Talk:Caro–Kann Defence

Classical variation
About this, in the Classical Variation:


 * Typically the black light-squared bishop is a problematic piece in that it has little mobility and can be confined to the awkward-looking h7 square after for example 6. h4 h6 7. h5 Bh7. Black will therefore seek to trade it off.

I don't play the Caro-Kann, so I may be missing something, but isn't the main line here 8.Nf3 Nf6 (or 8...Nd7) 9.Bd3, thus trading off the very piece which, according to the article, Black wants to get rid of? Strikes me that something is wrong here. --Camembert


 * You're right. The light squared bishop is hardly a liability and the white plan to trade it off is practically automatic now. --Malathion 5 July 2005 02:53 (UTC)

Again in the Classical Variation:


 * Should things proceed to an endgame, Black often stands well thanks to his solid pawn structure and king-side pawn majority.

I agree that the h5 pawn is weak, but otherwise I've been taught that a queen-side majority, rather than a king-side one, is an advantage in the endgame.Mbjoker 15:02, 4 September 2007 (UTC)

Time-wasting
From ''The Caro-Kann! In Black and White'' by Alexander Beliavsky and Anatoly Karpov:

Karpov on pg 153 says "The h5 pawn ties down the black kingside" and later says white has a "space advantage". Beliavsky writes that this "paralyzing" effect is "especially meaningful in the endgame". I don't think in makes sense to say that the h5 pawn push "develops nothing". --Malathion 8 July 2005 19:23 (UTC)

Correct but missing the point. Do Karpov or Beliavsky write that the Black moves ...h6 and ...Bh7 are time wasting? I don't have that book but I suspect that they don't. If there's no expert source that uses that sort of description, it can't go in the article. If there is a reliable reference, it can stay. Quale 8 July 2005 19:28 (UTC)


 * Do you have a reference (or even any reason to believe) that those move accomplish anything beneficial besides protecting the bishop from getting trapped? --Malathion 8 July 2005 19:42 (UTC)


 * Clearly that is the only reason Black plays those moves, but that doesn't make them "time-wasting," IMO. "Time-wasting" implies pointless moves. If White developed his Q-side instead of playing h4-h5 and Black played ...h6 and ...Bh7 anyway, THAT would be time-wasting. I agree that h4-h5 benefits White more than ...h6 and ...Bh7 benefit Black (that's why W usually throws in those moves before playing Bd3), but IMO that doesn't make Black's moves "time-wasting."  Krakatoa 8 July 2005 20:15 (UTC)


 * I changed the text to refer to a "king-side pawn bind that may be useful in an endgame." (Shades of Han Kmoch's "quartgrip" in Pawn Power in Chess.) I think this may capture what Karpov and Beliavsky said better than "time-wasting." See what you think. Note that this passage is now in some tension with the sentence in the next paragraph about how Black "usually" (I changed this to "often") does well if an endgame is reached. Maybe that sentence should be changed some more. Also, is it accurate to suggest that Black often leaves his king in the center? I'm not sure whether that's accurate. It might be; I'm not a great authority on this opening. I think there was a game where Beliavsky beat Larsen like a drum when his king stayed in the center too long. Krakatoa 9 July 2005 01:44 (UTC)


 * Sorry Malathion, what I wrote had a more combative tone than I intended. My point is that Black's ...h6 and ...Bh7 are in response to direct threats by White.  I think the benefit of avoiding loss of a piece alone makes the moves not a waste of time.  White's purpose is to gain space, not time.  The problem I had with the time wasting language is that it makes it sound like White wins by force, when in fact I think the variation is rather drawish.  I'm the furthest thing from an expert on the Caro-Kann, but my impression is that Black usually avoids 4...Bf5 and the Classical Variation not so much out of a fear that White is too likely to win; it's just that it's too hard for Black to win.  White has a safe position and will win or draw--Black's chances to win in this variation are practically nil unless White overextends.  I understand that MCO-14 is a few years old now, but it says this about the Classical Variation: "White can obtain a small kingside bind by advancing his pawn to h5, but to exploit this is a difficult task.  A problem with 4...Bf5 is the lack of winning chances against reasonable play by White."  MCO-14 then contrasts 4...Nd7 as favored by Karpov saying that its current popularity (as of 1999) is due to the double-edged positions that arise.  Of course we should record whatever the current thinking is on the Classical V., and it may have changed and I wouldn't know. Quale 9 July 2005 03:21 (UTC)


 * All of what Quale just said, and the quote from MCO-14, make a lot of sense to me. If you're Beliavsky, you might be able to squeeze out an endgame win by virtue of the extra space gained by h4-h5.  If you're a lesser mortal, or probably even a lesser GM, probably not -- and I'll bet even Beliavsky fails to win most of the time against GM's who play 4...Bf5. And yes, 4...Nd7! is the way to play. When I played the Caro-Kann, that's what I played.  I never lost a Caro-Kann, even against masters who were stronger than I. (Although now that I think about it, in one of those master games (against Marvin Dandridge) I stupidly played 4...Nf6 5.Nxf6+ gxf6 without knowing what I was doing, although I managed to scrounge a draw anyway. Krakatoa 9 July 2005 17:25 (UTC)


 * No problem about combative language. I was an avid CK player for many years. You two have given me a decent amount to respond to so I'll try to keep this organized.


