TWO KNIGHTS DEFENSE: FOUR POSTAL GAMES
by Mark Morss NM, NCM

The Two Knights is a cornerstone of the theory of the open games.  
It creates an imbalance between time and material was 
characteristic of many games of the Morphy era.  The opportunity 
to play the black side of the Two Knights arises frequently in 
postal chess, in part because many repertoire books have been 
written that advocating white's side of this opening.

The games given here were played in USCF postal sections during 
1992-1993, against oppenents who were all Postal Experts.


Game 1.  Angelo Calstaldo (Sunnyvale CA) - Mark Morss

1.  e4 e5  2. Nf3 Nc6 3. Bc4 Nf3  4. Ng5 ...
--------------------------------------------
Objectively, this is probably white's best move.  It's a difficult 
move for many white players to make, because it yields the 
initiative -- "the birthright of the white pieces," as Tarrasch 
put it.

4. ...d5  5. exd5 Na5
---------------------
An underrated system is 5...b5. 
     
6. Bb5+ c6  7. dxc6 bxc6  8. Qf3 ...
------------------------------------
This is extremely challenging and, like many lines of the Two 
Knights, requires both players to know some theory.  Objectively 
better is 8. Be2, when Melts - Boom, 15th World Postal Champ. 
Semifinals, 1984/89, went 8...h6  9. Nf3 e4  10. Ne5 Bd3  11. f4 
exf3  12. Nxf3 O-O  13. O-O c5  14. Kh1 Qc7  15. Nc3 a6  16. d4 
Bb7  17. d5 (equalizing is 17. Be3 Rad8  18. Bg1 Rfe8  19. Re1 Nc6  
20. d5 Nb4) 17...Rad8  18. Be3 Rfe8  19. Qd3 and now 19...Bxd5! 
would have established black's advantage. 

8. ...Rb8  9. Bd3 ...
---------------------
After 9. Bxc6+ Nxc6  10. Qxc6+ Nd7, Black's initiative is worth 
more than the two sacrificed pawns.

9. ...h6  10. Ne4 Nd5 11. Nec3 ...
----------------------------------
A dubious, unbooked move.  It is often true of postal chess that 
the first unbooked move decisively loses -- when it does not 
decisively win!  Bad are 11. O-O Nb4 and 11. Nbc3 Nf4.  
Theoretically recommended is 11. Ng3 g6, when Black plans a 
kingside pawn-storm and appears to have some advantage.  Some 
recent top-level experience is 11. b3 g6  12. Qg3 Bg7  13. Ba3 Nb4 
with advantage to black in van der Wiel - Torre, 1980.

12. ...Bd6
----------
This deliberately risky move attracted me because it looked so 
natural.  Probably better is 12...Bc5.  

13. a3 ...
----------
White can't afford to play this slowly.  I expected 13. Nxd5 cxd5  
14. Qxd5 O-O  15. Qf3 e4  16. Bxe4 Re8, when black throws his 
pawns to the winds but seems to have some
compensation (16. O-O Bxh7+).

13. ...O-O  14. b4 f5  15. Nxd5 ...
-----------------------------------
Or 15. bxa5 e5 and black regains the piece with a favorable 
opening of the position.

15. ...e4  16. Qe2 ...
----------------------
Or 15. Bxe4 fxe4  16. Qe3 (16. Qe2 cxd5  17. bxa5 Qxa5 followed by 
...Ba6) 16...cxd5  17. bxa5 Qf6 or 17...Ba6.

17. ...exd6
-----------
Also winning is 17...cxd5, but I thought it was best to open the 
e-file.  At this point in my analyses, I foresaw the actual 
conclusion of the game as one of the possibilities.

17.  Qxd6 cxd5  18. Qxd5+ ...
-----------------------------
Here white optimistically offered a draw.  

19. ...Kh8  20. Qxa5 Re8+  21. Kd1 ...
--------------------------------------
Now there is nothing left to defend white's kingside.  But 21. Kf1 
Qe7 is even more obviously losing.

21. ...Qg5  22. Re1 Qg4+
------------------------
An important finesse.

23. f3 Rxe8+  24. Kxe8 Qxg7  25. Qd8+ ...
-----------------------------------------
Not the best, but white is lost in any case.  The threat of 
26...Ba6 is very strong.

26. ...Kh7  27. Qxd6 Ba6
------------------------
A position fraught with irony.  White resigned.


