Two stand-out reasons:
(1) Flexibility: covers both c4 and f3
(2) Avoids the drawbacks of Nc3, which blocks the c-pawn.

I did not look at the Na3 line because it doesn't seem to accomplish much for white. It takes the N away from the action and allows black to take it with the d7 Bishop pulling our QR to a less desirable square and eliminating a potential b6 occupier.

12...o-o 13. Nc3?! Bd7 14.Bg5?! allows 14...Rc8 with control of the 1/2 open c-file or 14...Nfg4 15. Bxe7 Qxe7 16.Any reasonable move 16...Nxd3 and black looks better (coordinated rooks, c-file control and after pawn to e5 our knights have no scope at all.

i.e. 12.... 0-0 13. N1d2 Nxd3 and now

A) 14. Qxd3 e5 15. Nc4 Be6 16. Nb6 Rb8
This is a fairly typical Sicilian
position. White has some pressure,
and even a space advantage. However, black owns the long term trumps: a central pawn majority and the two bishops. I don't like White's prospects here despite black's shortterm discomfort.

B) 14. cxd3 e5 15. Nc4 Be6 16. Nb6 Rb8

I might be one of them. I hope f4 gets a forum.

Now that we are emerging from the opening, I have unboxed Fritz 8, Shredder 8, Hiarcs 9, Junior 8, and Chess Tiger 15. I let each engine analyze a position. Then I give the most positive engine the White pieces, and the most negative engine the Black pieces, and I let them play each other. Inevitably, one of them proves the other wrong.

Using this method, negative Shredder shattered positive Hiarcs optimism regarding 13.Bb6, which seems drawish. Cynical Shredder again deflated Hiarcs about 13.Nc3, which is tame. I have not had time to look at 13.N1d2 nor 13.Rd1.

Fritz 8 seems to consistently over-value White's position, here and for a few moves ahead. I think most of its evaluations are maybe half a pawn too high, eg it gives us +0.8 when +0.3 would seem more reasonable.

I assume this is simply the evaluation algorithms missing the subtle points a human can see. The engine sees white has more space, a solid castled position, no obvious weak points. Black gets points deducted for his hole-y kingside, his apparent weak pawn at d6, his (currently) uncastled state, and his overall cramp.

That's Fritz. But has <any> engine been configured to play (or even evaluate the strength of) a Hedgehog-type setup? I doubt it. Maybe, as you are apparently doing, adjusting the optimism parameters etc will help some...

What fun!

How do the engines solve the problem of 14. Qxd3 versus 14.cxd3?

Do they agree? And what length of time are we talking about?

Thorsson: <RookFile: How are we going to win this game? ...you can more or less plan on GM Nickel to play these moves, in this order: 12...0-0 13...Nxd3 14.... e5 15..... Be6 transposing black's defense into the time honored Najdorf setup. In researching the available games played at the GM level, I believe that the following game is a model example of a line of play the world wants to follow:

M Golubev vs Epishin, 2003>

I said something similar before, except I prefer GM Lane's d4-d5 push, again as long as we avoid f4, which ruins our pawns.

I believe, however, that Arno will deviate from Epishin's play here and go 13...Bd7. He will want to tempt us into playing f4 before chopping on d3.

It's a long shot, because in this game the Black knight didn't reach e5 until move 17, and went via c6 rather than d7.

RookFile argues that Black can head for a Najdorf-type position with moves like ...O-O, ...Nxd3, ...e5, ...Be6. By this logic, Nc3 seems stronger, covering d5 and e4.

Thorsson points out that ...Bd7 by Black would be more subtle. It has some queenside threats (c6, b5, a4), develops the Bc8 to allow ...Rc8, and tries to tempt us into a premature f4.

I think it's important to note that, even if AN plays RookFile's moves, we're still not in a 'real' Najdorf. There are similarities, but also vital differences.

One of these is that AN's Kan system gives us the luxury of choosing the best path for our Nb1. In all mainline open Sicilians it has to go to Nc3 early. In many positional-type Najdorf-esque games White later goes Nd1 or even Nc3-b1-d2, trying to get the knight onto better squares. We can get to the better squares in one shot by playing N1d2.

I haven't seen anything to persuade me that Nc3 is better, but I agree we'll probably have a fight on our hands persuading its longtime supporters.

