Stigma wrote on 07/11/10 at 21:31:25:

Why should "general understanding" and "pattern recognition" be opposed? This is not a rhetorical or semantic question. Euwe in his instructional books already isolated Steinitz' "positional elements" which lead to typical plans. One could say that he strived for pattern recognition on an elementary level.

Fromper wrote on 07/11/10 at 19:43:06:

Don't think so. The better your positional understanding the more likely you get promising attacking positions. This is especially true when playing the Dutch.
That game by Eric the Red shows at least one thing nicely: when White makes positional concessions to parry Black's attack Black must be able to beat White on another front.

ErictheRed wrote on 07/11/10 at 18:12:01:

Yup, that's me. There's a reason I'm only a 1700ish (USCF) player. This is also the reason I picked an opening that's known as a counter-attacking way to play against 1. d4. 

I used to play the QGD as black, but I never had any idea how to handle the closed positions. I know I should probably read some Silman books or something to learn how to handle different types of positions, but I've been mostly focused on tactics and endgames for a while now, which is how I went from 1300 to 1700 in 2 years as an adult player. The positional stuff can come later. In the mean time, I get by just playing a limited number of openings, and just trying to understand the positions that come from those openings better than my opponents. 

--Fromper

Fromper wrote on 07/11/10 at 17:06:39:


It sounds like you have a one track mind and aren't sure what to do when you can't get your standard attack going. 

You should study the closed openings more generally to see how to play these positions.  Personally, I think the Classical Dutch has a lot in common with the Queen's Indian and you have to be familiar with ideas of how to play in the center and Queenside as well as the Kingside.

At the risk of showing off, here's a game I annotated fairly deeply 5 years ago when I was rated about 2000 and trying to improve.  I had just taken up the Classical Dutch at the time.  The annotations are not all correct (I'm a stronger player now), but maybe you'll get some idea of the different plans available to Black.  Bear in mind, I was rated about 2000 and my opponent is about 1900, so it's not the highest quality game or anything, but hopefully it will help in your studies.  I probably spent about 6 hours delving deeply into analyzing this game (though it wasn't that comlicated) and I'm leaving all my comments as I wrote them at the time.  I hope you can learn from my method of self reflection, etc., more than anything else:

[Event ""]
[Site "?"]
[Date "2005.08.14"]
[Round "4"]
[White "Some guy"]
[Black "Eric"]
[Result "0-1"]
[ECO "A96"]
[WhiteElo "1907"]
[BlackElo "1991"]
[Annotator ",Eric"]
[PlyCount "52"]