Game 2.  Robert Clark (Austin TX)  - Mark Morss

1.  e4 e5  2. Nf3 Nc6  3. Bc4 Nf6  4. d4 ...
--------------------------------------------
This move is recommended by several repertoire books.  

4. ...exd4
----------
The same position can arise from 1. e4 e5  2. Nf3 Nc6  3. d4 exd4  
4. Bc4 Nf6!.

5. e5 d5  6. Bb5 Ne4  7. Nxd4 Bc5
---------------------------------
This is much better than 7...Bd7, but it requires a good deal of 
theoretical knowledge. 

8. Nxc6 ...
-----------
An extraordinary move that has recently attracted attention, after 
lying dormant in chess opening theory for more than 75 years.  

8. ...Bxf2+  9. Kf1 Qh4
-----------------------
The "Kaidanz Variation."  Dr. Hermann Kaidanz authored an 
extensive analysis of this line that appeared in the Wiener 
Schachzeitung in 1904.  Twelve years later, another lengthy 
analysis of 9...Qh4, credited to Kaidanz, appeared in the 1916 
edition of the Handbuch des Schachspiels (edited by Karl 
Schlecter), the ECO of its day.  Both the 1904 and 1916 analyses 
currently can be obtained at the Cleveland Public Library. 

Reflecting the obscurity of the variation, modern openings manuals 
contain just one or two brief continuations after 9...Qh4, not 
nearly enough to make sense of Kaidanz' idea.  In 1991, Rini Kuif, 
a Dutch IM writing in New in Chess Yearbook 20, expressed distrust 
of 9...Qh4 (without being aware of Kaidanz' analysis) and 
recommended 9...bxc6  10. Bxc6+ Kf8.  (Before the game, my 
opponent also considered this best for black)  The Handbuch 
claims that Kuif's idea is refuted by 11. Qd3! and now:

     I.  11...Qh4 (11...Rb8  12. Bxd5)  12. Nc3 Nc5  13. Qf3 Bd4  
     14. Bxa8 Bxc3  15. Be3.

     II.  11...Bf5  12. Qf3 Bc8  (12...Bd7  13. e6)  13. Nd2 Nxd2+  
     14. Bxd2 Bc5  15. Bxd5.

10. Qxd5 ...
------------
There are several alternatives.  I can only give the barest 
outline of Kaidanz' analyses:

     I.  10. Nd4+ c6 11. Nf3 (or 11. Nxc6 bxc6  12. Bxc6+ Ke7  13. 
     c4 Ng3+  14. Kxf7 Ne4+  15. Ke2 Qf2+  16. Kd3 dxc4+  17. Kxc4 
     Ba6) 11...Ng3+  12. Kxf2 Ne4+  13. Ke2 Qf2+  14. Kd3 Bf5!  
     14. Nd4 Bg6.

     II.  10. Nc3 O-O  11. Nxe4 (11. Qxd5 Nxc3  12. bxc3 bxc6 with 
     a strong attack) 11...dxe4  12. Qe2 (12. Nd4 c5 followed by 
     13...Bg4 with strong attack) 12...bxc6  13. Bxc6 Bb6 with 
     the threat of ...Ba6.

     III.  10. Qe2 O-O  11. Be3 bxc6 12. Bxc6 Bg4  13. Qd3 Rad8.

     IV.  10. Nxa7+ c6 11. Nxc8 Rxc8  12. Be2 Ba7  13. Qe1 Nf2  
     14. c3 O-O  15. Nd2 Qe7.

10. ...Bc5
----------
Kuif was unaware of this.  He gave 10...O-O  11. Nd2 and 10...Be3  
11. Nd4+ Ke7  12. Nxe6 fxe6  13. Qd7+ as good for white.

11.  Be3! ...
-------------
This great try to save white's game wasn't considered by Kaidanz.  
A skeletal outline of Kaidanz' analysis:

     I.  11. Nd4+ c6  12. Be3 O-O  13. Qb3 Be6!  14. Bc4 (other 
     moves also fail) 14...Bxd4  15. Bxd4 b5.

     II.  11. g3 Qh3+  12. Ke1 Bf2+  13. Ke2 Qg4+  14. Kd3 (14. 
     Kf1 O-O  15. Ne7+ Kh8  16. Nxc8 Qf3) 14...Nc5+  15. Kc3 O-O  
     16. Ne7+ Kh8  17. Qd1 Ne4+  18. Kb3 Qe6+  19. Qd5 Nc5+.  

