The Caro-Kann (1.e4 c6 2.

d4 d5) is a very solid defense, very thematic,

with easy-to-understand ideas for Black to work with. Whereas the French,
in general, can generate sharper play and the Caro can be a bit passive,
this defense can be feisty too. Names like Capablanca and Botvinnik
established this opening in the eyes of chessplayers.
If youre playing for a win as White, its solidity can be downright annoying!
When IM Dean Ippolito decided to vary from his trademark Petroff Defense
(another equalizing defense), he went over to the Caro-Kann with success.
The first thing you should not do is try to refute it. Recent years have
shown a surge in the Advance Variation with 3.e5. This is an extremely
sharp line, with both sides getting their shots in. However, you really need
to be constantly up on opening theory. Since were trying to pick openings
that have GM approval and good, natural and intuitive moves, we will avoid
that. You are welcome to investigate on your own if youre motivated to do
the studying.
Over the years, several chess writers have gone with a common-sense
approach rather than hit the 20-move-deep main lines that often involve
this defense.
For example, I.A. Horowitz used to recommend this
line: 1.e4 c6 2.d4 d5 3.Nc3 dxe4 4.Nxe4 Bf5 (there is a very double-edged
line with 4...Nf6 5.Nxf6 gxf6 that many aggressive Caro players
like) 5.Ng3 Bg6 6.Nf3 Nd7 (to deter
Ne5: 6...Nf6 7.h4 h6 8.Ne5 Bh7 9.Bc4 e6 10.Qe2 Horowitz):
XIIIIIIIIY
9r+-wqkvlntr0
9zpp+nzppzpp0
9-+p+-+l+0
9+-+-+-+-0
9-+-zP-+-+0
9+-+-+NsN-0
9PzPP+-zPPzP0
9tR-vLQmKL+R0
xiiiiiiiiy
The main line now goes 7.h4 h6 8.h5 Bh7 9.Bd3 Bxd3 10.Qxd3 e6. Any
number of chess engines will tell you White has an edge, but years ago
GM Ron Henley (when he was 17!) convinced me that Blacks position was
solid as a rock. He was right. Horowitz apparently felt the same way and
recommended:
7.Bc4 e6 8.Qe2
XIIIIIIIIY
9r+-wqkvlntr0
9zpp+n+pzpp0
9-+p+p+l+0

9+-+-+-+-0
9-+LzP-+-+0
9+-+-+NsN-0
9PzPP+QzPPzP0
9tR-vL-mK-+R0
xiiiiiiiiy
Its still a relatively equal game, with White having that slight advantage
(easy development, prevention of ...e6-e5, for example); however, read
Horowitzs thoughts here:
White is for choice. A good plan for future action is 0-0, followed by the
development of the queen bishop, posting a rook on the queen file and
then the retreat of the king bishop to QN3 [b3], making way for P-QB4 [c2c4]. It is not easy for Black to meet this proposed action.
Its not the sharper book-version plan with h2-h4. Its something that a
good many players could feel comfortable with.
Nor is it just club players who feel more comfortable. Grandmaster legend
Paul Keres, in his games collection (which should be in every
chessplayers library), found himself playing former British champion Harry
Golombek in Moscow, 1956. Keres very much respected Golombeks
knowledge of the opening and relates that he wanted to deprive Golombek
of one of his trump cards. Keres, although far superior as a player, didnt
want to enter the British players den. Heres how it
opened: 1.e4 c6 2.d4 d5 3.Nc3 dxe4 4.Nxe4 Bf5 5.Ng3 Bg6 6.Bc4:
XIIIIIIIIY
9rsn-wqkvlntr0
9zpp+-zppzpp0
9-+p+-+l+0
9+-+-+-+-0
9-+LzP-+-+0
9+-+-+-sN-0
9PzPP+-zPPzP0
9tR-vLQmK-sNR0
xiiiiiiiiy
Keres doesnt put the knight on f3. What is he saying by moving the
bishop? The answer is logical. He doesnt want to play Bd3 with the
subsequent exchange because he feels Golombek knows how to simplify
and solidify. He also has another square in mind for the knight: e2. Can
you see why?
6...e6
6...Nd7 7.N1e2 e5 8.0-0 Keres; he notes that opening up the center
helps the fellow who is castled as opposed to the one who isnt.
7.N1e2:
XIIIIIIIIY