 * I think the "time wasting" comment is okay, but I'm also satisfied with the current edit, so I am not disputing this any longer.
 * Saying that the Petrosian-Smyslov (as Karpov and Beliavsky call it in their book) is "the way to play" might be a bit specious. Leko played the main line against Kramnik and drew. Anand and Bareev have also figured prominently as a common CK main line players in recent history. Leko even beat Topalov with it in 2002. To be more concrete, my database (using the same restrictions as before: Both players rated over 2500 and within the last 5 years) shows that the main line appeared in 375 games. White wins 22% Draw 59% Black wins 19%. The Nd7 line appeared in 126 games. White wins 24% Draw 59% Black wins 17%. So really, it seems fine to say that the Nd7 line is sound, and certainly works well for you, but I would not agree with Wikipedia preferring it. Strong master level players don't prefer it at all, and I think we should regard their opinions and experience as more relevant than our own. Imo, Wikipedia should not make a preferetial judgment of any kind between these lines.
 * In response to Karatoka's query:  Also, is it accurate to suggest that Black often leaves his king in the center? Yes, to my surprise. I searched my database for all main line games in the last ten years with players rated over 2500.
 * Out of 593 games, White wins 21%, draws 61%, Black wins 18%.
 * There were 329 games where black castled kingside. White wins 23%, draws 55%, Black wins 22%.
 * There were 140 games where black castled queenside. White wins 25%, draws 60%, Black wins 15%.
 * There were 124 games where black did not castle at all. White wins 12%, draws 78% (!), Black wins 10%. --Malathion 9 July 2005 18:08 (UTC)


 * Thanks for the stats. You selected games that meet the standards that give interesting numbers--recent games by very strong players.  Although 4...Bf5 has been played since around 1900, only experience in the last 5-20 years is helpful to try to figure out where the line is now, so using 5 and 10 years as filter criteria is good.  It's interesting that Black wins more often with Bf5 than Nd7 since MCO-14 suggests that it expects the reverse.  Maybe those numbers were different when De Firmian was writing MCO-14 or perhaps he had a mistaken impression.  One of the things about the single volume opening books (MCO, NCO, BCO) is that coverage of the individual openings is of varying quality depending on the authors' knowledge and interests.  Perhaps the Caro-Kann isn't a specialty of De Firmian or his collaborators.  (And all the printed volumes are 5+ years old too.) Quale 04:57, 10 July 2005 (UTC)


 * Yes, thanks for the stats, Malathion. Interesting. My remark about 4...Nd7 being the "way to play" just reflected my own good results with it in a small number of games. I certainly wasn't suggesting that that should be the official Wikipedia verdict. Krakatoa 06:06, 10 July 2005 (UTC)

The Exchange Variation
The text says that the Panov-Botvinnik Attack "often leads to wildly tactical play thanks to the open diagonals and rapid White development. It is often thought that this type of play scores exceptionally well against the Caro-Kann . . . ." I am doubtful about the accuracy of this, and have two questions. First, I'm no great expert on the Panov-Botvinnik, but is it accurate to say that it "often leads to wildly tactical play"? That's not my impression. Second, can't someone check ChessBase and find out whether the P-B A actually "scores exceptionally well against the Caro-Kann"? Again, rightly or wrongly, that's not my impression -- but in any event we ought in this age of huge chess databases to be able to find out whether the P-B A really does score exceptionally well, instead of talking about what is "often thought" (by whom?) on that subject.


