Rebuilding the Repertoire

Jun 17, 2008, 1:52 PM |

The process of rebuilding my opening repertoire begun conceptually several years ago.  Although I have been generally dormant in USCF tournaments over the past several years, with my membership lapsing on more than one occasion, I have always kept acquiring opening books whenever I found one that I thought might be of later interest.  Consequently, my library has grown considerably over the years.

In rebuilding from the ground up, there was a key question that I had to honestly answer of myself before determing which path(s) I wanted to take.  And that was:

What did I expect from my opening repertoire?

In other words, what was the goal I was attempting to have an "improved" repertoire achieve?

And the answer took three forms:

  1. To consistently yield better positions against lower-rated players
  2. To consistently yield at least equal positions against players of my own strength, and
  3. To consistently yield positions that give me a fighting chance (as well as drawing opportunities) against players stronger than I

If my opening repertoire is accomplishing those three goals, then I believe it's doing its job.  My current trappy repertoire is delivering on (1) in spades, hit-and-miss on (2), and not doing so well on (3).  The other issue with (3) is that even when I'm fortunate enough to get a good position, I find the stronger players routinely outplaying me because of a better command of the strategic elements called for in the position.  In short, my analysis may be as good as theirs, but their strategic judgement was head-and-shoulders above mine.

Before I can figure out where I'm going, let's take a look at where I'm starting.

My repertoire, as of the beginning of 2008 was:


  1. 1. d4 with an eye to the Blackmar-Diemer Gambit (1. d4 d5 2. e4 de 3. Nc3 Nf6 4. f3).  2. Bg5 versus the Dutch (1. d4 f5 2. Bg5)
  2. On occasion 1. e4 looking for the Danish (1. e4 e5 2. d4 ed 3. c3 dc 4. Bc4) or Halloween Four Knights (1. e4 e5 2. Nf3 Nc6 3. Nc3 Nf6 4. Nxe5?!) Gambits.  Against French, Caro-Kann, Alekhine's and Center-Counter, transposing into BDG variants.  Austrian Attack (4. f4) against the Pirc.  Against Sicilian, Smith-Morra Gambit (1. e4 c5 2. d4 dc 3. c3) or...who the heck knows?

As for black, it was looking like:

  1.  Against 1. e4, 1...e5, with an eye to Ruy Lopez Schliemann Defense (1. e4 e5 2. Nf3 Nc6 3. Bb5 f5) or the Two Knights Defense (1. e4 e5 2. Nf3 Nc6 3. Bc4 Nf6).  Also various forms of the Philidor, including the Larsen lines (1. e4 e5 2. Nf3 d6 3. d4 ed 4. Nxd4 Nf6 5. Nc3 g6) and the Philidor Counter Gambit (1. e4 e5 2. Nf3 d6 3. d4 f5?!).  Against the Scotch, the wild 4...Qh4 line and against the King's Gambit, my own pet Falkbeer Counter-Gambit declined line (1. e4 e5 2. f4 d5 3. ed c6 4. dc Bc5?!)
  2. Against 1. d4, 1...Nf6, looking for a Budapest Defense (1. d4 Nf6 2. c4 e5?!); alternatively 1...d5, looking to play Bf5 against the Q's gambit (1. d4 d5 2. c4 Bf5?!) or the Von Hennig-Schara Gambit (1. d4 d5 2. c4 e6 3. Nc3 c5 4. cd cd?!)
  3. Against everything else, a Classical Dutch setup (1...f5, 2...e6, 3...Nf6, 4...Be7, 5...O-O)

On the whole, I can say this is a very fun repertoire, leading to exciting chess, reminiscent of the Romantic period.  Open lines, piece attacks, very Morphy-esque.  However, I can also say it wasn't an optimal repertoire, since goals (2) and (3) were clearly not met.

As best I can tell, the best way of assuring (2) and (3) get met is having a better strategic understanding of the opening (or variation) than my opponent.  That is, pick my battlegrounds more wisely, understand the strategic goals more deeply and appreciate better which plans yielded the best results (those most consistent with the strategic goals of the opening) and those that were full of crap.

