lichess.org
Donate

Sicilian Najdorf, English Attack 8...h5 *Question*

I'm playing black on April 15th. My opponent is roughly 1800 USCF and will play the English Attack vs my Najdorf. I want to prepare the h5 variation to prevent him from going into typical pawn pushing (g2-g4-g5). Could advanced players help me out with the ideas for black in the position below? Thanks!
Ok, looking closer those games are all 2008 or older. As you'd know the theory in the Najdorf does not sit still!
Still it might give you some good pattern recognition. For more modern games of a super GM to follow I'm not sure if any play the 8...h5 sideline. Gelfand dropped the Najdorf for many years, but has recently returned to it. MVL would be the other one to check out.
#2. Thanks! The najdorf is always changing and h5 definitely isn't the mainline, but that's part of the reason I like it. I haven't found any written ideas on h5 and how to attack/defend (part of what I'm looking for in the forum) but playing over lots of games will be helpful.
Ok, I fired up my database (big2015 +TWIC) to take a look at some recent games. Remember, I have absolutely no practice in this line!! Hopefully someone else can take over giving advice here.
First off a stack of super GMs have dabbled in it for black in the last five years; Naka, Giri(!) Gelfand, Karjakin. But Leinier Dominguez Perez seems to be its main defender as black (although he also drew a game with Carlsen from the White side too). He'd be the man it'd look at to learn the line.
As for specifics I really can't give any, although glancing over the games a few things stood out;
•White still castles 0-0-0 in the majority of games. If so black just doesnt castle and keeps his King in the centre.
•If white does 0-0 then black often feels happy to castle Kside too
•Blacks Qknight belongs on d7. It almost always goes there on move 9 or 10
•Nd5 is almost immediately answered by ...Bxd5 in almost every case. Nd5 is always combined with a 0-0-0 plan by white.
•black Kbishop often ends on e7, but another idea is g7 & Bh6 to exchange, or even Bg7 in the event of a Qd2
•Qc7&Rc8 look like a pretty standard idea

Once again, this is all very superficial with no knowledge of the specifics!!! Own risk!!!
Sorry, in the second to last point replace 'g7' with 'g6'.
I've been playing this line for black almost exclusively with pretty decent results overall. Now, there are some specific move orders you have to be familiar with if white castles long. First, if white goes h3 it is often recommended you go h4 to freeze the structure. Normally if white expands on the kingside they go g3-h3-g4. I recommend postponing castling for a while, here you have an example game showing how quickly it can go wrong for white:
http://pl.lichess.org/7hTHov21#0

If white puts a knight on d5 you don't always take back with the bishop, sometimes you take with a knight on b6. Very often white goes for some positional plans with Bg5 eliminating one of the knights and then trying to occupy d5. If that happens always think about activating your e7 bishop through d8-b6 in the endgame or the middlegame. Also always be aware white can switch plans and go h3-f4 challenging your center. Here is another interesting game showing more/less how to play witha closed center and a bishop pair:
http://pl.lichess.org/4ZjPDmq56opG

Both games obviously have mistakes, sometimes even significant, but they might help you a bit. I heavily recommend this line, it is positinally rich, also similar structures can arise in different Najdorf variations. Classical games by Topalov are probably your best bet on getting a good feeling for the position.

GL with the line, I hope it sticks with you for a longer time :)
Thanks to you both. I appreciate all the research you put into learning the ideas for black and hope I can use this variation for a long time. :)
Great game! Looks like you came out of the opening at least equal with an interesting middlegame ahead. Doesn't get much better than that. I'd say stick with it mate.

This topic has been archived and can no longer be replied to.