{(Time Control: G/120)} 1. d4 f5 2. c4 e6 3. Nc3 Nf6 4. Nf3 ({I've played about everything
under the sun against 1.d4, and never really been happy.  I tried to play the
Nimzo Indian for a while, but I only got a Nimzo a couple of times; the rest
were Bogos or various London/Colle/Torre/Trompowsky junk systems.    For a
long time I played the Abraham/Noteboom variation of the semi-Slav (and the
main line Semi-Slav or Stonewall when White avoided it), but I never liked the
positions very much.  There are many lines (for instance, 1.d4 d5  2.c4 e6  3.
Nc3 c6  4.Nf3 dc  5.g3!?) where Black surrenders the center for a pawn.  In
many of them Black should be fine or even better, but I've always had a hard
time against White's initiative and imposing center; it's just not how I
wanted to play as Black.  I tried the KID, Gruenfeld, and Benko also (I've had
quite good results with the KID), but I decided to give the Classical Dutch a
try about 6 months when I realized that I never gave it a chance before.  So
far I'm fairly happy with it (I've only played the Dutch in 3 tournament games,
scoring 2/3), but all my experience has been in the main-line Classical.  I'm
not really sure what to play against systems where White avoids g3, for
instance:} 4. e3 b6 (4... d5 $5) 5. Bd3 Bb7 6. f3 (6. Nf3)) (4. Bg5) (4. Qc2)
4... Be7 ({Maybe} 4... Bb4 $5 {takes advantage of White's move order, but I
don't have any experience in these types of positions (for instance, I've
never played the Nimzo line 1.d4 Nf6  2.c4 e6  3.Nc3 Bb4  4.e3 b6  5.Bd3 Bb7 
6.Nf3 Ne4 followed by ...f5).}) 5. g3 {I was happy to see this transposition.}
O-O 6. Bg2 d6 7. O-O Ne4 $5 ({Simon Williams recommended this move in his book
Play the Classical Dutch; I'm not sure it's better than} 7... a5 {, though
some lines transpose.}) 8. d5 $2 {This is a big mistake which allows Black his
ideal pawn formation in the center and weakend White's queenside.  Better was:}
(8. Qc2 Nxc3 9. Qxc3 (9. bxc3 Nc6 10. e4 e5 $11) 9... a5 {, though after
something like} 10. a3 Bf6 11. Re1 Nc6 12. e4 e5 {I'm very happy with Black 
(though of course there are alternatives).}) 8... Nxc3 $1 9. bxc3 e5 10. Rb1 $6
({During the game I thought White should try} 10. c5 {, though after} Nd7 $1
11. cxd6 cxd6 {Black has good play against the c-pawn and control of the c4
and c5-squares.}) ({Best is} 10. Nd2 {, preparing e2-e4.  I didn't think 10.
Rb1 was good, because Black will probably play ...b7-b6 anyway, when the rook
isn't doing anything on the b-file.  Also, if the bishop gets to f5 (after
White's e- and Black's f-pawns are exchanged) then Black can win a tempo by
attacking the rook.}) 10... Nd7 {(W 1:47, B 1:54)} ({I was pleased with this
move during the game, because I almost played} 10... Na6 {instinctually.  On
a6, however, the knight's only future is to be traded for the dark-squared
bishop on c5, which I didn't want to allow.}) 11. Ba3 {White should be trying
for e2-e4} b6 {(W 1:39, B 1:50) Controlling c5 so that I can play ...Nf6.} 12.
Nd2 Nf6 $6 {(B 1: 39)} ({I considered} 12... Bg5 $1 {for a while, and I don't
know why I rejected it now:} 13. e3 (13. Nf3 f4 $15 (13... Bh6 $5)) (13. e4 $2
Bxd2 14. Qxd2 f4 $17) 13... e4 $1 14. f4 {(To stop 14...Ne5)} exf3 15. Qxf3 Nf6
{is much better for Black}) 13. e3 ({Better was} 13. e4 fxe4 14. Nxe4 $15 {;
Black has the better structure but White is much less passive than in the game.
}) 13... Qe8 $1 {(W 1:31, B 1:36) eyeing the Kingside and the important
a4-square} 14. f4 $2 {I think White was worried about a future K-side attack;
he also probably hoped to get the d4-square for his knight.  I thought for a
while (14 minutes) about how to deal with the center.  I considered 14...ef,
when I could put my bishop on f6 and try to utilize the e-file, but didn't
think it was that promising.  I wanted to play 14...e4, but was concerned
about locking the center pawns and especially of giving White the d4-square
for his knight.  I eventually realized that I could still play the pawn breaks
...g7-g5, c7-c6, and ...b6-b5 to try to infiltrate White's position.  I
calculated that I could keep White's knight out of d4 for a long time: 14...e4
15.Nb3? Qa4;  15.Bc1 Ba6  16.Qe2? Nxd5; 16.Rb4? c6  17.dc Qxc6 18.Nb3? d5!  19.
Nd4 Qc8, winning a pawn.  Once I realized that White couldn't use the
d4-square, my choice was simple:} e4 $1 {(W 1:30, B 1:23) Now both White's
bishops are bad.  The next phase of the game consists of trying to prevent
White getting his knight to d4, which I felt was the "thread" of the position.}
15. Bc1 Ba6 ({maybe} 15... c6 $5 {is better:} 16. Qb3 $1 (16. Nb3 $2 Ba6 $17)
16... cxd5 (16... Bb7 $2 17. c5 $1) 17. cxd5 Bb7 18. c4 Nd7 $15) 16. Rb4 $2 ({
This looks like a mistake; the rook is terribly placed here.  Perhaps White
intended} 16. Qe2 $4 {and only now saw} Nxd5 $19) ({Best must be} 16. Re1 {to
facilitate Bf1, but after} c6 17. Qb3 cxd5 18. cxd5 Bd3 $1 {Black is clearly
better.  On b4 the rook is very clumsily placed; it's only job is to defend c4
and it can be attacked by the bishop on e7.}) 16... c5 $2 {(W 1:20, B 1:02) I
was enticed into attacking the rook after debating between this or 16...c6.  I
thought White would take en-pessant in order to keep control of the d4-square 
(which he did), but in fact White should just play 17.Rb1!.  I thought I could
still get an edge after 17.Rb1 b5, but 18.cb Bxb5  19.c4 is only slightly
better for Black.} (16... c6 $1 {would have forced White to take on c6 and
transpose to the game, if} 17. Ra4 $2 Bb7 18. dxc6 Bxc6 {White is in trouble.})
17. dxc6 $2 (17. Rb1 $1 b5 18. cxb5 Bxb5 19. c4 Ba6 $15 {I think both myself
and my opponent were so concerned about the d4-square that we didn't seriously
consider that White could allow the pawn to stay on c5.}) 17... Qxc6 $17 18.
Nb3 $2 ({This loses a pawn, thought it's hard to suggest anything else; perhaps
} 18. Qa4 {?  Black is still much better though, due to the weak c-pawns.})
18... d5 $1 {Around here I was actually pretty amazed that a non-forcing line
I had looked at before playing my 14th move had arisen on the board!  Although
both sides have made mistakes, this is the type of strategic, prophylactic
game that I'm fond of.} 19. Nd4 Qd7 20. Ra4 $2 ({I don't know what White
thought he was doing with this rook;} 20. Rb1 {already!}) 20... Bxc4 $19 21.
Re1 Bc5 22. Ba3 Rfd8 23. Bxc5 $2 {White must have been demoralized that he
lost the c4-pawn after getting his knight where it wanted to go; I don't
understand why he would let me chase it away now.} bxc5 24. Ne2 Rdb8 25. Qc2 $4
Bd3 26. Qd1 Rb1 {This was a sloppy game and White didn't put up any resistance,
but I feel good about it because I feel that I understood the "essence" of the
position from move 14 onward.} 0-1