9rsn-wqkvlntr0
9zpp+-+pzpp0
9-+p+p+l+0
9+-+-+-+-0
9-+LzP-+-+0
9+-+-+-sN-0
9PzPP+NzPPzP0
9tR-vLQmK-+R0
xiiiiiiiiy
At this point, Keres comments: The commencement of an interesting idea.
In the ensuing phase of the game White wants to set in motion an attack
against the enemy kings position by f4-f5, and for this reason avoids
developing his knight on f3, so as not to get in the way of the pawn
advance. In addition White threatens when the occasion arises to play his
knight to f4 followed by h2-h4, which would leave Blacks bishop on g6 in
somewhat of a dilemma.
Keres did get his 0-0 and f2-f4-f5 in by sacrificing the f-pawn: 7...Nf6 8.00 Bd6 9.f4 Qc7 10.f5 exf5 11.Nxf5 Bxh2+ 12.Kh1 0-0
12...Bxf5 13.Rxf5 Bd6 14.Bh6!!
13.g3:
XIIIIIIIIY
9rsn-+-trk+0
9zppwq-+pzpp0
9-+p+-snl+0
9+-+-+N+-0
9-+LzP-+-+0
9+-+-+-zP-0
9PzPP+N+-vl0
9tR-vLQ+R+K0
xiiiiiiiiy
Do you see how Keres carried out his plan? Black is playing a game where
an unfamiliar position has arisen. This cost Golombek the game. Keres
played naturally and forcefully. If Black tries 13...Qd7,
then 14.Bd3 Ng4 15.Nc3 h5 16.Rf4 looks good. Notice how both Horowitz
and Keres went with off the beaten track but sound opening lines with real
plans and chances to win.
Now, youre going to be thrown a curveball! Theres another opening line
which you should consider. Why? Black can play that very
difficult 4...Nf6 5.Nxf6 gxf6 we mentioned, or he can play 4...Nd7, which
has a very devoted following. In both lines, there is a lot of theory. It can be
learned. You should be aware of that commitment.
The approach we are offering looks odd at first sight, but it was played by
Smyslov (even in a world championship match against Botvinnik!) and in

recent years by GMs Polgr, Hodgson, and Ivanchuk. It takes Black out of
all his main line Caro-Kann prep and gives White easy development. It just
looks weird at first. Unnatural. Looks can be deceiving, though, about what
is natural. Heres how we get there:
1.e4 c6 2.d4 d5 3.f3
The Fantasy Variation!
XIIIIIIIIY
9rsnlwqkvlntr0
9zpp+-zppzpp0
9-+p+-+-+0
9+-+p+-+-0
9-+-zPP+-+0
9+-+-+P+-0
9PzPP+-+PzP0
9tRNvLQmKLsNR0
xiiiiiiiiy
That f3-pawn looks awkward and that looming queen check at h4 doesnt
look like fun. White hasnt developed a piece. On the other hand, neither
has Black. The f3-pawn does support the attacked e4-pawn and its
doubtful Black can take advantage of ...Qh4+ if White is alert.
Smyslov, in his My Best Games of Chess, comments on 3.f3 in his game
against Makogonov (see below) from the 13th USSR Championship: This
move does not enjoy a wide popularity, although it has its virtues. White
strengthens his centre which causes his opponent to be extremely
accurate in the choice of replies. As well as this, the system gives to the
whole game a sharper and more individual character than in the usual
variations of the Caro-Kann, where early on the tension in the centre is
removed by pawn exchanges.
The first game I found after Smyslov was the wild encounter Boros
Krdy, Budapest
1932: 1.e4 c6 2.d4 d5 3.f3 e6 4.Be3 dxe4 5.Nd2 exf3 6.Ngxf3 Nf6 7.Bd3
Nd5 8.Qe2 Nd7 9.0-0 Be7:
XIIIIIIIIY
9r+lwqk+-tr0
9zpp+nvlpzpp0
9-+p+p+-+0
9+-+n+-+-0
9-+-zP-+-+0
9+-+LvLN+-0
9PzPPsNQ+PzP0
9tR-+-+RmK-0
xiiiiiiiiy