Simply, I'd like to put more trust in my pieces, "listen" to where they "say" they want to go (is that intuition?), understand the pawn structure and its implications (chains, levers, blockades) and just be more familiar than my opponent on what the strategic objectives are in middlegames that arise from an opening of my choosing. 

As white, I like a repertoire with some variance.  As white, I like to be able to play 1. e4 or 1. d4 (or even 1. Nf3) indiscriminately, and my opponent not know what he's going to be confronted with.  At the same time, there is a practical need for some economy, as I don't intend to become a full-time chess player, or spend every waking moment of free time studying openings. 

Starting on the black side of the ball, against 1. d4, I'm going to anchor my repertoire with 1...d6, based on my reading of "An Explosive Chess Opening Repertoire For Black" (although it's altogether unclear how "explosive" and "1...d6" could be used to describe the same opening).  There are a couple of critical lines in this book, namely those that spring from:

  • 1. d4 d6 2. c4 e5 3. Nf3 e4 4. Ng5 f5
  • 1. d4 d6 2. Nf3 Bg4 3. e4 Nf6 4. Bd3 e6
  • 1. d4 d6 2. c4 e5 3. de de 4. Qd8+ Kd8

What I found utterly amazing was that, according to ChessBase, in the last line black scores an incredible 66%.  The authors extend a number of potential explanations for that anomaly, but exist it does.

The authors propose the Pirc against 1. e4, something I just can't bring myself to play.  There is a Philidor line consistent with black's repertoire choice (1. e4 d6 2. d4 Nf6 3. Nc3 e5) if white obliges by playing 4. Nf3.  However, white can also simply play 4. de de 5. Qd8 Kd8 with a queenless middlegame which, while great against higher-rated players, really hurts one's winning chances against lower-rated opposition. Since I already have the Philidor in my repertoire and know many of the lines, there is a certain attraction to just rolling up my sleeves and getting deeper into the strategic aspects of the opening.

Against 1. e4, I would love to keep playing 1...e5 having enjoyed the open positions which typically arise since playing chess as a junior.  That means something having a repertoire element against King's Gambit, Vienna Game, Italian Game (1. e4 e5 2. Nf3 Nc6 3. Bc4), Ruy Lopez (3. Bb5), Danish and Half-Danish Gambits, Scotch Game, Ponziani, Four Knights, just a whole a mess of stuff.  THAT's the problem with wanting to play 1...e5.  You have to find simple, effective lines that allow you to meet a wide range of white plans.

    Therein lies some of the attraction of keeping a basically sound (if somewhat cramped) line like the Philidor's Defense in one's repertoire, because it allows you to eliminate so many white options from consideration.  The Petroff's Defense (1. e4 e5 2. Nf3 Nf6) and Latvian Gambit (1. e4 e5 2. Nf3 f5?!) also provide the same benefit.  All that said, I still enjoy the struggles and conflict of ideas that arise from 1. e4 e5 2. Nf3 Nc6.

So, putting together some open defenses, let's start with the King's Gambit.  Although I like my pet decline line (1. e4 e5 2. f4 d5 3. ed [not 3. fe? Qh4+] c6 4. dc Bc5 5. cb Bxb7), Richard Francisco, a young Atlanta master, showed me a line in late 2006 that still intrigues me, namely (1. e4 e5 2. f4 Nc6 3. Nf3 f5!?).  Don't believe this has value?  Search for this position on Chessbase Online and see how many 0-1 results you find.  I assure you there are plenty.

For economy, the Ponziani (1. e4 e5 2. Nf3 Nc6 3. c3) and Danish/Half-Danish family can be met with the same reply, namely 3...Nf6 as put forth by Nigel Davies in his book "Play e4 e5!". 

The Vienna (1. e4 e5 2. Nc3) can be met with 2...Nf6 and there are good lines both for addressing white's two most common replies 3. Bc4 and 3. f4.  Bc4 is best met by 3...Nc6 while against 3. f4 Black can equalize comfortably using Karpov's suggestion of 3...Be7.  There is also nothing wrong with 3...Bc5 where black finds equality with 4. fe Nxe4 5. d4 Bb4.

