and worse - there is 16.Re1!! fg5 17.Kg1! when black is just lost. White's attack is too strong.
Not published in NIC, but published in this forum.

Matemax in the Yearbook:  25. Qe7 Qe6 26. Qc7 Kf6 27. Qf4 Rg8 28. g4 with a decisive advantage.

The verdict of the riga now really hangs on Matemax' brilliant analysis. It goes very deep and one may wonder how relevant it is for a practical game ... but then it's not nice to play a variation that is busted.
The endgame is not much for white and in the long term he may be in serious risk of losing. My results with it have been excellent. Sufficient replies against white's major tries here have been given in YB85. Same for 8.c4 which isn't much after 8. ... dc3 9.Nc3 Bb4 and black is OK. So it has to be 8.Bg5 - a move hardly ever played! (I never had it in a game)  
The whole analysis of Matemax is pretty forced so all depends on the endgame after 8.Bg5 f6 9.Nd4 Bc5 10.Nc6 Bf2 11.Kf1 Qd7 12.Nc3 bc6 13.Ne4 Be1 14.Nf6 gf6 15.Qh5 Kf8 16.Re1!! fg5 17.Qh6 Kf7 18.Bc6 Qd6 19.Qh5 Qg6 20.Bd5 Be6 21.Qf3 Qf6 22.Qf6 Kf6 23.Re6 Kf5 24.g4 Kg4 25.Ba8 Ra8 26.Rh6.
If any white would ever get this far, do we really believe this rook endgame would be lost for black?! I don't. Black's king is very well placed (strongly supporing any future K-size past pawn[s]) he only needs to activate his rook ... so there goes ...
26. ... Rf8+ 29.Ke2 (29.Kg2 Rd8=) Re8+ 30.Kd3 Re7 31.a4 Rf7! threatening Rf3-h3 with an about equal position. (32.Ra6 h5 or 32. ... Rf3+-h3 both about equal)
The ending is admittely complex but there are loads of rook endings worse than this one which are drawn. Which places the riga right back in the set of playable lines!   
 

Matemax wrote on 05/08/08 at 06:38:11:


So, it looks like the Spanish game is done for, then.  Dear me, what to do?

CheckMate wrote on 05/07/08 at 15:55:55:


White is simply better in the endgame with Tarrasch's 15.c3, and probably also in other endgames. It's not such difficult chess; at best Black can grovel for a long time and draw. It's hard for me to understand why anyone would want to play into that.  In part I think many here simply overestimate the rook and pawns versus the two minor pieces.

I mean, think about it.  If the Riga were adequate, the Spanish would be a big bust.  How likely is that?

JEH wrote on 04/26/08 at 15:12:17:


I suggest to have a look at 13.Re4 de4 and now either 14.Qd7 or 14.Bf4

I think the Riga Variation isn´t playable, because of 8.Bg5! :

1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Nxe4 6. d4 exd4 7. Re1 d5 8. Bg5! f6 (Be7!? 89.Bxe7 Qxe7 10.c4! ) 9.Nxd4 Bc5 (Bd7 10.Nc3!) 10.Nxc6 Bxf2 11.Kf1! Qd7 12.Nc3! bxc6 13.Nxe4 Bxe1 14.Nxf6+ gxf6 15.Qh5 Kf8 16.Re1! fxg5 (Qf5+ 17.Kg1 Qg6 18.Qe2) 17.Qh6+ Kf7 18.Bxc6 Qd6 (Qf5+ 19.Kg1 Be6 20.Rf1) 19.Qh5+ Qg6 20.Bxd5+ Be6 21.Qf3+ Qf6 22.Qxf6+ Kxf6 23.Rxe6+ Kf5 24.g4+ Kxg4 25.Bxa8 Rxa8 26.Rh6 
This endgame should be winning for White. For example:
26...Rf8+ 27.Ke2 Rf7 28.Rxa6 h5 29.Ra3 Re7+ 30.Kf2 Rf7+ 31.Kg2 +-   

I analyse the rook endgame a bit and White should win without any problems with the queenside passed pawn

So I wouldn´t play the Riga Varíation, because of Bg5! and I think the main variation is bettert for White, too, because of the two peaces.

@ Anonymous:

Mr Khalifman does analyze the Riga endgame. It's interesting that he almost considers it playable for Black with the best lines leading to a small advantage for White (doesn't he come to the same conclusion about every line in the Ruy Lopez?).


@ TopNotch:

While I agree with you in principle I feel some sympathy towards our anonymous "bad guy". Such small details as how to face the Riga variation and other offbeat lines may influence my decision to buy the book in the first place. Let's say that I would buy the book if he treats 8. Bg5 against the Riga but not if he only treats the endgame variation.
Personally I never buy a book without knowing exactly what it contains. I invariably looks through  the book at the book shop or discuss it with a friend who also have it before I buy anything. Such an attitude is essential these days with so much trash out on the market. 

A related question is: If I would to review a book how much of the content is I allowed to expose without infringing copyright law? I think the reader deserves to know some about what theoretical lines are covered. For instance if an important line is missing it's my responsibility as a reviewer to report this fact. The published list of content is usually too coarse grained to be really useful. Good book reviews are unfortunately very rare.