Wednesday, August 05, 2009

Evolutionary view of the Main-line Caro-Kann to the 4…Nd7

This essay is a brief cursory look at the evolutionary track of this relatively young Defense for black as explored by an amateur chess historian hack.

Max Weiss plays with Mikhail Chigorin

Pre-historic Caro-Kann
Looking back over a 100 years, I decided to dig into the history of who in their right minds would play the Caro-Kann. As the story goes, back in 1886 Horatio Caro printed an analysis that he and Marcus Kann did on the move order 1.e4 c6 and it’s merits. This came at a time when the swashbuckling era of the romantic age of chess was being tempered by the dawn of the classical era known more for it’s positional style. Steinitz was transitioning in this period from his aggressive attacking style to what some claimed as heresy with his new “weaker” positional style. At the time, the French defense was the solid alternative to a mirrored e-pawn game but that came with the “burden” of developing Black’s problem child, the Queen’s Bishop.

Doing some data-mining on earlier C-K games shows a Skotish Master John Cochrane to be one of the first to encounter this defense as white ( search in with Somacarana employing it in Calcutta. in 1856. One of these early games transposed to the Bird as King’s gambit was still in flavor back then. For the most part, white played an exchange variation with 2.Nf3

Winawer shows up in 1883 on the Black side and successfully beats Josef Noa in an early exchange variation that has Nf3 before Nc3. Max Weis in 1883 had the most success with 1…c6 in an early exchange variation with Bd3 played before Nf3. His draw against Paulsen, shows a stronger exchange variation with Nc3 played before Nf3 and resembles most of what we see today in the exchange variation with themes around going after the Bishop via Nh4 etc.

Dawn of the Caro-Kann:

In 1885, Marcus Kann beats Jacques Mieses with what appears to be a close resemblance to the modern advanced variation (1.e4 c6 2.d4 d5 3.e5 Bf5 ). Mieses exchanges bishops on d3 and plays f4. Not having seen the translation of Die Bruederschaft, the publication where Horatio Caro published the analysis in 1856, I suspect this game along with the explorations of Max Weis and others was a frame of reference.

Steinitz makes no mention of this opening in his landmark book, The Modern Chess Instructor.

Oddly enough, in looking at the games of Horatio Caro, he didn’t play this opening all that much. The first occurrence of Horatio playing this is against Pillsbury in 1905. He employs the 4…Nf6 version and loses in a game that left his king in the center. In he same year, he plays the same variation and loses to Moritz Lewitt at the Berlin Championship in 1905 in round 6 of the match. You see a now classic maneuver of the white knight being allowed to capture the Bishop on h5 followed by Qa5 check and sliding over to capture the piece back. But he gets into trouble by leaving his queen vulnerable for a double attack. He comes back in round 8, he sticks to the same variation and wins with more cautious play. By round 12 he beats Lewitt with a variation now seen today in the main line 4….Bf5.

Mainline Caro- Kann with 4…Bf5 ( Tarrasch Talking the Caro-Kann)
By this time, Siegbert Tarrasch mentions this opening in his 1931 book, The Game of Chess ( available online at google books for free) as “…cannot be considered as theoretically correct since (1…c6) does nothing towards development. In actual practice, however, as recent experience has shown, it can quite well be played since it sets the first player the enormously difficult problem of maintaining and increasing his slight advantage from the opening, and in attempting to solve the problem it is very easy to go astray”
This doesn’t come as no surprise since his rivals, mainly the hypermoderns, were experimenting with this defense. Nimzovitch, Reti, and Tartakower all played this line in the 1910’s. It wasn’t until Jose Raul Capablanca came along and had some successes with this line that some wanted to call this the Capablanca Variation. It really was Flohr who popularized this defense in the 1930’s and 40’s.

The problem with the main line, 1.e4 c6 2. d4 d5 3. Nc3 dxe4 4. Nxe4 Bf5 , is that if White plays aggressively enough and goes after the Bishop, Black’s game is centered around trying to equalize. Central control is temporarily suspended as he must find a safe haven for the Bishop. White’s strongest lines are to go after the Bishop with the h-pawn and create further weaknesses in Black’s pawn structure. White must maintain constant pressure though as Black can recoil once given the initiative.