So now, with a minimum of preparation, we arrive at the meat of the 1...e5 repertoire, a defense to the Italian Game (1. e4 e5 2. Nf3 Nc6 Bc4) and Spanish Game (Ruy Lopez) (3. Bb5). 

With respect to the Italian game, I'm looking forward to trying out Davies' two recommendations, which essentially diverge at white's fifth move.  After 1. e4 e5 2. Nf3 Nc6 3. Bc4 Nf6 4. d4 ed, white can try 5. O-O or 5. e5.  5. O-O is an invitation to steer into the Max Lange Attack with 5...Bc5, while 5. e5 invites  the modern Two Knights Defense treatment 5...d5 6. Bb5 Ne4.

Davies suggests an Anti-Max Lange line that seems to completely neutralize 5. O-O.  A well-known position from theory, representing what was heretofore best play by white, goes 5...Nxe4 6. Re1 d5 7. Bxd5 Qxd5 8. Nc3 Qh5 9. Nxe4 Be6 10. Bg5 Bd6 and now, best has been long held to be 11. Bf6 (D). 

Play follows 11...O-O 12. Nxd6 cd 13. Bxd4 with a slight advantage for white, oweing the weak black pawn on d6. 

However, Davies points out a new resource for black from the diagram in 11...Bxh2+! 12. Nxh2 Qxd1 13. Ra1xd1 gf 14. Nf6+ Kf8 15. Nf3 Rd8 16. Ng5 h6! 17. Nf3 (Nxe6+ fe 18. Rxe6? Kf7) Kg7 and black is just flat-out a pawn to the good.  This means white has to deviate into one of the "equality" lines earlier, like 10. Nxd4 -- although looking at the results of this position on Chessbase Online doesn't inspire confidence in the white position.

Against 5. e5, Davies recommends 5...Ne4 6. Bd5 Nc5!? as the route to equality.  The idea is the c5 knight settles in on e6, defending the d4 pawn, and ...Be7 (or ...Bc5) follows.  For his part, white should recapture the d-pawn as soon as possible, which affords black an equal game without much fuss.

Lastly, of course, is the linchpin of any 1...e5 repertoire, a defense to the Ruy Lopez.  My choice for this is also a Nigel Davies recommendation, not from "Play 1. e4 e5!" but rather his excellent book "Gambiteer II".  In it he recommends the Schliemann Defense (which I already played), but instead of the tantalizing complexities of 1. e4 e5 2. Nf3 Nc6 3. Bb5 f5 4. Nc3 fe 5. Ne4 d5 6. Ne5 de 7. Nxc6 Qg5!?, Davies rather espouses the modern approach favored by a number of GM's who currently feature the Schliemann in their repertoire, namely 5...Nf6.  This line involves a sacrifice of the e-pawn either by 6. Nxf6 Qf6 7. Qe2 or 6. Qe2 d5 7. Nxf6 gf 8. d4 Bg7 9. de O-O.  These lines are sharp but sound.

Davies also features an excellent treatment of white's most common practical reply against the Schliemann, 4. d3, where the first player attempts to eke out a small but real positional advantage.

Against stronger players, I will still maintain the Philidor's Defense in my repertoire.  The reason is that 1. d4 d6 2. e4 Nf6 3. Nc3 e5 leaves white with an invitation to go into the queenless middlegame mentioned above (4. de de 5. Qxd8 Kxd8) or to go into a standard Philidor's with 4. Nf3 Nbd7 or 4...ed.

I am quite fond of the excitement generated by 4....ed 5. Nxd4 g6 6. Be3 Bg2 7. f3 O-O 8. Qd2 Re8 9. O-O-O.  Compare this with the analogous Sicilian Dragon position below:









There are some critical distinctions (for instance, the half-open c-file for black against the half-open e-file) that make attempts to play one like the other problematic.  Most white players don't recognize the subtle differences and don't play against the Philidor version often enough to appreciate them. Another advantage of the Philidor version is the theory is pretty constant, while Sicilian Yugoslav theory can change by the day.  Suffice it to say neither opening is for the faint of heart, and the only thing clear is the game is going to end with someone being checkmated.   

