’t great.

(13. Bg2 Bb5 14. c4 Nd5! threatening …Bf6, and White is in trouble.)

13… Qc7 14. Bg2

corr8.jpg

14… Ng4

I wasn’t exactly sure how to follow this up, but White’s play so far felt like it needed to be punished.

(14… e5 15. Nd2 d5 was a less forcing way to push the black advantage.)

15. Bd4

Luring the e6 pawn to e5 may not be stupid. For now it eliminates all black ideas connected to …Bf6 and …Ne5.

(15. O-O Bb5 16. c4 Nxe3 17. Qxe3 d5 (threat …Bc5) and Black is much better.)

15… e5

corr9.jpg

16. Bg1

(16. O-O Qb7 was possible, not especially tempting though. White has achieved castling, but remains under pressure.)

16… f5

According to Stockfish this spoils the black advantage. The machine very much wants Black to calmly improve via …d5, …Rfd8, …a5 and so on and claims something like -0.5. However, with the white king trapped in the centre for now, it felt to me like tearing things up in the middle of the board was the logical course of action (and prepared by 14…Ng4 after all). Now that I annotate the game and have Houdini running, I am relieved to see that other engines don’t hate 16…f5 quite as much as Stockfish does.

(16… a5 17. Nd2 Rfd8 18. Rd1 Nf6 19. Bf2 d5 =+ is how Stockfish would play as Black.)

17. Nd2

corr10.jpg

And now? The engine correctly insists that Black needs to close the b1-f7 diagonal before he can force the action. The choice between 17…Rf7 and 17…Kh8 is not easy, both lead to incredible complications, and after both I couldn’t find a clear path to killing the white king (or any other kind of decisive advantage). In the end I chose the one that seemed to offer White more chances to go wrong.

17… Rf7 18. Rd1 Bg5

corr11.jpg

Now everything is activated maximally. Next step: open lines against the white king. Watch out for the e3 square, it will be the key square in the final sequence of the game. It’s hard for White to find a constructive defense.

19. h4

(19. h3 It would be understandable if White wanted to chase away the Ng4, but after 19…Nf6 followed by …Nh5 he would soon feel the weakness of his g3 square. (19… e4 20. fxe4 Ne5 is another nice option for Black.))

19… e4 20. fxe4 Bh6 21. Bd4

corr19.jpg

Only move, preparing 0-0. White needs to get his king out of the centre, before … Re8 and/or …d5 hit him.

21… Re8 22. O-O d5

corr13.jpg

The e file will open up, Black will gain access to e3, and the queen starts to eye towards g3.

23. exd5?

Now that he has managed to castle White becomes careless. Opening the e file is deadly.

(23. e5 needed to be played in order to stay in the game. 23… Nxe5 24. Bxe5 Rxe5 25. Rf3 Qb6+ 26. Kh1 g6 White’s position isn’t fun to play, but he remains a pawn up and is still fighting.)

23… Re3!

corr14.jpg

The Bd4 is White’s key defender of the black squares. Now it is cut off from defending the kingside with …Qxg3 is coming.

(23… Be3+? The obvious move is not good enough to crack the White position. 24. Bxe3 Rxe3 25. Qd4 Qxg3 26. Rf3 Qh2+ 27. Kf1 Bb5 28. Qxe3 (28. Rxe3?! f4 29. Rf3 Ne3+ 30. Qxe3 fxe3 31. Rxf7 Kxf7 32. c4 Qxh4 33. Ne4 Bd7 with a large black advantage) 28… Nxe3+ 29. Rxe3 Qxh4 and White may hold.)

24. Bxe3 Bxe3+ 25. Kh1 Qxg3 26. Nf3 Bd7

corr15.jpg

Eventually the Bc6 needed to move, so White has gained a tempo to organize some kind of defense, but he is helpless with all his pieces (except the queen) being tied down.

27. Qc4

Black would like to go …R7-f6-h6xh4 to finish the game, but unfortunately after 27…Rf6 28.d6+ the white queen on its own can still create a bit of counterplay. However, there is another way to get the rook into the attack:

27… g5! 28. hxg5 Rg7

corr16.jpg

Threatening …Rg7xg5-h5+

29. d6+ Kh8 30. Rd3

corr17.jpg

Now after 30…Rxg5 31.Qf7 holds, but there are other options…

30… h5 

…like …h5-h4-h3. White remains helpless.

31. Qxg4

(31. gxh6 Rg6)

31… fxg4 32. Rxe3 h4

corr18.jpg

White resigned.

0-1

Kommentar verfassen

Trage deine Daten unten ein oder klicke ein Icon um dich einzuloggen:

WordPress.com-Logo

Du kommentierst mit Deinem WordPress.com-Konto. Abmelden /  Ändern )

Google Foto

Du kommentierst mit Deinem Google-Konto. Abmelden /  Ändern )

Twitter-Bild

Du kommentierst mit Deinem Twitter-Konto. Abmelden /  Ändern )

Facebook-Foto

Du kommentierst mit Deinem Facebook-Konto. Abmelden /  Ändern )

Verbinde mit %s