This position, though presenting equal chances, will warm every BlackmarDiemer players heart.
10.Ne4 0-0 11.Rae1 f6
11...Nxe3
12.Bc1 (12.Bd2) 12...Re8 13.Neg5 Nf8 14.Ne5
Leading to a draw
is 14.Nxh7 Nxh7 15.Bxh7+ Kxh7 16.Ng5+ fxg5 17.Qh5+ Kg8 18.Qf7+ Kh7
19.Rf3 Nf4 20.Bxf4 gxf4 21.Re5 [21.Rh3+ Bh4 22.Qxf4 Kg8 23.Rxh4 Rf8].
14...fxg5 15.Qh5 Bf6
Black should have taken the draw
with 15...g6 16.Nxg6 hxg6 17.Rxf8+ [17.Bxg6 Nf6 18.Qxg5 Qxd4+ 19.Kh1
{19.Be3 Qd5} 19...Rd8 20.Bd3+ Kf7 21.Qh5+ Kg8] 17...Bxf8 18.Qxg6+ Bg
7 19.Bxg5 Nf6 20.Bxf6 Qxf6 21.Qxe8+ Qf8 22.Qg6 Qf7 23.Qg5 Qf6 24.Qh
5 Qf7 25.Qg5 Qf6.
16.Bxh7+ 1-0. However, a wild draw was not enough to sway me.
I found another game right after that with a different line in Tatai Mariotti,
Reggio Emilia 196768: 1.e4 c6 2.d4 d5 3.f3 dxe4 4.fxe4 e5 5.Nf3 exd4 6.Bc4 Bb4+?:
XIIIIIIIIY
9rsnlwqk+ntr0
9zpp+-+pzpp0
9-+p+-+-+0
9+-+-+-+-0
9-vlLzpP+-+0
9+-+-+N+-0
9PzPP+-+PzP0
9tRNvLQmK-+R0
xiiiiiiiiy
Even masters make mistakes like this! When you consider the potential
threats to f7 other than the bishop (Ng5, 0-0, Qh5), you would think a red
flag would have popped up in Blacks mind. Thats why you need to review
the Primer. These opening mistakes happen over and over again at all
levels.
7.c3 dxc3 8.Bxf7+ Kxf7 9.Qxd8 cxb2+ 10.Ke2 bxa1Q 11.Ng5+ Kg6 12.Q
e8+ Kh6 13.Nf7+ 1-0.
Its either mate after 13...Kg6 14.Nxh8+ Kf6 15.Rf1+ Bf5 16.Rxf5#, or
heavy material losses
with 13...Kh5 14.Nxh8+ g6 15.Qxc8 Qxa2+ 16.Nd2 Kh6 17.Rf1 Kg7 18.Qx
b7+ Kxh8 19.Qxb4, when Bb2+ or Qc3+ is devastating.
However, there was Smyslov. By following the history of an opening and
the players who use it, you get a pretty good feel for the system. Smyslov
is also good for the Closed Sicilian with his very clear plans. This game is
Smyslov Gereben, Moscow vs. Budapest 1949:

