I think 13...b5 is probably OK as well against a Kb1 setup, though it's easy to drift into a passive position in the queenless middlegame if you're not careful. The main worry I have about these positions is that it's not just the Kb1 lines that are suffering for Black: the straight-out hack with h4 and g4 is pretty brutal too. It's covered in "Play the Sicilian Dragon", which improves on some analysis in WWTD2 - I can post the line if you like.

Adam

and the main problem is
if you dont play 15...Qc7, you're not entering the variations that Chris Ward has prepared, with a great deal of effort, to handle the yougoslav attack (after 16.Bd4 Be6 17.h5 a5 18.a4 b5!? etc)

so the variation is closed to be dead

i'm now working very hard in topalov variation (interesting but very deep and not that easy to play)
and on the chinese variation (so exciting new line !)

If i can say something more on this question : "What exactly is wrong with the Qa5 lines?"
I just want to answer that i felt on this forum just because i was in trouble with my Qa5 pet lines

In the main line that Chris Ward gives in his book, you can find :
1.e4 c5
2.Nf3 d6
3.d4 cxd4
4.Nxd4 Nf6
5.Nc3 g6
6.Be3 Bg7
7.f3 Nc6
8.Qd2 00
9.Bc4 Bd7
10.000 Qa5
11.Bb3 Rfc8
12.Kb1 Ne5
13.h4 Nc4
14.Bxc4 Rxc4
15.Nb3 Qc7
and now
as an alternative to the main 16.Bd4 on which Chris found some really great moves, there is 16.g4!? with a very sensible position for black
In WWTD2, CW just gives 16...Rc8 17.h5? Bxg4! with a winning attack
But if i'm right, white has a simple, almost winning plan :
16...Rc8 17.e5! (used by Karjakin already, and on my database there is 81% of win for white !)

Here 17...Ne8 18.Nd5 Qd8 19.exd6 leads to nightmare :
a) 19...Nxd6 20.c3 +/-
b) 19...Rxc2 don't change anything because of 20.dxe7! (stronger than 20.Nxe7+ as played in some games) 20...Rxd2 21.Rxd2 Bf5+ 22.gxf5 Qd6 23.Bf4 +/-
c) 19...exd6 isn't sufficient to hold the balance either, because now that black pawn structure has been weakened, white can simply play 20.c3 to ensure the position of his king and begin to harass black's position.

Then i looked at early deviations and nothing was good enough to be played in serious games : a good part of my energy passed in the search for a new 16th move (instead of 16...Rc8) but the only serious one i found was 16...Be6 (removing the bishop of the storm)
I calculated some crazy lines beginning by :
16.g4 Be6 17.h5 Bxg4!? (woot)
18.fxg4 Rxc3 (woooooot)
19.Qxc3 (the only move :p) 19...Qxc3 20.bxc3 Nxe4
(this gives black a strange position, prolly quite loosing, but counting on some tactics with the h6 square available for the bishop, all my lines were "almost" playable

Unfortunately, even this "bad but personal" idea can't be even dream't of being seen in practice, because White can play a rather stronger move than 17.h5...
17.Nd4!
and everything is clear

This is the main reason why i just give up the Qa5 line and if someone knows how to improve somewhere, please, help !

(now i focus my attention on the Chinese variation)

By Qa5 lines just for those who don't know I am referring to the Yugoslav attack 1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 g6 6.Be3 Bg7 7.f3 0-0 8.Qd2 Nc6 9.Bc4 Bd7 10.0-0-0 Qa5 (or 10..Ne5 11.Bb3 Qa5), planning Rfc8