 * In my ChessBase database (Mega 2005), in all games after 1. e4 c6 2. d4 d5 3. exd5 cxd5 4. c4 where both players were rated higher than 2500, white wins 23%, draws 62%, and loses 15%. In the advance white wins 34%, draws 46%, and loses 20%. In the main line (1. e4 c6 2. d4 d5 3. Nc3 dxe4 4. Nxe4 Bf5) white wins 34%, draws 39%, loses 27%. So it would seem that at the master level, the Panov is the most drawish of the available lines. --Malathion 9 July 2005 00:15 (UTC)


 * Thanks. In light of your ChessBase research (hope my reliance on it doesn't violate the dreaded NOR guideline or policy or whatever it is), I've taken out the stuff about the Panov-Botvinnik being the sharpest line and being thought to score exceptionally well, and added statements about it being drawish at top levels and scoring less well than 3.e5. I left in the stuff about the open diagonals and rapid white development, which someone else can ax if appropriate. Krakatoa 9 July 2005 01:25 (UTC)

I would say that it is not wild, but dynamic. — Preceding unsigned comment added by Puzzlesbrony777 (talk • contribs) 18:13, 16 March 2014 (UTC)

4... Nd7 Variation Name
Is this variation definitively called the Smyslov Variation? MCO-14 and NCO have no particular name for this line, while Chessgames.com, Gameknot.com, and a number of other online sources call it the Steinitz Variation. -- Gestrin 12:58, 15 February 2007 (UTC)
 * Well, apparently somebody wrote a whole book entitled "CARO-KANN: SMYSLOV SYSTEM 4...ND7". However, variations can have several names, so if you have sources giving alternate names, by all means put it in. Sjakkalle (Check!)  13:35, 15 February 2007 (UTC)
 * The List of chess openings page, with sources, gives it as the Steinitz variation as well. Given that "Steinitz Variation" is the official ECO name, I'm changing the article to include both names for this variation. -- Gestrin 22:15, 15 February 2007 (UTC)
 * I'm very late to this party, but does anyone know why 4...Nd7 is sometimes called the Steinitz Variation? Steinitz thought the Caro-Kann was a junk opening and AFAIK never played it. FWIW, the first game chessgames.com has with 4...Nd7 is a Nimzowitsch game from 1913, by which time Steinitz had been pushing up daisies for over a decade. Krakatoa (talk) 08:44, 4 February 2010 (UTC)
 * Have looked with Google, can't find anything about why it was called the Steinitz Variation other then the obvious it was named after Steinitz. SunCreator (talk) 11:55, 4 February 2010 (UTC)
 * In the 1980s, I played 4....Nd7 a few times, but don't remember seeing a game where Steinitz tried it, either. At least the Rubinstein Variation (4.e3) vs the Nimzo-Indian was actually played once by Rubinstein! Hushpuckena (talk) 16:51, 4 February 2010 (UTC)

canonical position in main diagram?
I used the opening template to illustrate the opening here. However, I see a minor dilemma. The formal designation of the opening is based on the position after Black's first move. However, in popular parlance, it de facto begins after Black's second move; deviations from d-pawn moves on either player's second move are sometimes referred to as irregular king pawn or irregular semiopen games etc., i.e., not a Caro-Kann. I tried to split the difference by noting the second move position in the diagram, while noting formally that the opening consists of just the first moves.

I would not mind anyone who is more current than me on opening nomenclature (beyond my MCO 1983 ed) having a look and changing the moves in the template (or affirming them here) accordingly. Thanks. Baccyak4H (Yak!) 17:42, 20 August 2007 (UTC)
 * I would prefer the position after 1.e4 c6 as the main diagram. The reason is that 1.e4 c6 is more general, 2.d4 d5 is most usual, but White can deviate with 2.Nc3 or 2.d3, and it is still a Caro-Kann. While it is true that deviations from 2...d5 can lead over to other openings (in particular the Pirc Defence or Modern Defence), 2...d5 is not necessary, one can play the Caro-Kann while abstaining from that move. According to Encyclopedia of Chess Openings, codes B10 to B19 are devoted by the Caro-Kann. The most general of these chapters B10 is defined only by 1.e4 c6, while an opening sequence like 1.e4 c6 2.d4 e5 goes under code B12, still in the ten chapters devoted to the Caro-Kann. True, the opening can transpose to something else, Pirc, Modern or even the Nimzo-Indian(!), but this is less serious than neglecting the possibility for 2.d3 or 2.Nc3. Sjakkalle (Check!)  06:16, 21 August 2007 (UTC)
 * Thanks for the feedback. I will edit the template accordingly.  Baccyak4H (Yak!) 13:48, 21 August 2007 (UTC)
 * Thank you. :-) Sjakkalle (Check!)  14:08, 21 August 2007 (UTC)

Merger proposal
I came across a page called Caro-Kann Defense Quick Checkmates. I suggest that page should be merged into this article. Bluewave (talk) 13:21, 26 January 2010 (UTC)


 * Caro-Kann Defense Quick Checkmates is not an encyclopedia article and certainly cannot stand as it does now. I am in fact tempted to recommend the article be deleted, but the miniature, Reti-Tartakower is at least quite famous, even though White's line is not really the critical one. The smothered mate Alekhine demonstrates is also often used as an illustration of that theme, although 5.Qe2?! is really just playing for a cheapo and leaves a lot to be desired in terms of harmonious development. Since the red tape and bureaucracy involved in merging is less than what is required in deletion, I'm OK with merging in some, if not all the content, but it should not be a copy-paste job. Sjakkalle (Check!)  15:11, 26 January 2010 (UTC)