Model Game 13
1.e4 c6 2.d4 d5 3.f3 e6
3...e5 4.dxe5 Bc5 5.Nc3 Qb6 6.Na4 Bf2+ 7.Ke2 Qd4 8.Qxd4 Bxd4 9.f4 dxe
4 10.Be3 Bxe3 11.Kxe3 f5 12.Rd1.
3...Qb6 has interesting
possibilities: 4.a4 e6 5.a5 Qd8 6.Nc3 or 4.c3 e6 5.Bd3 c5 6.exd5 exd5 7.Q
e2+ Be7 8.dxc5 Qxc5 9.Be3 Qc7 10.Na3 a6 11.Bd4 Nf6 12.Bxf6 gxf6 13.N
c2; or the
wild 4.Nc3 dxe4 5.fxe4 e5 6.dxe5 Bc5 7.Nf3 Bf2+ 8.Ke2 Be6 9.Qd3 a5 10.
Be3 Bxe3 11.Qxe3 Qxb2 12.Rb1 Qxc2+ 13.Ke1 c5 (13...Nh6 14.Nd4 Ng4
15.Qg3) 14.Rxb7 Ne7 15.Bb5+.
3...g6 4.Nc3 (4.c3 Bg7 5.Bd3 c5 6.Ne2 Nf6 7.e5 Nfd7 8.0-0) 4...Bg7 5.Be3.
4.Be3
XIIIIIIIIY
9rsnlwqkvlntr0
9zpp+-+pzpp0
9-+p+p+-+0
9+-+p+-+-0
9-+-zPP+-+0
9+-+-vLP+-0
9PzPP+-+PzP0
9tRN+QmKLsNR0
xiiiiiiiiy
The world champ was fond of this move. It develops a piece and indirectly
hits Blacks natural freeing move of ...c7-c5. There is an
alternative: 4.Nc3 Nf6 (4...Bb4 5.Nge2 [Judit Polgr played 5.Bf4, with Qd3
and Nge2 to
follow] 5...dxe4 6.a3 Ba5 [6...exf3 7.axb4 fxe2 8.Bxe2] 7.fxe4 Qh4+ 8.Ng3
e5 9.dxe5 Bc7 10.Qd4 Ne7 11.Be3 0-0 12.0-0-0) 5.e5 Nfd7 6.f4 c5 7.Nf3. It
is no longer a Caro-Kann in any case.
4...dxe4
Makogonov against Smyslov (mentioned above)
went 4...Qb6 5.Nd2 Nd7 6.Bd3 c5 7.c3 c4 8.Bc2 Qxb2 9.Ne2 Qa3 10.0-0 N
b6 11.exd5 exd5 12.Re1 Bd7 13.Nf1 (13.Bf4 0-0-0 14.Nb3 cxb3 15.axb3 Q
e7 16.Rxa7) 13...0-0-0 14.Bc1 and, with a backwards bishop move (a littletaught concept), White makes Black regret his grabbing the b2-pawn.
5.Nd2 exf3
5...Nf6 6.fxe4 c5 7.Ngf3 cxd4 8.Nxd4 e5 9.N4f3.
6.Ngxf3 Nf6 7.Nc4
XIIIIIIIIY
9rsnlwqkvl-tr0
9zpp+-+pzpp0

9-+p+psn-+0
9+-+-+-+-0
9-+NzP-+-+0
9+-+-vLN+-0
9PzPP+-+PzP0
9tR-+QmKL+R0
xiiiiiiiiy
You can see Smyslovs idea: he heads for the e5 strongpoint with his
knight supported by the other knight. If unchallenged, White may be able to
afford Ng5 with a double attack on f7. We are seven moves into the game,
and you can see your middlegame forming: Bd3, 0-0, Qe2/e1, Rd1/e1.
7...Nd5 8.Bd2
White actively preserves the bishop. It also prepares for a possible c2-c4
down the road, hitting the black knight which will find it harder to go to b4
as an option. This is not played in the game, but its worth knowing about
the thought behind this move.
8...Be7 9.Bd3 Bh4+
An utter waste of time, perhaps with the vague idea of loosening up the
white kingside. This underestimates Whites initiative.
10.g3 Bf6 11.Qe2 a5 12.a3
12.0-0-0 was probably more accurate, and would have saved White the
need to be clever later.
12...0-0 13.h4
XIIIIIIIIY
9rsnlwq-trk+0
9+p+-+pzpp0
9-+p+pvl-+0
9zp-+n+-+-0
9-+NzP-+-zP0
9zP-+L+NzP-0
9-zPPvLQ+-+0
9tR-+-mK-+R0
xiiiiiiiiy
Maybe he has to catch a bus! More prudent is 13.Ne5 and 14.0-0-0. He
points out that it threatens the famous combination 14.Bxh7+ Kxh7
15.Ng5+ Kg8 16.Qh5 Re8 17.Qxf7+ Kh8 18.Ne5 Bxe5 19.dxe5 winning
(Smyslov). With his next move, Black enables the knight to go to f6 to
defend that threat.
13...Be7 14.Ng5 Nf6 15.Ne5 b5 16.Rf1 h6
XIIIIIIIIY
9rsnlwq-trk+0
9+-+-vlpzp-0
9-+p+psn-zp0

