14 August 2017

A Training Tie-break

Day one of rapid chess at the Gelfand - Inarkiev match in Nazran in the review of Vladimir Barsky

Following the first two classical games in a matchup in Magas there took place a kind of tie-break: two rapid games featuring 25 minutes plus a 10-second increment per move followed by two games with 10 minutes plus a 10-second increment. The following is shared by Ernesto Inarkiev:

- This is a drill for the upcoming World Cup, in which a tie-break (in the case of equal points gained in the "main time") features a similar time control. Honestly, it has also come to me that this formula might help revive the classical matchups. For example, a match of 10 games is split into 5 rounds, 2 games each. We have seen the world championships in which people would adhere to the following strategy: the one with "+1" starts playing very safe chess, even as White. If the match is split into rounds, there is no sense in doing this any longer.

Let me clarify my point of view. If two classical games end in 1.5:0.5 (as in our case with Boris), there will be no rapid as he has won round one. If games three and four result in two draws, or the opponents trade blows, then a tie-break takes place until the positive result is achieved, as in the World Cup. If need arises, the onus may be placed on the Armageddon. Each round unveils its winner, while the total score is a sum of victories. Each game’s price will be completely different then, so that even a 2-0 lead is not a plain sailing to the final victory yet. In my opinion, this is an interesting enough system to impart a combative meaning to each classical game.

Gelfand – Inarkiev (m/3)

Queen's Gambit Accepted

1.d4 d5 2.c4 dxc4 3.e4 Nf6 4.e5 Nd5 5.Bxc4 Bf5

I hoped this line would catch Boris unawares, but it was quite the opposite - he was aware of the strongest continuation.

6.Qb3 c6 7.Bxd5 cxd5 8.Qxb7 e6

I was unwilling to liquidate into an endgame with 8...Qc8, as was the case with rapid game three. Could he have reacted stronger, I wonder?

9.Qxa8 Qb6


The engine points out to a stronger 10.Ne2! Bb4+ (also bad is 10...Be7 11.Nbc3 0-0 12.Nxd5 exd5 13.Qxd5 Bb4+ 14.Nc3 Nc6 15.Be3 Bd3 16.Rd1 Rd8 17.Qf3 Ba6 18.a3) 11.Bd2 Bxd2+ (11...0-0 12.Bxb4 Qxb4+ 13.Nbc3 Qb6 14.Na4 Qb4+ 15.Nec3 Qxd4 16.0-0) 12.Nxd2 0-0 13.Rc1 Bd3 14.a4 – and White gradually sets his queen free while retaining an extra rook.

10...Bb4 11.Nge2 0-0 12.0-0 Nc6 13.Qxf8+ Kxf8

While Black has a queen for a pair of rooks, White’s extra pawn is not felt at all, to my mind.

14.Be3 Bxc3 15.Nxc3

I do not think that 15.bxc3 is any stronger.

15...Qxb2 16.Rfc1 h6 17.Na4 Qb5 18.Nc5

The setup is a kind of fortress and is approximately equal. At first I decided against repeating moves, but then realized that I had no meaningful follow-up: he has closed the queenside and there is no headway on the kingside either. We have marked the time, but then Gelfand decided to push forward.

18...Kg8 19.h3 Qb2 20.Na4 Qb5 21.Nc5 Na5 22.Bd2 Nc4 23.a4 Qb6 24.Be1 a5 25.Ra2 h5 26.Re2 Kh7 27.f3 Kg8 28.Kh2 Bg6 29.g4

This is a normal move, to my mind. However, the subsequent play called White to handle his rooks delicately. Conversely, White allowed the queen into his camp.

29...Qd8 30.Bg3 Qb6 31.Be1 Qd8 32.Rc3 Kh7


Stronger is 33.f4!, taking care of the g5-square.

33...Qg5 34.Kg2 Qc1 35.Bf2 Nd2

In the end we just resigned ourselves to making self-suggesting moves, whereas it was crucial to have everything calculated precisely instead. However, there was no longer time for that.

36.Re1 Qc2 37.Rb8

Stronger is 37.Re2!

37...Qc3 38.Re3 Qc4 39.Re1 Qc3 40.Re3 Qxd4 41.Nd7 Bd3

Winning was 41...Qxa4! 42.Nf8+ Kh6 43.Re2 Nc4.


Here White could have turned the scales in his favor via: 42.Nf8+ Kh6 43.g5+! Kxg5 44.h4+ Kf5 45.Rb7 with an attack.


Winning nicely was 42...Bf1+! 43.Rxf1 Qd3 44.Re1 Qxf3+ 45.Kg1 hxg4.