Thanks for the suggestions. A lot of what you guys have said is very similar to how I did play in these games, or how my opponents played. 

I've already figured out that Qh5 doesn't work if the Nf3 is the only thing between the white and black queens. And I did have one opponent play Nge2 instead of Nf3, then push f4, after which I tried too hard to continue on the king side with g5, leaving the center too weak, which he was able to exploit.

In one of my games, my opponent let me play e5 very early (he played the English without d4), and I just wasn't sure where to go from there. Usually, I'd be looking to use e5 to support f4, getting my c8 bishop into the game, especially since his light squares are so weak. I'd love to trade light squared bishops and invade the king side with my other pieces. That was the game where I played f4 as a pawn sac, and it wasn't quite strong enough for a king side attack. I did get enough compensation for the pawn that I was able to win back my pawn later to return to material equality, probably still maintaining a slight positional edge, though I couldn't figure out how to make anything out of it. In that case, I think I may have just started my attack a little too soon.

I just want to point out, not too condescendingly I hope, that these are things you really need to practice working out yourself.  If you want to become a better player, you should set up your board with the position after 8.e3, get a paper and pencil, and start making notes.  Ask yourself what is different about the current position than other positions you are familiar with.  You started to do that, noting the Black will have a hard time playing ...f5-f4, but you didn't continue to discover that Black should not put his Queen on h5 or that White has a very strong idea in f2-f4.

After writing down some notes, listing imbalances, jotting down a few tentative lines, etc., you should go to Chessbase and "check" your work.  In this case, it can be hard to know what to search for, because you're not looking for a specific position but a general strategic idea.  What I would recommend is just searching for a pawn structure, not caring where the pieces are.  Search for a position with White pawns on g3, f2, e3, d4, and c4, and Black pawns on f5, e6, and b7.  I omit some of the pawns, because Black may or may not have played ...a7-a5, his pawn might still be on d7, White may or may not have played b2-b3 or a2-a3, etc.  If your results are too broad add a few other features in, like Kings on g8 and g1, a Bishop on g2, etc., so you don't end up with a bunch of endgame positions.  Now compare your notes to the Master games you see.

