(7.a3 exf3) Yes, I thought the misplaced bishop (on b3) was a key factor that I thought favored black.  This is one reason I think the pawn on c6 is helpful for black, since it covers d5.  I think the bishop is a bit stranded on b3, since ideas with d5 seem to be cut out and I think black can easily defend against pressure against e6 and f7.  For one, black might aim for ...Nd5 at some point, for instance black might be able to develop, Nd7-b6-d5.  Of course, it needs to be examined since black needs to watch out for sacrifices on e6, but I don't think black has many problems if he is careful.  Also ...b5 serves a purpose in cementing a knight on d5, as c4 advances are covered.  If black can successfully post the knight on d5, I think he is clearly better.

Has anyone looked further at the continuation of the line 7.Bg5 Be7 8.fxe4 Nxe4 that I analyzed (in particular 10...Kxe7)?  I think black has a clear advantage here.  Anyone notice an improvement for white?

At the moment, it looks like 5...b5!? would be my favorite choice for OTB play, especially if my analysis after 7.Bg5 is correct.  To me, it seems most in the spirit of the counterattacking nature of the O'Kelly Defense, in that it seems to put the most pressure on white at an early stage.

In my opinion, it looks like white has more practical chances after 5...Bf5.  I'll have to look at Patrik's analysis of this when I have more time.

Hello. Short reply to some of the questions.

BDG - O'Kelly Defense [D00] 
 
1.d4 d5 2.e4 dxe4 3.Nc3 Nf6 4.f3 c6 5.Bc4 Bf5 6.fxe4 Nxe4 7.Nf3 Nxc3 8.bxc3 e6 

X wrote: „Is given by Kaufman where he comments "Its hard to see much compensation for the pawn."  I tend to agree.  Black looks very solid here to me.  Maybe white has an improvement earlier.  Perhaps this is another starting point for discussion in this line.“

My reply: Here, Kaufman is perhaps wrong. I consider 7..Nxc3?! to be a dubious move already as my analyse could show:

7...Nxc3?! c2,c3,d4 pawn centre seems to be for white´s good. It strenghtens d4, open b-file for rook and sometimes also Ba3 comes into consideration. And of course, black knight on e4 should be rather coming back to defence retreating to f6 or d6. 8.bxc3 e6 9.0–0 Bd6 10.Ng5 Bg6 11.Qg4! Qe7 (11...Qc7 12.Bxe6! fxe6 13.Nxe6+-; 11...0–0 12.Nxe6!) 12.Nxe6 fxe6 13.Bg5© with dangerous initiative for a piece.


X wrote: „5...Nbd7!? Lastly, this might be a good unambitious attempt to frustrate BDG players.  Does white have anything here?“ 

My reply: Yes, this move is interesting, indeed. I have no analysis in hand considering this line as it was never played against me. 

X wrote: 1.d4 d5 2.e4 dxe4 3.Nc3 Nf6 4.f3 c6 5.Bc4 Bf5 6.fxe4 Nxe4 7.Nxe4 Bxe4 8.Nf3 e6 9.0-0 Be7  In the analysis of 5...Bf5, I think I have found an improvement for black in the 9...Be7 line following 7.Nxe4“

My reply: Yes. It is an improvement. Comparing with 9..Nbd7?! line. 1.e4 c6 2.d4 d5 3.Nc3 dxe4 4.Bc4 Nf6 5.f3 Bf5 6.fxe4 Nxe4 7.Nf3 e6 8.0–0 Nd7?! 9.Nxe4 Bxe4 10.Ng5! Bg6 [10...Nf6 11.Rxf6! Bxc2 12.Rxe6+ fxe6 13.Qxc2 Qxd4+ 14.Kh1±] 11.Bxe6! [11.h4 Be7 (11...Nf6!?) 12.Bxe6 fxe6 (12...0–0!=) 13.Nxe6 Qb6 14.h5‚ with attack.; 11.c3 h6 12.Nxe6 fxe6 13.Bxe6 Nf6 14.Bf4÷ unclear] 11...fxe6 [11...Nf6! 12.Bc4˛ And white has only a slight advantage because of development advantage or the position is maybe = already.] 12.Nxe6± Line

I haven´t find an advantage yet in your 8..Be7! line.