9zpp+-sN-sN-0
9-+-zP-+-zP0
9zP-+L+-zP-0
9-zPPvLQ+-+0
9tR-+-mKR+-0
xiiiiiiiiy
This is a very important position for understanding whether you should play
this opening or not. White has a decisive win here. The opening has been
a huge success. However, if you have not studied kingside attacks on
castled positions, whats the point of playing this line? If I show this to any
number of my attacking-style friends, the first move they will look at
is 17.Rxf6!. They understand that Black will have to take with the pawn,
17...gxf6, and the kingside will be torn apart for the final attack. How do
they know Black wont play 17...Bxf6?
Because 18.Bxh7+ Kh8 19.Nexf7+ wins the queen. You will not get good at
chess just studying the opening. In the Primer I noted all the other stuff you
had best pay attention to. These sacrifices become less genius-like the
more you study similar positions. Now Smyslov wraps him up.
17.Rxf6 gxf6 18.Qh5 fxg5 19.Qxh6 f5 20.Qg6+ Kh8 21.hxg5 Bxa3
This looks a little odd, but the idea is to get the bishop off the second rank
to allow the a8-rook to play ...Ra7 for defense. Nice thought, but too late.
22.Ke2 Qd5 23.Nf3 1-0
There is one more line of considerable interest. Because of an annotated
game in the October 2010 issue of Chess Life, a large number of CaroKann players have made this their main line. The games occasion was the
U.S. Junior Championship of that year. The two combatants were two
superbly talented players, Ray Robson and Sam Shankland. Shanklands
notes as the winner with the black pieces had great influence around the
country. After Chess Life was published, I noticed it on the ICC as the line
started to emerge as the regular weapon of Caro-Kanners.
Its a fine game, and, even though the wrong side won, its worth looking
at. Oh, by the way, theres an improvement for White in the notes!
Model Game 14
Robson Shankland
U.S. Junior Chp. 2010
1.e4 c6 2.d4 d5 3.f3 dxe4 4.fxe4 e5
XIIIIIIIIY
9rsnlwqkvlntr0
9zpp+-+pzpp0
9-+p+-+-+0
9+-+-zp-+-0
9-+-zPP+-+0

9+-+-+-+-0
9PzPP+-+PzP0
9tRNvLQmKLsNR0
xiiiiiiiiy
The system of choice follows a common-sense chess idea of challenging
the center. One way not to challenge the center is by
playing 4...Nf6 5.e5 Nd5 6.c4. Its not an Alekhines Defense. Many people
might avoid 5.dxe5 in our game because of the queen exchange and
losing castling privileges, but White would have a decent chance of holding
onto the pawn or making Black lose time to get it back! The problem
with 5.dxe5 however, is 5...Qh4+. Thus, White plays a logical move:
5.Nf3 Bg4
Other moves are also seen:
5...exd4? 6.Bc4 Bc5 7.0-0 Nf6 8.Ng5 d3+ 9.Kh1 Bg4 10.Bxf7+ Ke7 11.Nf3
Kxf7 12.Ne5+ Ke8 13.Nxg4 Nxg4 14.Qxg4;
5...Be6!? 6.Nc3 exd4 (6...Nd7 7.d5) 7.Nxd4 Bb4 8.Be2 Nf6 9.Bg5 Qd6 10.
Nf5 Qxd1+ 11.Rxd1 Bxc3+ 12.bxc3 Bxf5 13.exf5 Nbd7 14.0-0 0-0.
White continues with his f7 theme, a version of which you saw above.
6.Bc4 Nd7
One example of a horror show would
be 6...Nf6 7.Bxf7+ Kxf7 8.Nxe5+ Kg8 9.Nxg4 Nxe4 10.0-0.
XIIIIIIIIY
9r+-wqkvlntr0
9zpp+n+pzpp0
9-+p+-+-+0
9+-+-zp-+-0
9-+LzPP+l+0
9+-+-+N+-0
9PzPP+-+PzP0
9tRNvLQmK-+R0
xiiiiiiiiy
An early big decision moment for White. He might
consider 7.h3 Bxf3 8.Qxf3 Qh4+ 9.g3 Qf6 10.0-0 Qxf3 11.Rxf3 Ngf6 12.c3
with very slightly the better of an equal position nothing to write home
about. Robson chooses the critical move.
7.c3 Bh5
XIIIIIIIIY
9r+-wqkvlntr0
9zpp+n+pzpp0
9-+p+-+-+0
9+-+-zp-+l0
9-+LzPP+-+0
9+-zP-+N+-0