<if he does play 13...Bd7, what are we going to do?>

Which sequence is that after 12.Be3?

Team WHITE has been all about smart moves and originality...

It appears that 14. Bb6 is a reliable reply, first of all forcing the queen onto an uncomfortable square. Black probably replies 14....Qe8. Now 15. f4 Nxd3 16. cxd3

And his domain (where you'll also find game lines with playable boards, and many other goodies): http://www.nickel.nohyphen.com/

Note: a couple of additions below in [square brackets] are mine - points I was unclear about. If I'm wrong, [apologies].

Monad:
Rightly or wrongly, I have noted the following about our position:

We have made some progress on the queenside, but we're doing nothing active on the kingside.

We've lost the chance to get control of the -d5-square and little prospect of winning it back. Playing through the game it is striking to see how Black homes in on control/possession of -d5-. [should this be e5 - ??]

The a5 pawn needs defending by pieces; at present both the Queen's Rook and the King's Knight are there to protect it.

Presumably the idea behind pushing the a-pawn up that far was to engage in a skirmish with the black b-pawn. If that was indeed the case, then we must leave b6 free for that pawn to travel over. This means no Bb6.

If we play Nc3, then we have both our b-pawn AND our c-pawn blocked.

Black's d-pawn may be a little weak, but the a- and g-pawns are enviable.

Are we after a material advantage at this stage? If so, what? Given the black pawn structure, we might attempt to eliminate Black's DSB, but I don't see how we could achieve that, now that the move 12.Bg5 has been nixed.

Maybe our first objective ought to be attacking the pawn on -d6-.

All in all I can't help feeling that Black has not only freed himself already, but has a fine position which is ready to attack as well as defend. I see no evidence of a 'bind' that we are supposed to have applied and, quite frankly, I am getting a little scared.

Junior's outlook at 20 ply differs from Rybka although the values between Junior lines are equally close. The sequence of expected replies is rather unhelpful.

As far as 12....Nxd3 is concerned, I suppose it is not unreasonable for Black to eliminate our LSB, seeing his present pawn structure, even if that means his leaving the d5 [again: e5 - ??] square. In addition it presents us with a difficult choice in how to take it: Queen or pawn. That decision could have far reaching consequences. I noticed Junior and Rybka have opposing ideas on this. [Monad ends]

'Course it should be.

Just testing if you were paying attention.

But after 12...O-O 13.Nc3/d2 Bd7, I'm not so sure. If 14.Bb6 Qe8 is in some ways a good square for the queen. Our bishop has gone queenside, abandoning the threat of Bh6 that otherwise stops ...Qe8. And from e8 the black queen powers up the Bd7 - possible threats on c6/b5/a4 - while also providing some force down the e-file.

Thorsson: I've found another CC game (not on the Ultracorr CD): [Event "BdF MS-014"]
[Site "www.remoteschach.de"]
[Date "2004.05.26"]
[Round "?"]
[White "Bekemann, Uwe"]
[Black "Riemer, Wolfgang"]
[Result "0-1"]
[ECO "B42"]
[WhiteElo "2125"]
[BlackElo "2265"]
[PlyCount "60"]
[EventDate "2004.??.??"]

We now move on to analyse our new topic, 13.N1d2

**MARKER DIAGRAM**

This is the position if we play 13.N1d2

click for larger view

A lot of debate this time will focus on where the Knight should go. Leading contenders are likely to be 13.N1d2, 13.Nc3, 13.f4 and 13.Bb6.

The ideas behind 13.N1d2 have been pointed out a little way back, so I won't repeat them just now. Basically, it seems a more flexible and subtler way of bringing out the knight, with future options of kingside and queenside, eg Nc4 and Nf3.

to use as a starting point.

<line 5 = (0.37): 13.N1d2 Bd7 14.f4 Nxd3 15.cxd3 d5 16.e5 >

The effect of 15.cxd3 versus 15.Qxd3 will have to come into it soon.

What does this mean? It means that when we play 13. Nc3, and he replies ...Nxd3, we can reply 14. cxd3.

Remember the criticism of 3. d4? it was that we were trading a central pawn for a wing pawn, and black had a central pawn majority. With 14. cxd3, we rectify that problem and pave the way for a CENTRAL pawn advance - d4! - of our own, since the c3 Knight so nicely controls e4.