     III.  11. Qd8+ Qxd8  12. Nxd8+ Kxd8.            

11. ...Ng3+
-----------
Safe, solid and probably best.  My opponent had prepared 11...Bxe3  
12. g3, when it isn't easy for Black to continue his attack.  

12. hxg3 Qxh8+  13. Kf2 Bxe3+  14. Kxe3 O-O
-------------------------------------------
Black has only a rook versus two minor pieces, but white's 
insecure King, weak pawns, and difficulties along the back rank 
weigh strongly in black's favor.

15. Ne7+ Kh8  16. Nxc8 Qc1+  17. Qd2 ...
----------------------------------------
Otherwise Black retains the threat of ...c6.

17. ...Qxb2  18. Qd4 ...
------------------------
Or 18. Nc3 Qxa1  19. Ne7 Rad8.

18. ...Qxb5  19. Ne7 Rae8
-------------------------
Black intends to break through on the e- and f-files.

20. Nc3 Qb7  21. Rb1 Qxc2  22. Ned5 Qxg2
----------------------------------------
I also thought about 22...f6  23. e6 Rxe6+  24. Kf6.

23. Ne4 ...
-----------
Now white seems to have a pretty solid setup.


23. ...c5!  24. Qxc5 f5
-----------------------
Here white resigned, albeit somewhat prematurely.  I expected 25. 
Nd6 Rxe5+  26. Kd4 and 26...Qf2! wins.

Game 3.  Jack Maxfield (Wichita Falls TX) - Mark Morss   

1. e4 e5  2. Nf3 Nc6  3. Bc4 Nf6 4. d4 exd4  5. e5 d5  6. Bb5 Ne4
7. Nxd4 Bc5  8. O-O
-----------------------------------------------------------------
This was once considered the main line, and was recommended in 
the first (1984) edition of Harding's "Evan's Gambit: and a System 
versus the Two Knights".  Recent theory (notably the second, 
1991, edition of Harding's book!) recommends 8. Be3.  There is a 
great deal of experience with this, and the critical line is 
8...Bd7 (8...O-O is unsound)  9. Bxc6 bxc6  10. Nd2! Nxd2!  11. 
Qxd2 Qe7! (11...O-O  12. O-O f6  13. Nxc6!)  12. Nb3 Bb6  13. Qc3 
O-O  14. O-O Rfe8 (not 14...Rae8  15. Bc5)  15. f4 f6  16. Bc5 
fxe5  17. fxe4 Qh4 (not 17...Qg5  18. Rae1) and black was fine in 
Rinkis-Poleschuk, USSR postal 1984.  Harding claims advantage for 
white based on 18. Nd2 d4  19. Qa3 and "black cannot play 
19...Rxe5 because of 20. Bxb6 and 21. Nf3 winning the exchange."  
This is false, because after 20. Bxb6 black has the zwischenzug 
20...Re6. 

In the foregoing line, white tried to batten down the d4 and c5 
squares, fixing black's weak c-pawns.  Black's counter was to use 
...f6 to rip the position open and gain play with his major 
pieces.  This dynamic is typical of the whole variation with 4. d4 
and 5. e5. 

8. ...O-O  9. Nxc6 ...
----------------------
The brother of 8. Nxc6 (Game 2), and just as dubious.  Theory 
used to recommend 9. Bxc6 bxc6  10. Be3.  The critical 
continuation is 10...Qe8  11. Nd2 Nxd2  12. Qxd2 Qxe5  13. Nxc6 
Bxe3  14. fxe6 (14. Qxe6 Qxe6  15. fxe6 Bb7! is good for black) 
14...Qxb2  15. Rb1 (15. Ne7+ Kh8  16. Nxd5 Ba6 favors black) 
15...Qa6 (15...Qxa7  16. Qxd4 favors white according to Harding)  
16. Qd4! Ba6!  17. Rf4 (17. Rb3? drops a piece to 17...Qd6, and 
17. Rf3 Qxa7  18. Ra1 Qc4 is also good for black) and Harding 
considers this position as favoring white in his 1984 edition.  He 
gives 17...Qxa7  18. Ra1 Qc4  19. Ne7+ Ke8  20. Qe5 and white has 
a strong initiative.  However, black is in no hurry to take the a-
pawn.  He can play 17...Bc4, defending the d-pawn and preparing 
...Rae8.  In my opinion, white lacks sufficient compensation for 
his weak pawns.  

In his 1991 edition, Harding doesn't say why he gives up on 8. O-O 
and recommends instead a completely different branch (8. Be3).