Correct was 43.Be3!, upon which Black needs to come up with precise measures to maintain balance. Here are some basic lines: 43…h4! 44.Nf8+ Kh8 45.Nxe6+ Kh7 46.Nf8+ Kg8 47.e6 Qc7! 48.Nd7+ Kh7 49.Bf2 Nxf3 50.exf7! (but not 50.Kxf3? Be4+) 50...Nxe1+ 51.Bxe1 Qc2+ 52.Kg1 Qe2 53.Rh8+ Kg6 54.Re8 Qf1+ 55.Kh2 Qf4+ (there is no taking a pawn: 55...Kxf7 56.Rf8+ or 55...Qxf7 56.Ne5+) 56.Kg1 Qf1+ with a perpet.

43...Kh6 44.Be3+ g5 45.h4 Nxf3 46.hxg5+ Nxg5 47.Rc1 Qxe5 White resigns.


 Inarkiev – Gelfand (m/4)

Sicilian Defense

1.e4 c5 2.Nf3 Nc6 3.Bb5 g6

As far as I understand, the last year’s match of ours has set a trend as the Rossolimo system with g6 has become very popular. Prior to that, Dubov had had a very interesting game with Rublevsky, but the system has really caught on as a result of our match, tested in as many as four of its games.

References to those four games are given below.

4.Bxc6 bxc6 5.0-0 Bg7 6.Re1 Nh6 7.c3 0-0 8.h3

8.d4 cxd4 9.cxd4 d5 10.e5 f6 11.exf6 exf6 12.Nbd2 Re8 13.Nb3 Nf7 14.Nc5 Bg4, as in Inarkiev – Gelfand, Magas 21.7.2016 (rapid).


8...d5 9.d3 c4 10.dxc4 dxe4 11.Qxd8 Rxd8 12.Rxe4 e5 13.Re1 f6 14.Nbd2 Nf7 15.Nb3 a5 16.Be3 a4 17.Bb6 Re8 18.Nc5 g5 (18...Bf5 19.Nh4 Bc8 20.Rad1 f5 21.Nf3 e4 22.Nd4 Ne5, as in Inarkiev – Gelfand, Magas 21.7.2016, rapid) 19.b3 g4 20.hxg4 axb3 21.axb3 Bxg4 22.Nh4 f5, as in Inarkiev – Gelfand, Magas 19.7.2016.

9.e5 Nf7 10.d3 d6

10...Ba6 11.c4 e6 12.Nc3 g5 13.Be3 h5 14.h4 g4 15.Ng5 Nxe5 16.Bxc5 Ng6 17.g3 Re8 18.Qa4 Bb7 19.Qb4 Ba6 20.Qa3 Bb7 21.Qb4 Ba6 22.Qa3 Bb7 23.Qb4, draw, as in Inarkiev – Gelfand, Magas 15.7.2016.

11.Bf4 Ba6


This is an interesting idea. All in all, this is a complex setup with many lines possible. Boris opts for immediate simplifications.

12.e6 Ne5 13.Bxe5 dxe5 14.Nxe5 Qb8 15.d4 Bxe5 16.Rxe5 cxd4 17.cxd4 Qxb2 18.Nd2, as in Nepomniachtchi – Gelfand, Geneva 2017.

12.Nbd2 Qb8 13.exd6 exd6 14.Qc2 Qd8 15.h4 Qd7 16.Nc4 Rfe8 17.Rad1 Bxc4 18.dxc4 Rxe1+ 19.Rxe1 Re8 20.Rd1 Re4, as in Grischuk – Radjabov, Geneva 2017.

12...dxe5 13.Bxe5 Nxe5 14.Nxe5 Bxe5 15.Rxe5 Bxc4 16.Rxc5 Qxd3 17.Nc3 Qxd1+ 18.Rxd1 Bd5 19.Nxd5 cxd5 20.Rdxd5 Rfd8

A four-rook ending has appeared in a semi-forced fashion. I have not put Black up against any serious problems, allowing him to consolidate. It is difficult to say where exactly some heavy spanners could be thrown into the works of Black, but my actions were clearly underwhelming.

21.b4 Kf7 22.f4 a6 23.Kf2 Rxd5 24.Rxd5 Rc8

The black rook enters the game with a tempo.

25.Ra5 Rc4 26.Kf3

Risky is 26.a3 Rxf4+ 27.Kg3 Rf1!, and it is only White who is at danger of going down in this position as the e-pawn is running fast towards the queening square.

26...Rxb4 27.Rxa6 Rb2 28.g3 Rc2 29.Ra3 Kf6 30.h4 h6 31.Ke3 g5 32.hxg5+ hxg5 33.Ra6+ e6 34.fxg5+ Kxg5 35.Rxe6 Draw.