9PzP-+-+PzP0
9tRNvLQmK-+R0
xiiiiiiiiy
Black had a critical choice to make. Obviously bad is 7...Ngf6 8.Qb3.
Much play has been given to 7...b5, but 8.Be2! Ngf6 9.a4 (9.0-0 may even
be better remember the
Purdy Rule!) 9...a6 (9...Nxe4 10.axb5 cxb5 11.Bxb5) 10.0-0 (10.axb5 cxb5
11.h3 Bh5 [11...Bxf3 12.Bxf3] 12.Nxe5 Bxe2 13.Qxe2 Nxe5 14.dxe5 Nd7 1
5.Bf4 [15.e6 Qh4+] 15...Nc5 16.0-0 Qd3) 10...Bd6 11.h3 Bh5 12.axb5 cxb5
13.Nh4. There is still much to be discovered here, but it does follow our
rule of being playable with a chance to win.
8.0-0
Another possibility is 8.Qb3 b5 9.Be2 Ngf6 10.00 Bd6 11.dxe5 Bc5+ 12.Kh1 Nxe4 13.Nfd2 Nxd2 14.Nxd2 Qh4 15.Ne4 0-0
16.Bxh5 Qxh5 17.e6 fxe6 18.Qxe6+ Kh8 19.Bf4 Qd5 20.Qxd5 cxd5 21.Nxc
5 Nxc5 22.Rad1, but there are too many moves here that could go a
different way (e.g., 15.Bxh5 and 16.e6), so well leave that to your
investigation and tastes.
8...Ngf6
XIIIIIIIIY
9r+-wqkvl-tr0
9zpp+n+pzpp0
9-+p+-sn-+0
9+-+-zp-+l0
9-+LzPP+-+0
9+-zP-+N+-0
9PzP-+-+PzP0
9tRNvLQ+RmK-0
xiiiiiiiiy
At this point, White plays 9.Nbd2, which Shankland properly criticizes as
blocking the development of the queenside. He
suggests 9.Bg5 Bd6 10.Nbd2 0-0 with an interesting and balanced
middlegame. Also, by playing Nbd2, it causes problems when the c4bishop retreats to d3, putting him under isndirect fire from the black queen
and his own queen is blocked from protecting him. As a result, Black can
play ...c6-c5 or even ...Nc5 (as Shankland points out) and the white dpawn cant capture.
Maybe you can look for something else. Heres one example to play with,
following the theme on
f7: 9.Qb3 Qc7 10.Ng5 Nb6 11.Bxf7+ Bxf7 12.Nxf7 Qxf7 13.Qxf7+ Kxf7 14.
dxe5 Bc5+ 15.Kh1 Rhe8 16.exf6 Rxe4 17.Nd2 Re2 18.Nb3 Bf2 19.fxg7 Kx
g7 20.Na5 Rb8 21.Rb1 Na4 22.g3 Have fun! Homework! Also look
at 9.Be2 and 9.Bb3.

The rest of the game demonstrates how Black counters when White
falters.
9.Nbd2 Bd6 10.Bd3 0-0 11.Qe1 Re8 12.Kh1 Bc7 13.Bc2 exd4 14.cxd4 c
5 15.dxc5 Nxc5 16.e5 Ncd7 17.Qh4 Nxe5 18.Nxe5 Bxe5 19.Nc4 Qd4 20.
Qxd4 Bxd4 21.Nd6 Re6 22.Nf5 Bc5 23.Bb3 Re5 24.Bd2 Ne4 25.Rae1 R
ae8 26.Rxe4 Rxe4 27.Ng3 Bg6 28.Nxe4 Rxe4 29.Bc3 h5 30.Bd5 Re7 31.
b4 Bb6 32.a4 a6 33.a5 Ba7 34.Re1 Rd7 35.Bf3 Kh7 36.Re8 Rc7 37.Bd2
h4 38.g3 h3 39.Bg4 f5 0-1
You now have three different ways to approach the Caro-Kann. There are
others. These were good because most of the moves were natural and the
good ones concentrated either on development or on preventing
something bad from happening. While this last game will be subject to
further theoretical scrutiny, its not the hellhole of memorization that so
many lines are.