 * Merge to Checkmates_in_the_opening. SunCreator (talk) 15:30, 26 January 2010 (UTC)
 * Ah, yes that may indeed be a better location. --Sjakkalle (Check!)  15:40, 26 January 2010 (UTC)
 * I've no strong opinion on what happens to it...I came on it more as a reader, looking for help with the Caro-Kann, than as an editor. It just looked like an orphan in need of a better home (or no home at all, maybe). Merging here looked plausible, and I thought it would give people who have an opinion the chance to express it. I'll leave it to you guys to decide what to do with it. Bluewave (talk) 15:50, 26 January 2010 (UTC)
 * Think I'm with SunCreator on this one. Hushpuckena (talk) 08:19, 4 February 2010 (UTC)


 * merge. Bubba73 (You talkin' to me?), 15:51, 26 January 2010 (UTC)
 * The content started out in Checkmate, and was later broken out into Checkmates_in_the_opening. At some point, someone broke it out further into Caro-Kann Defense Quick Checkmates. I agree with SunCreator that it should go back into Checkmates_in_the_opening. Krakatoa (talk) 08:38, 4 February 2010 (UTC)

✅ Bubba73 (You talkin' to me?), 05:54, 24 February 2010 (UTC)

ECO section clarification
Under the ECO section, B12, two variations are arrived at through the same moves. Shouldn't white's 4th move be added to distinguish the two? —Preceding unsigned comment added by 83.249.134.194 (talk) 17:12, 22 May 2011 (UTC)
 * 1) Maroczy Variation: 1.e4 c6 2.d4 d5 3.f3
 * 2) Fantasy/Lilienfisch Variation: 1.e4 c6 2.d4 d5 3.f3

Does anyone have a reference for the name: "Lilienfisch Variation". I have found it repeated all over the net, but Lilienfisch has never played this variation! I presume they have all copied it from wikipedia. I believe no one calls this anything other than The Fantasy Variation. In the article the variation is also called The Tartakower Variation, which makes some sense as Tartakower has player this in 6 games, but I also believe no one really uses this name for the variation. Is The Maroczy Variation the continuation: 3...dxe4 4.fxe4 e5 ?, Maroczy has played this continuation twice. — Preceding unsigned comment added by 83.95.160.234 (talk) 11:29, 10 December 2014 (UTC)

Any non-internet sources for the name "Mindeno Variation"?
I couldn't find this name in a google books search. The well-known Noteboom-Mindeno game was included in Chernev's "1000 Best Short Games of Chess" and "The Bright Side of Chess" but I couldn't find any printed references that use the name "Mindeno Variation" for the common move 3...Bg4. I find it difficult to believe that an obscure Dutch player was the first person to play it. If this is an "internet name" I don't think it should be used in the article. MaxBrowne (talk) 05:11, 1 May 2016 (UTC)
 * Can't find anything in my books. Looks like it's been named based on the 1927 game but has no real credibility as a proper name. Nothing substantial to justify it. Jkmaskell (talk) 10:30, 1 May 2016 (UTC)

Hillbilly Attack
Obviously the name "Hillbilly" in reference to the 2. Bc4 line isn't a politically correct one but it is in use. Checking my Chessmaster 10th edition, it does appear under that name in the program's opening book. A Google search also pulls up some comments about its early days. It looks like it has been written about in printed form, "Chess Horizons"and "Randspringer" named as having published articles on it some time ago. The name was translated from the German "Hinterwäldler Angriff". I'm posting this here in light of the edit-warring taking place today. It's clearly a valid name. Jkmaskell (talk) 16:01, 29 May 2016 (UTC)
 * See This thread at NPOV noticeboard. Also Edit Warring Noticeboard. MaxBrowne (talk) 16:07, 29 May 2016 (UTC)
 * Added Watson source, and another one. Left comment at notice board. - DVdm (talk) 17:48, 29 May 2016 (UTC)

Kann-Mieses game
The article notes an "impressive" 17-move Kann victory over Mieses in Hamburg, 1885.

However, there's no victory in the image - and the link suggests a 24-move game.

Am I missing something? — Preceding unsigned comment added by 220.233.102.178 (talk) 14:28, 4 February 2018 (UTC)

Game results graphic
The graphic in the 'Other Lines' section is confusing. It says 3.d4 is the main line, but that is impossible since 2.d4 is the move before! I am not sure what it means but it needs altering/clarifying or removing. Knucmo2 (talk) 14:19, 29 July 2022 (UTC)