Gelfand – Inarkiev (m/5)

Queen's Gambit Accepted

1.d4 d5 2.c4 dxc4 3.e4 Nf6 4.e5 Nd5 5.Bxc4 Bf5 6.Qb3 c6 7.Bxd5 cxd5 8.Qxb7 Qc8

I decided to play it safe this time.

9.Qxc8+ Bxc8 10.Nc3 e6 11.f4 Nc6 12.Nf3 Ba6

The light-squared bishop is a very powerful piece, and I believe that Black is well compensated for the missing pawn. Nevertheless, White’s play is, perhaps, easier, while Black needs to handle this position with caution. This said, Gelfand has not followed up by carrying out any dangerous plan, honestly speaking.

13.Kf2 Rb8 14.Rd1 Kd7 15.b3 Be7 16.Na4 Rhc8 17.Be3 Nb4 18.Ne1 Bb5 19.Nc5+

Although the knight transfer to с5 seems logic, it has ended up offering White nothing in terms of benefits. It is hard to say if there was any headway for him on the kingside. With Black having a bishop pair, there is always a danger of overlooking bishops’ counterplay associated with opening up the position. Gelfand’s performance was very competent and reliable for the rapid chess, but it was not enough to bring the point home anyway. White has an extra pawn on the queenside, and my pieces are so full of life there that I am virtually up a bishop. Therefore, an extra pawn should not be a factor at all.

I was rather apprehensive of the kingside pawn storm, but the extent of White’s potential achievements there is not obvious.

19...Bxc5 20.dxc5 a5 21.Rd2 Rb7 22.Rc1 Ke8 23.Rc3 Rcb8 24.Nf3 Nc6 25.Rb2 Ba6 26.Nd4

It is clear that White has no potential with the knights off the board.

26...Kd7 27.Rd2 h5 28.g3 g6 29.Kg2 a4 30.bxa4 Rb4

Although my pieces are a lot more active, White has no weaknesses for me to hook up to.

31.Rb3 Nxd4 32.Rxb4 Rxb4 33.Rxd4 Bc4 34.Rd2 Rxa4 35.Kf2 Kc6 36.Rb2 Bb5 37.Rd2 Bc4 38.Rb2 Bb5 Draw.


Inarkiev – Gelfand (m/6)

Sicilian Defense

1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 Nc6

The Rauser system is an interesting approach. In principle, there is logic behind striving at a more complex type of struggle when playing a 10-minute game. I think Boris has long since faced the Rauser variation other than in blitz, maybe.

6.Bg5 e6 7.Qd2 a6 8.0-0-0 Bd7 9.Kb1 Nxd4 10.Qxd4 h6 11.Bc1

This is a relatively rare continuation.

11...Bc6 12.f3 Qc7

Boris opts for a long castle, but this is rather risky since а7-а6 has already been thrown in and his king is not going to be safe for that matter.


I understood that my opponent was about to castle long and I declined g2-g4 for that reason. My guess was: why play 12...Qc7, if you intend b7-b5? This is a purely rapid chess type of logic! This said, after 13.Be2 the advance 13...b5 should gain in strength since the bishop has deprived the knight of the retreat square е2.

13...0-0-0 14.Qf2 d5

This is an attempt to cut the Gordian knot. It is either going to work, or Black will be in bad shape before long.

15.Be3 Rd7 16.Bb6 Qb8 17.exd5 Nxd5 18.Nxd5 Rxd5 19.Rxd5 exd5

I wanted to meet 19...Bxd5 with 20.Qd4, threatening 21.Qc3+ Bc6 22.Bxa6. Besides, the queen starts eyeballing the g7-pawn.


This important lever accentuates the black king’s exposure.

20...Qe5 21.cxd5 Qxd5

If 21...Bxd5, then 22.Rc1+ Bc6 23.Bxa6.

22.Rd1 Qf5+ 23.Bd3

The engine points out to an immediate victory after 23.Ka1 Be7 24.Qg3! Bd8 25.Rxd8+.

Indeed, what a resource I missed, it would have been spectacular!



An ultimate finesse.

24...Qf6 25.Bd4 Qe6 26.Qc2

White’s pieces enter active positions with tempi.

26...Kb8 27.Bf5 Qe8 28.Qc3 f6 29.Qa5 Bxf3 30.Re1 Be7 31.Qb6 Qf7 32.Rc1 Black resigns.

* * *

In a classical time-control game, opponents get two points for a victory and one for a draw; in rapid - one and half a point respectively. Thus, the score in the match has become 4:4 (3:1 in Gelfand’s favor in the classical chess and 3:1 in Inarkiev’s favor in the rapid).

Pictures by Vladimir Barsky