It's OK to ask questions of other players, of course, but this is the best method I know of to figure this stuff out on your own, and will make you a better player in the long run.  It's best to do this first and then get a strong player to check your work.  Good luck!

I'm not sure I'd call e2-e3 a mistake, though it's clearly not best.  I think it was played a lot in the pre-Fischer era.  It slows the tempo of the game down a bit and you have to adjust.  Think about how the move changes the position and you'll be able to come up with a plan:
1. As you noted, it makes it more difficult for Black to play ...f5-f4 (though certainly not impossible, especially if White later moves his dark-squared Bishop to b2).
2. It puts White's Queen in communication with the h5-square, so that a ...Qd8-e8-h5 maneuvre can be met by moving the f3-Knight, offering a Queen exchange and messing up Black's attacking chances.
3. If Black isn't careful, White will move his King's Knight (say, Nf3-e1-d3) and play f2-f4!.  This is a very powerful idea that can leave Black with a strategically lost game if he isn't careful.

Here is an illustrative game to show exactly what White is looking for.  This game is in the introduction to Golombek's translation of The Art of the Middle Game by Keres and Kotov:

[Event "?"]
[Site "?"]
[Date "????.??.??"]
[Round "?"]
[White "Golombek"]
[Black "Parr"]
[Result "*"]
[ECO "A96"]
[PlyCount "59"]

1. d4 f5 2. g3 Nf6 3. Bg2 e6 4. Nf3 Be7 5. O-O O-O 6. c4 Qe8 7. Nc3 d6 8. b3
Qh5 9. e3 g5 10. Ba3 a5 11. Ne5 Qe8 12. Nd3 Nbd7 13. f4 Ng4 14. Qe2 Bf6 15. Nb5
Qd8 16. h3 Nh6 17. fxg5 Bxg5 18. e4 fxe4 19. Rxf8+ Qxf8 20. Qxe4 Nf6 21. Qe2
Nf5 22. g4 Nh4 23. Nxc7 Rb8 24. Nxe6 Qe7 25. Re1 Bxe6 26. Qxe6+ Qxe6 27. Rxe6
Re8 28. Rxe8+ Nxe8 29. Bxb7 Be3+ 30. Kf1 

Black was basically strategically lost at move 13.

It's late and I had to enter that game by hand because it's not in my database, so I'm going to go to bed without explaining in depth what I think Black should do.  But here are some pointers:

1. When White plays e2-e3 and his Queen is still on d1, Black needs to abandon the Qd8-e8-h5 idea because it's just a huge waste of time.  That means he should play in the center!

2. In general, Black should arrange to play ...e7-e5 if White plays Nf3-e1 (threatening f2-f4).

3. Black should always be willing to consider changing to a favorable Stonewall setup if the opportunity arises.

4. Since ...Qd8-e8-h5 isn't on the table, ...Ne4 and ...Bf6 should be considered.   

5. Actually ...Qd8-e8 is still a good move, supporting ...e6-e5 (possibly with ...Be7-d8) and sometimes allowing a x...Nc6 x2.d5 Nd8 maneuver.  Just don't go to h5 anytime soon!   

5. Lastly, it's possible to play a weird sort of "hedgehog" position with ...Na6, ...c6, etc., but I don't usually trust this plan.

The position gets more interesting after you manage to play ...e6-e5, because now you have the option to continue on with ...e5-e4, which can be very double-edged but is worth considering in a number of positions.  Whether or not White throws in d4xe5 will affect things, obviously.  We can't go much further talking about these types of positions without looking at a concrete line, and I'm too tired for that right now.

Which plan you choose will depend on the exact position.  In your particular game, at first glance I would favour 8...Ne4.  Compared to the 7...Ne4 line, I think Black's ...a7-a5 is more useful than White's e2-e3.