The start of the story
Hi everybody ! I was about to call it a day after playing mindless chess 960 for quite some time. And then I saw that it was 16:00 and I remembered that there should be a simul hosted by the Improver's group. So I checked out, to see how it was going. It turns out that it hadn't started and some participants didn't show up ! So I quickly messaged the organiser to see if I had a chance to play ... and I managed to join the simul. I was really unprepared, my brain was on auto-mode and I just finished quite a few 960 games. Not sure if I could do my best but nonetheless, it was a unique opportunity !
The actual game
As I was too lazy to prep during the 10/20 minutes before the game, I didn't really know what to expect. The game started with a surprise at move 2.Nc3 ! It turns out we were heading into a closed sicilian, something that I am not used to seeing [to be honest, something that I have never seen]. As you probably know from my previous blog, I love playing the sicilian against 1.e4 for one reason : the OPEN sicilian !! At below 2000, most players go for wing-gambit sicilians [or just any unsound sicilians] or anti-sicilians .... and I don't remember the last time I saw a closed sicilian. And so, I didn't know any of the meta / ideas of this opening. But this is what we got :
Ok so my initial reasoning was : On move 2, I played the usual 2. ... d6 because I was still hoping that it transposes to an Open Sicilian, maybe a Najdorf .... even though it is probably unlikely if my opponent played 2.Nc3.
After we get this position, what is my plan ? I went for 5. ... g6, not because I'm a fan of the dragon setup .... I studied some uncommon sicilians / anti sicilians like the Moscow / Canal attack and going for g6 and fianchettoing the dark squared bishop is by far the best structure. Sure, going e6 and a scheveningen setup is also possible, but I'd say black is probably too passive to hope for a win.
Up to here everything has been played thousands of times according to the master database.
So, actually the next thing I did was incorrect. My idea is to go e4, creating some sort of Maroczy bind structure and have strong control of d4 and f4 squares. Say for instance, if white goes in with Nd5, I can just take and black is completely fine. The best move here for white is f5, a positional sacrifice ! The reason is simple : white has a kingside attack, which could already be started here. The point of f5, is after black takes, white simply castles. Notice that black's g6 pawn is gone, and the bishop is left without protection. Furthermore, black needs to think if they want to castle kingside ??? [Probably not]
My opponent decided to castle instead of f5, which is also completely fine. Here I saw the kingside pressure, but I didn't see a concrete threat, so I castled. Again, white's best plan is to go f5 for the same idea. [Eventually, you will see that my opponent did go f5 to claim a big advantage]
Here's the most chaotic part of the game : After Nh4, I had a chance to seize an advantage. Even though Nh4 seems to be a reasonable move to play (bringing one more attacker closer to the black king), black can actually start the attack before white, in which case, it is hard to tell who is attacking who ! For instance, say exf4, white needs to make a big decision : with which piece do they take back ? gxf4 falls into Nxe4, picking up a the knight on h4 and winning a pawn ; if Bxf4, then what about h6, Nf3, g5 ? [according to the engine, the a knight sac is beneficial for white ; but this is unclear to me] also in this case g5 is a fork if white plays something else.
So let's continue. I played Bg4, winning a tempo, and my opponent went for Qd2. Here, again the best logical continuation is exf4 and questionning white's kingside attack and also how they want to take back. I went Bh6, I thought that it was a smart move to limit the f pawn advancement with the pin, but actually white just needs to move their queen and everything's alright. xD
The next move I played is the FATAL mistake of the game. I played Bg7 as the bishop doesn't belong to this diagonal but just after I played this move, I realised that my bishop is trapped after f4 !!! Here I thought that the evaluation was at least +1 [My king is being attacked and I lose a important defender but I get 2 useless pawns in exchange ; can be interesting if we ever go to an endgame which is unlikely]
After exf5, I played Kh1. The idea is simple : I want to play Rg8 and argue that I'm not completely lost and counterplay is coming on the white king. But just look at the resulting position, let's count the number of attackers and defenders : white has 5 pieces attacking and black has 4 pieces defending ........ so the game is lost unless white blunders. After a few moves, we get :
This is the decisive phase of the game. I had to find this beautiful idea of Qc8 followed by Qc6 followed by the nasty d5 trap! It's really hard to find this because I looked at other things such as sacrificing the rook for counterplay and also because I didn't have time to calculate all of that obviously ..... these defensive moves remind me of some puzzles from the "hard" section of the woodpecker .... And so, I didn't find that and I lost the game.
How to play the closed sicilian as black ?
We saw in this game that black was super passive, didn't have a plan and was just trying to hold on for dear life. White's plan is clear and is always the same : launch the f,g,h pawns, bring out the pieces towards black's king. So, I did some research.
-First off, playing e5 is a mistake because it insists on that white plays f4-f5 and the black squared bishop has no scope. But this is true UNLESS white plays Ne2, in which case there is no Nh4 trying to support f5 pawn advancement.
- Our main plan is to find counterplay on the queen side, by going a6/a5, b5/b4, Rb8. Of course, Rb8 can be played before or after b5, depending if white goes Nc3, early on or not.
- a6 is played if and only if white proceeds to go a4, trying to stop our queenside attack
- Of course at some point, we will castle kingside despite white's attack.
- Another important idea for black : at some point, going Nd7 and then Ne5 (if needed i.e. if white goes all in). Nd7 opens up the dark squared bishop and also after, say f5, the diagonal is very weak, not well protected. Furthermore, f5 will allow our knights to jump to the e-square.
- Once the queenside pawns are cleaned, black can sometimes procceed to Nb4 and Ba6, overtaking the queenside.
How to procceed : Often times black's queenside attack is more deadly than white's kingside attack. So going b5,b4 is the main priority. Most of the time, the queenside pawns will be exchanged and black will take over the open files, and maybe go for the white king, by going to white's kingside. Defensive measures should be taken only when necessary (i.e. to prevent mate).
Final notes
Now that I checked the master database, I probably shouldn't be so scared of the closed sicilian : After 3.g3, win rates are the following : 26% for white / 38% draw and 36% for black. So yeah, black is also playing for a win and should have good chances if they know what they are doing ! Of course, I didn't know what I should be doing during the game ....
An additional note : I just checked on openingtree.com, what my opponent plays. On a sample of the 2000 most recent games, they mainly go for 1.d4, 1. c4 (logical as it can often transpose to playing d4 openings), and 1.e4. And I was curious, what exactly does he play against the Sicilian ? Well he mostly just goes for the closed sicilian ...
[EDIT : In the closed sicilian, after fianchettoing the dark squared bishop we can also go Nh6 to stop f4 and then go f5, maybe e5 at some point and bringing the knight on f7]
