IN round one of the Colonia de Sant Jordi U2400 I had white against a grandmaster rated 2354.
The full game can be seen at
M1 but here I want to concentrate on the opening, which I generally take as lasting until one player has connected rooks.
Sicilian Bb5(+)
The game began 1.e4 c5 2.Nf3 Nc6 3.Bb5 g6, and I continued 4.c3!?
There are 2,040 examples of this move in ChessBase's 2022 Mega database, well behind the 13,656 examples of 4.Bxc6 and the 20,009 examples of 4.0-0.
Joe Gallagher in Beating The Anti-Sicilians (Batsford, 1994) states: "White is hoping to lure Black into a dangerous gambit." He cites the game David Bronstein - Efim Geller, Interzonal (Gothenburg, Sweden) 1955, which continued 4...Bg7 (this is the commonest move in Mega22, but Gallagher calls it dubious) 5.d4 Qb6 6.a4 ("White has to give up a pawn to retain the initiative, but at least it will be a dangerous one") cxd4 7.0-0 a6 8.Bxc6 Qxc6 9.cxd4 Qxe4 10.Nc3, which gives White lots of compensation for a pawn (1-0, 23 moves). Gallagher says "the safest reaction" is 4...Nf6, which is what my opponent played.
Steffen Pedersen in Easy Guide To The Bb5 Sicilian (Everyman Chess, 1999) by implication recommends 4...Bg7, meeting 5.d4 with 5...Qa5!?, stating: "I am surprised I have been able to find so few games with this move." He treats 6.Bxc6 as the main reply, for which see Richard Palliser below, but says 6.Qe2!? "might be a better try [Stockfish15 and Komodo13.02 strongly disagree]."
Dorian Rogozenko in Anti-Sicilians: A Guide For Black (Gambit, 2003) claims that after 4.c3!? "it is not so easy for Black to emerge from the opening with a good position." He adds: "Compared to 4.0-0 Bg7 5.c3 here Black cannot castle so quickly and must fight against White's initiative with his king in the centre. On the other hand, White's king also remains on e1 - a factor that can be exploited by Black if he plays energetically enough." Rogozenko, like Gallagher, recommends 4...Nf6.
Palliser in The Bb5 Sicilian (Everyman Chess, 2005) also, by implication, recommends 4...Nf6, but says 4...Bg7 "is also possible," adding: "This was considered dubious due to an immediate 5.d4 when Black is advised to steer clear of Bronstein's 5...Qb6. Pedersen drew attention to Black's best option of 5...Qa5! and this has subsequently received a few more tests, although it remains surprising that this aggressive queen move isn't a lot more popular. Practice has focussed on 6.Bxc6 dxc6 7.0-0 when 7...cxd4! 8.cxd4 Bg4 9.Be3 Nf6 10.Nbd2 0-0 11.Qc2 Nd7! and ...e5 gave Black fully sufficient counterplay" in Kiril Georgiev (2532) - George-Gabriel Grigore (2508), Olympiad (Istanbul) 2000 (½–½, 23 moves).
Summing up, Black's commonest response in ChessBase's 2022 Mega database is 4...Bg7, but the theoretical preference is, narrowly, to opt for 4...Nf6, challenging the e4 pawn at a time when it cannot be defended smoothly by Nc3.
The replies 5.e5 (the only move covered by Palliser) and 5.Qe2 are almost equally popular. Ideas behind the latter include playing for two pawns abreast on e4 and d4, and vacating the d1 square for the king's rook to shadow the black queen.
After 5...Bg7 games usually continue 6.0-0 0-0 and either 7.d4 or 7.Rd1, but I preferred immediately occupying the centre with 6.d4 (not covered by Gallagher).
There followed 6...cxd4 7.cxd4, when the question becomes: can Black exploit White's lack of castling?
 |
| How should Black continue? |
*****
*****
*****
*****
*****
Checking with the queen by 7...Qa5+?! merely helps White develop with 8.Bd2 as the white light-square bishop is protected by the white queen.
7...Qb6!? tries to overload the white queen, which is defending b5 and e4, and adds a second attacker to d4. But the engines reckon White can happily reply 8.Nc3 as 8...Nxd4 9.Nxd4 Qxd4 10.Be3 gives White a lot of compensation for a pawn, eg 10...Qb4 11.0-0 0-0 12.e5 Ne8 13.Nd5.
Accordingly the game saw 7...d5, immediately challenging the white centre.
Developing with 8.Nc3, and so allowing ...dxe4, is not liked by the engines. not least because the advance 8.e5 comes with gain of tempo by hitting the black king's knight.
After 8...Ne4 9.0-0 0-0 it is time to decide what to do about the advanced king's knight.
Trying to trap it with 10.Bxc6 bxc6 11.Ne1? leaves Black well on top after both 11...Qb6 and 11...f6!?, according to the engines, eg 11...f6!? 12.f3 runs into 12...Qb6, when 13.fxe4 Ba6 14.Qe3 Bxf1 15.Kxf1 fxe5+ 16.Nf3 exd4 is horrible for White.
I played 10.Nc3, reaching the following position:
 |
| Rogozenko calls this "complicated (and) double-edged" |
*****
*****
*****
*****
*****
10...Bf5!?
Rogozenko gives 10...Nxc3 11.bxc3, commenting: "Black plays ...Bg4 at some point, and then takes on f3, followed by ...Re8 and ...Na5 with counterplay on the c file. White will have to take back with the queen on f3, which means that he lost a tempo by playing first Qe2 and then Qxf3. This is certainly a reasonable option for Black."
The downside of 10...Nxc3 is that it strengthens the white centre, albeit giving White a backward c pawn. Stockfish15 likes the game continuation; Komodo13.02 also does for a while, but comes to marginally prefer 10...Nxc3.
I replied 11.Bxc6!?, giving up the bishop-pair but ensuring White will not get a backward c pawn. More popular in Mega22, but the sample size is small, is the passive-looking 11.Be3, when 11...Nxc3 12.bxc3 gives Black c-file counterplay after 12...Rc8 or 12...Na5.
Following 11...bxc6 I went after the light-square bishop with 12.Nh4 (Bronstein in a 1975 game played 12.Na4).
In Levon Aronian (2758) - Quang Liêm Lê (2708), Fide World Cup (Khanty-Mansiysk, Russia) 2019, Black allowed the bishop to be captured, viz: 12...e6!? 13.Nxf5 exf5 14.Na4 c5!? 15.Nxc5 Nxc5 16.dxc5 Qc7 17.f4 Qxc5+, when White has at best a slight pull, according to the engines (½–½, 32 moves).
My opponent preferred to preserve the bishop-pair with 12...Nxc3 13.bxc3 Bc8, and I continued with 14.Ba3.
 |
| White has connected rooks - how would you assess the coming middlegame? |
*****
*****
*****
*****
*****
Black has the bishop-pair but White has a lead in development and more central-kingside space. Stockfish15 gives White the upper hand, but Komodo13.02 reckons the position is equal. I suspect the truth is somewhere in-between, and, given much more time, Komodo13.02 comes to rate White as slightly better.