This led Flohr to explore new methods. Going back to old Nf6 lines drew him to 4… Nd7 to set up 5…Nf6. It was a means to keep the center active and allows similar piece set ups like the main line.

Moving to 4…Nd7

Smyslov, Petrosian and Flohr were all playing this line in the 1950’s and 60’s. It was at first considered a drawing line. White’s early successes found 5. Bc4 to be a strong line especially if followed up with 6Ng5 , castling long and opposite of Black’s kingside castle only to be followed by a pawn storm.
Enter, Anatoly Karpov, who took this project on and discovered the solid pawn structure for Black and no early h-pawn pushes from White lead to some better endgames for Black. It also opened up strong attacking lines with opposite side castling and piece sacrifices.

Recently, the sharpest line against the 4….Nd7 line is 5Ng5. The most important game is Gary Kasparov’s defeat against Deep Blue. 1.e4 c6, Normally a Sicilian player, Kasparov decided to play this closed line against the computer 2.d4 d5 3.Nc3 dxe4 4.Nxe4 Nd7 5.Ng5; Joel Benjamin was very helpful in fine tuning the algorithm for this project and programmed this sharp line.
This relatively recent innovation breaks one of the classic opening principles ("don't move the same piece twice in the opening"), but puts pressure on the weak f7 square. Kasparov had played this move himself as White at least three times earlier. 5...Ngf6 6.Bd3 e6 7.N1f3 h6??

A strange blunder by Kasparov, one of the most theoretically knowledgeable players in chess history. Did Kasparov get his opening moves mixed up, playing ...h6 a move too early? He argues that the played this because the computer played “materially” in earlier games. The normal 7...Bd6 8.Qe2 h6 9.Ne4 Nxe4 10.Qxe4 was played in Kasparov(!)-Kamsky, 1994 and Kasparov-Epishin, 1995, among other games. The upcoming sacrifice is well known to theory and Kasparov knew about it (in fact, there are some reports that he even wrote an article supporting 8.Nxe6 as a refutation). He didn’t know that Joel programmed it to play the next move without calculating.

8.Nxe6! The rest of this sad turn of events follows: 8...Qe7 9.0-0 fxe6 10.Bg6+ Kd8 11.Bf411...b512.a4 Bb7 13.Re1 Nd5 14.Bg3 Kc8 15.axb5 cxb5 16.Qd3 Bc6 17.Bf5
17...exf5 18.Rxe7 Bxe7 19.c4 Black Resigns

This wasn’t a deep dive by any means but I felt like sharing my evolutionay study in the mainline of this relatively young opening known as the Caro-Kann. I play the main line and now am eager to play the 4…Nd7 variation in an upcoming event.

Yes, I detail my repertoire liberally here on my blog. When I play you, I expect a good game now. You’d better be prepared, I want to learn this one!


dfan said...

This is really interesting, thanks! I love opening history, and I wish there were more written on the subject. The recent book The Greatest Ever Chess Opening Ideas was pretty fun but it was more about taking 50 snapshots than following a historical line.

Linuxguyonfics said...

I like the Nd7 variation as Black, if Black is content to equalize first and withstand White's withering attacks - White usually does get a great looking attack, so it's better to be happy with the draw as Black, if you can get it.

I don't play well against ...Bf5 variation as White. It's very hard to gain advantage and very hard to avoid losing to it even, in my experience. Main line that is. I've switched to the Advance variation, but play it quietly. I don't see many Caro's, so this works for me.

Joshua said...

I am under the lasting impression that 4...Nd7 is under a cloud. There is much discussion of possible refutations in the analysis columns on, and the only serious grandmaster I have seen play the line in recent years in Alexander Morozevich... in a game he lost badly. I think that the lines where black is forced to sidestep with his king instead of castling are pretty bad for black.

Anonymous said...

I thought he had heard that move wasn't in its opening book.

Rolling Pawns said...

Lately I follow recommendation of GM Bareev which he gave during his lecture and play 3. f3 (1. e4 c6 2. d4 d5 3. f3). It gives pretty good results. I think it takes Black out of their intended play, sometimes it transforms to French.

Lu Egner said...

Which Max Weiss is the player on the photo? Max Ignaz Weiss or Miksa Weiss?