8...Be7! 9.Nxe4 Bxe4 10.c3 0–0 

a) 10...Nd7 11.Qe1 Bxf3 (11..Bg6!? M.Nb) 12.Rxf3 0–0 13.Bd3 Nf6? 14.Qh4 h5 15.Bg5+-; 
b) 10...Bxf3 11.Qxf3±; 

11.Qe1 Bg6 

a) 11...Bxf3 12.Rxf3 Nd7 13.Bd3 Qa5 (13...Nf6 see 10..Nd7) 14.Bh6!; 
b) 11...Bf5!?; 

12.Ne5 Nd7 

(12...Bf5!? 13.Qg3 

(13.g4!? Bg6 14.Nxg6 hxg6 15.Rxf7!? Rxf7 (15...Kxf7! 16.Qxe6+ Ke8 17.Bg5 (17.Qxg6+!? Kd7 18.Qxg7 Kc8 (18...b5 19.Bg5 bxc4 20.Re1 Re8 unclear 21.Qf7 Kc8 22.Bxe7 Qd7 23.g5!? Qg4+ 24.Kf2 (24.Kh1 Rh8 25.g6 Qg3 26.Re2 Qd3 27.Qe6+ Kb7 unclear) 24...Rh8 25.Bd6 Nd7 26.Qf3 Qxf3+ 27.Kxf3©) 19.Be6+ Nd7 (19...Kc7 20.Bg5 Re8 21.Rf1+-) 20.Bd2 Bf6! (20...Kc7 21.Re1 Bh4 22.Re4 Re8 23.Bf4+ Kb6 24.d5‚; 20...Bd6 21.Bg5 Qe8 22.Re1 Kc7 23.Kg2÷) ) 17...Qd6 18.Qc8+ (18.Re1 Qxe6 19.Rxe6 Rf6 20.Bxf6 gxf6–+) 18...Qd8=) 16.Qxe6 Qe8 (16...Qf8 17.Bd2©) 17.Bg5©) 

13...Nd7÷) 

13.Nxg6 hxg6 14.Bf4 c5 15.Bxe6 fxe6 16.Qxe6+ Kh7 unclear



M.Nb wrote: „7.Nf3 Nxc3 8.bxc3 Qa5! and Rajmund on his website has shown, that White's attack is not strong enough. At the other hand 8...e6 was played in Keeling-Okosieme, St Helier 1998. White could have established a clear advantage with 13Qe2 Be7 14Bf4 followed by 15Rae1. Therefore 6.g4 Bg6 7.g5 Nd5 8.Nxe4 is White's best, but Black surely can maintain equality.“

My reply: In my personal analysis I have following..

7...Qa5 8.Ne5! e6 (8...f6 9.Qh5+! g6 (9...Bg6 10.Bf7+ Bxf7 11.Qxf7+ Kd8 12.Nc4± with advantage) 10.Bf7+ Kd8 11.Nxg6± with advantage) 9.0–0 Nxc3 (9...Nd6 10.Qe2 with a good compensation for a pawn.(10.g4!? Bg6 needs to be analysed in detail.(10...Nxc4 11.Nxc4 Qb4 12.gxf5 Qxc4 13.fxe6‚ with attack.) ) ) 10.bxc3 Qxc3 (10...f6 11.Rxf5!) 11.Rxf5 Qxa1 12.Rxf7 Ba3 (Bc5! This move is on the ED website. Well, this might be as an improvement 13.Rf4 Rf8 14.Rxf8+ Bxf8 15.Bxe6 Bc5 16.Qh5+ g6 17.Bf7+ Kd8 18.Qg5+ Be7 19.Qe3 Nd7=/+) 13.Rf1 Bxc1 14.Qg4 Be3+ 15.Kh1 Qxd4 16.Qxg7+-] 