So, enough about handling things for the black pieces.  Let's take a look at the white side of the board.

I like to play both 1. e4 and 1. d4 on occasion.  Sometimes, perhaps even 1. Nf3, transposing back later into my repertoire.  Thus, there is some work ahead.

On the 1. d4 side of things, the book I bought for this purpose is "Win With The London System" by Sverre Johnsen and Vlatko Kovacevic.  This opening seems to be a great balance of aggression and solidity and performed well in my ICC training games with it.  The London System features an early Bf4 by white.

When playing 1. e4, I'm cobbling together a repertoire from a few different sources.  White must obviously be prepared for a number of responses.

One advantage of playing the London is against the Caro-Kann, a 1.e4 player can adopt an approach that transposes from the London, namely 1. e4 c6 2. d4 d5 3. ed cd 4. Bf4.   My ICC training games with this variation have been quite successful from the white point of view.

Against the French Winawer (1. e4 e6 2. d4 d5 3. Nc3 Bb4), I am adopting the line recommended by Lev Alburt and Roman Dzindzichaschvili (lol, or however the hell you spell  his last name...could there be a more random collection of vowels and consonants?) in their book "Chess Openings for White Explained".  Their selection is 4. e5 c5 5. Bd2 !?.  The motivation for this move is both tactical and positional, and is not so popularly played, thus it fits in well with my repertoire and goals.

I also keep the Kupreichek line of the Advance variation in my head, namely 1. e4 e6 2. d4 d5 3. e5 c5 4. c3 Nc6 5. Be3 !? (instead of 5. Nf3).  The idea is that when black plays the inevitable ...Qb6, white responds with Qd2 protecting both d4 and b2.  Any exchange on d4 allows white to play Bxd4, presenting an opportunity to eliminate his "bad" bishop.

With respect to the Classical French (1. e4 e6 2. d4 d5 3. Nc3 Nf6) I haven't decided.  However, I don't like the notion of learning separate lines for each variation, so I may end up coming back to the advance or some generic French line over time.

In looking at the Pirc, I rather like Nigel Davies recommendation in "Gambiteer I".  The Austrian Attack with quite a twist at move 5, a3.  Thus, the line is 1. e4 d6 2. d4 Nf6 3. Nc3 g6 4. f4 Bg7 5. a3!? -- the idea is black can't play ...c5 now because of 6. dxc5, Qa5 7. b4! After the other obvious option 5...O-O, white plays 6. Nf3 c5 7. dc dc 8. Qd8 Rd8 9. e5 Nd5 10. Nxd5 Rxd5 11. Be3 b6 12. Bd3 Nc6 13. O-O-O (followed by Be4) with a better endgame for white.

Against the Sicilian, I enjoy the Smith-Morra Gambit.  There is a new series of videos out on ICC by a young IM that shows a very modern treatment which refreshes many of the older lines.  However, I am reluctant for a couple of reasons.

Firstly, the Smith-Morra Accepted (1. e4 c5 2. d4 cd 3. c3 dc) is a thicket of complications in which analysis is favored over judgement, exactly the problem I"m trying to break free of.  Secondly, the declined line with 3...Nf6 is difficult to confront, as it has the earmarks of an Alekhine's Defense where black has already gotten in ....c5.  I don't regard this as a net-plus white.

So, maybe Smith-Morra until something better comes along...but it better come along quick.  I particularly am not fond of the the lines put forth in Lev Alburt's and Dzindzi's book, the Grand Prix Attack, which occurs after 1. e4 c5 2. Nc3 Nc6 (or d6) 3. f4.  These Closed-Sicilian type positions are just not my cup of tea, although some people have made a career out of playing it.  So this is an area for more research...maybe English System or Bb5 lines?  We'll have to see.

Much of the same holds true for attacking 1...e5.  I'm honestly trying to stay away from the Danish, although I don't take it to be thoroughly unsound.  It's just not a direction I'm trying to take my chess now on a routine basis.  I'm open to suggestions...