X, M.Nb prefers this line 1.d4 d5 2.e4 dxe4 3.Nc3 Nf6 4.f3 c6 5.Bc4 b5!? 6.Bb3 e6 7.a3 exf3! X wrote: „Oddly enough 7.a3 exf3 is not considered on Rajmund's website.  Maybe it's because of the switch in mode from counterattack to acceptance.  MNb's suggestion is very logical, since white has spent a move on the non-developing a3.  So essentially, it's an Euwe defense where black has gotten in the moves ...c6 and ...b5 for free.  Since the bishop is developed to c4, the c6 pawn has a useful function in covering d5, making the bishop look ineffective at its post.  Since the Euwe is solid in its own right, I think this postion should be clearly better for black.“

My reply: Well, after 8.Nxf3 white has developed 3 pieces, black one. It is very difficult to compare this position to Euwe. You think that moves ..c6 and ..b5 are here for free. Well, in Euwe, the best setup includes Be7 and Nc6, playing against d4. Here, in my opinion, c6-b5 weakens black´s position and cannot be recommended for real, strategically based chess. White has good compensation for a pawn here, yet the position would need a deep analysis because at first glance, black seems to have an edge (at least according to computer´s engines). But anyway, 7.a3 looks a bit odd. Maybe the idea is to open an a-file, because it is very probably that black will be playing on queenside (with a5 etc.). E.g. in b4 variant it is obvious. Still I wonder if there is some other move in this position, because a3 can be played e.g. only after a5 is played. If you reject fxe4 (Short _ Bareev), we still have some moves in hand. I guess M.Nb recommendation with Bg5 will suit our case. It is a developing move unlike a3. I hope someone will come up with a whole new idea for white. 

Here is a summary.  Looking closer, after 10.Qg4, I think 10...Kxe7! is strong for black, while 10.Qxe7 is not so clear:

1.d4 d5 2.e4 dxe4 3.Nc3 Nf6 4.f3 c6 5.Bc4 b5!? 6.Bb3 e6 7.Bg5 [7.a3 exf3! MNb] 7...Be7 8.fxe4 Nxe4 9.Bxe7 Nxc3 10.Qg4!? 
 
[10.Bxd8 Nxd1 11.Rxd1 (11.Bc7 Nxb2µ) 11...Kxd8 12.Nf3 Nd7 13.0-0 h6µ] 

10...Kxe7! 

[10...Qxe7!? 11.Qxg7 Rf8 12.bxc3 Qh4+ 13.Qg3 Qe4+ is not so clear.(13...Qxg3+!? 14.hxg3 Rh8 unclear) ] 

11.bxc3 [11.Qg5+ Kd7! 12.Qxd8+ Kxd8 13.bxc3µ] 11...Kf8µ

I think black has a clear advantage here, although at first his king may not look so safe.  With ...Nd7-f6 coming, I think black has a solid defense, especially since the dark-squared bishop has been exchanged and there are fewer pieces on the board.

7.Bg5 is not mine, as it was already played in Van Schaardenburg-Boersma, Groningen 1989. After Be7 8.fxe4 Nxe4 9.Bxe7 Nxc3 10.Bxd8 Nxd1 11.Rxd1 Kxd8 12.Nf3 it is a bit too early to give Black an advantage, as Black's position is quite cramped. But White should play 10.Qg4! of course, which looks at least equal after Qxe7 11.Qxg7 Qf8 12.Qxf8+ Rxf8 13.bxc3.

At the other hand 7.a3 loses a tempo, so Black can play exf3! in a favourable version: Ohtake-Mercadal Benejam, em 2000. This is more solid than 7...b4 (why?) 8.axb4 Bxb4 9.fxe4 Nxe4 10.Qf3 Nxc3 11.bxc3 Be7 12.Nh3 o-o 13.o-o Ba6 14.Rf2 and White is ready to attack.

BTW I have noticed, that Rajmund likes to lose a tempo for development (like 7.a3 above), being a pawn down. Is this a consequence of his ED-theory?

1.d4 d5 2.e4 dxe4 3.Nc3 Nf6 4.f3 c6 5.Bc4! b5 is not a problem for white. Rajmunde Emanuel has very good and convincing analysis on this particular line at Caro Gambit section here:
http://www.mujweb.cz/www/rajmunde/ec_ena_car.htm
They contain e.g. improvement over Short vs Bareev game, which was cited here. Feel free to check the analysis, in my opinion they are very good.

Yes, I think the most critical line in O´Kelly (or Millner-Barry) is 5.Bc4 Bf5!? Positions resembling Caro-Kann arise and it is quite difficult to find any advantage. I will have a look at my personal analysis I did some year ago and I hope I will be able to send them here soon.

And yes, 5.Bc4 e5?! is also dealt with somewhere on ED web page, again, in a very convincing way.

In the analysis of 5...Bf5, I think I have found an improvement for black in the 9...Be7 line following 7.Nxe4:

1.d4 d5 2.e4 dxe4 3.Nc3 Nf6 4.f3 c6 5.Bc4 Bf5 6.fxe4 Nxe4 7.Nxe4 

7...Bxe4 8.Nf3 e6 9.0-0 Be7

I think this move deserves further consideration based on the following idea of rapid castling. 

10.c3 0-0! 

This move is my improvement for black.  I think it is logical for black to the castle as quicly as possible, while at the same time preventing methods of attack by Ng5.  Especially, considering that in many lines where Nd7-f6 is played immediately, the king is awkwardly stuck in the center due to ideas with Ng5 and/or Bxe6. 

11.Ne5

[11.Bd3 Bg6µ looks solid for black.] 

11...Bg6 12.Qe2 Bf6!µ 

Another key idea.  Sacrifices on e6 fail after this move.  Otherwise black has smooth development with Nd7-b6, perhaps with a preliminary Bxe5, if necessary. 

13.Nxg6 [13.Bxe6? Bxe5-+] 13...hxg6 14.Bxe6? Re8-+

Here are some notes I made.  I think 5...Bf5 is worthy of attention.  Also I think 5...b5 and 5...Nbd7 should also be considered.

BDG - O'Kelly Defense [D00]

1.d4 d5 2.e4 dxe4 3.Nc3 Nf6 4.f3 c6 5.Bc4 

[5.Bg5 likely enters Veresov lines.] 

5...exf3 

Maybe this is the best move, but I would like to consider other moves that give the O'Kelly Defense independent play.  Perhaps this is the reason the O'Kelly Defense has had little attention in the forum.  Perhaps many of the BDG players only consider it is as a  variation primarily suitable for transpostion to the Ziegler Defense, while viewing independent deviations as inferior.  Still, I am of the opinion that there are other ways besides 5...exf3 that are perhaps safer in a practical sense for black to play for a slight edge (or at the very least, dynamic equality).  However, it does not suffice to give an exclamation mark to a black move and say white cannot achieve an advantage (as Schiller does).  Bias from the white players is not the only side to blame...
There are a couple of extra moves mentioned by TopNotch that I think should be considered.  Also I would like to examine 5...Bf5 further, in order to draw a comparison with the Ziegler Defense (especially the Alchemy variation as it seems critical).  As for Patrik Schoupal's lines (A) and (D), I tend to agree that these are not that worthy of attention.

[
-- 5...Qa5? (Schiller's move)  I have to be harsh here; I think this may allow to white demonstrate a significant advantage. 6.Bd2 Qb6 (6...Qf5 7.g4 Qg6 8.g5) 7.Qe2 Qxb2 8.Rb1 exf3 9.Nxf3 Qa3 I have to agree with you here; this looks flat out bad for black.  Rather convenient for Schiller to stop analysis here and say black has two pawns in a line he recommends for black!; 

--5...e5?! is another move, but I don't think it's that great.  White probably has a slight advantage.; 

--5...Bf5 is recommended by Larry Kaufman in his repertoire book.  I think this move especially deserves consideration, as it may be a safer way for black to play for an advantage than 5...exf3. 

A) 6.g4 Bg6 7.g5 b5 (7...Nd5 8.fxe4 Nxc3 9.bxc3 e5) 8.Bb3 b4 9.gxf6 bxc3 10.fxg7 Bxg7 11.bxc3 e5µ Abbasifar-Filippov, Golden Sands 2000;

B) 6.fxe4 6...Nxe4 7.Nf3 (7.Qf3 Nd6 8.Bxf7+ Kxf7 9.g4 e6 also looks good for black.) 7...Nxc3 8.bxc3 e6 is given by Kaufman where he comments "Its hard to see much compensation for the pawn."  I tend to agree.  Black looks very solid here to me.  Maybe white has an improvement earlier.  Perhaps this is another starting point for discussion in this line.; 

--5...b5!? 

I think this should be considered, especially since it was played by Bareev in a rare instance of high level tournament play in this variation.  The following is the sample game provided by TopNotch played between Short and Bareev.  Perhaps white's play can be improved, but this seems like a good attempt at steering the game out of traditional BDG territory, while allowing black good chances to outplay white.  Perhaps a new starting point for discussion?  I am not looking for a direct tactical refutation, but am rather happy with a postional edge for black that allows good chances to play for a win.  Maybe there is a viable way for white to mix things up? 

6.Bb3 e6 7.fxe4 b4 8.Nce2 Nxe4 9.Nf3 Ba6 10.0-0 Bd6 11.c4 bxc3 12.bxc3 Nd7 13.Qc2 Nef6 14.c4 0-0 15.c5 Bc7 16.Bg5 h6 17.Bh4 Qc8 18.Rfe1 Bxe2 19.Rxe2 Nd5 20.Rf1 Qa6 21.Re4 Rae8 22.Rfe1 N7f6 23.Bxf6 Nxf6 24.Rh4 Qa5 25.Re2 Re7 26.g3 Rb8 27.Kg2 Rbe8 28.Qd3 Nd5 29.Ne5 Qc3 30.Qxc3 Nxc3 31.Rd2 Bxe5 32.dxe5 Nd5 33.Ra4 Rb8 34.Ra5 Kf8 35.Kf3 Reb7 36.h4 Ke7 37.Rd4 f6 38.exf6+ gxf6 39.Rda4 Nc3 40.Ra3 Nb5 41.R3a4 Rd7 42.Bc4 Nc3 43.Ra3 Nb1 44.R3a4 Nd2+ 45.Ke3 Rg8 46.Be2 Rxg3+ 47.Kf2 Rc3 48.Rxa7 Ne4+ 49.Kg2 Nxc5 50.Rxd7+ Kxd7 51.Ra7+ Kd6 52.a4 Rc2 53.Kf1 Ke5 54.a5 Kf4 55.Rg7 Ne4 56.Bd3 Ra2 57.a6 Nd2+ 58.Kf2 Nf3+ 59.Be2 Nd4 60.Rg4+ Ke5 Short-Bareev, Sarajevo 2000; 

--5...Nbd7 

Lastly, this might be a good unambitious attempt to frustrate BDG players.  Does white have anything here?  TopNotch give the following lines:  
6.Nxe4 (6.fxe4 e5 7.Nf3) 6...e6 7.Ne2 Be7 8.0-0 Qb6 9.Kh1 0-0 10.c3 c5 11.dxc5 Nxc5 12.Nxf6+ Bxf6 
This seems like a no-nonsense way to obtain a game still full of play with at least equality for black.] 



6.Nxf3 Bf5 transposes to the Ziegler Defense, where I suppose black must be prepared to face the Alchemy variation.