A very recent challenge to black's system is 9. Bxc6 bxc6  10. 
Nxc6, conceived by Australian GM Ian Rogers.  

Rogers-Pavlovic, Sokobanja 1989, went 10...Qd7  (10...Qh4  11. 
Be3 Ba6  12. Bxc5! Bxf1  13. Be3! Ba6!  14. Na3! favored white in 
Esteves-Peres, Cuba 1991)  11. Nd4 Qe7!  12. Bf4  (12. Be3 Qxe5  
13. c3 Bd6  14. Nf3 Qh5 is good for black) 12...f6!  13. e6 (13. 
exf6 Qxf6  14. Be3 Ba6) 13...Bxe6 and now:

     I.  The game with Pavlovic went 14. c3 Bf7  15. Nd2 Rfe8  
     16. N2b3 (best is 16. a4 a5  17. N2b3 Bb6) 16...Bb6  17. a4 
     c5!  18. Nf3 a5 with somewhat better play for black.

     II.  Rogers later recommended 14. Nxe6 Qxe6  15. Bxc7 Rac8  
     16. Bg3 (16. Bf4 Nxf2!  17. Rxf2 Bxf2+  18. Kxf2 Qf5) 16...f5  

     17. Bf4 as favoring white.  I think this claim is pretty 
     optimistic.  For example, 14...Qb6!  15. Qd5+ Kh8 and black 
     has terrific play.

9. ...bxc6  10. Bxc6 Ba6  11. Bxa8
----------------------------------      
Or 11. Qxd5 Bxf1  12. Qxe5 Bb5!  13. Nc3 Bxc6  14. Qxc6 Bd4. 

11. ...Bxf1  12. Kxf1 ...
-------------------------
This unbooked move looks reasonable, though it is passive.  Losing 
outright is 12. Qxd5 Bc4!.  Also 12. Be3 greatly favored black 
after 12...Bxe3  13. fxe3 Bxg2  14. Qg4 Bh3  15. Qxh3 Qg5+  16. 
Kh1 Rxa8 in Hermann-Keres (!) postal (!!) 1936.

12. ...Qh4  14. Qf3 Rxa8  15. Nc3 Nxc3  16. bxc3 Qxh2!
------------------------------------------------------
With the following cute idea:

17. Qxd5 Re8
------------
Indirectly defending the bishop.  Black has the advantage based on 
his better development, better pawns, and more secure king.

18. Bb2  Rxe5  19. Qa8+ Bf8  20. g3 Rf5
---------------------------------------
This move took some thought.  In spite of black's awkwardness on 
his back rank, it seems best to claw away at white's king.

21. Qg2 Qh6
-----------
A conservative choice would have been 21...Qxg2+ with an 
advantageous ending, but it seemed best to try to exploit white's 
trouble with his king.  As it turned out, White could not find the 
solution to the resulting problems.

22. Qe4 Qf6  23. Qe3 g5  24. Kg2 Bc5  25. Qe8+ Kg7  26. f4 gxf4
---------------------------------------------------------------
and White resigned.

Game 4.  Gerard Soricelli (Staten Island NY) - Mark Morss

1. e4 e5  2. Nf3 Nc6  3. Bc4 Nf6  4. d4 exd5  5. O-O Bc5
--------------------------------------------------------
Entering the Max Lange Attack.  A perfectly sound, equalizing move 
is 5...Nxe4, but if black wants to win, he should play the text.  

For the next several moves, the game follows the critical path of 
the Max Lange.  There are so many offshoots that it isn't possible 
to comment on them here.  The thoretically-minded should see ECO-
II (C55) or Zagorovsky's Romantic Chess Openings.

6. e5 d5  7. exf6 dxc4  8. Re1+ Be6  9. Ng5 Qd5  10. Nc3 Qf5
11. Nce4 Bf8
------------------------------------------------------------
Rubinstein's move, which is absolutely critical.  The point is 
that white is not to be allowed to establish a pawn on g7.  The 
books all recommend 11...O-O-O, but the game gets very difficult 
after 12. g4! Qe5  13. Nxe6 fxe6  14. fxg7! Rg8  15. Bh6!  Again 
there is too much material to treat here - see Soltis' Winning 
with the Guico Piano and the Max Lange Attack (a superb book), 
where the author argues that white is better.

12.  g4 ...
-----------
Not the best.  The test is 12. Nxf7 Kxf7  13. Ng5+ and now:

     I.  13...Kg8 (the best winning try) 14. g4 Qg6 with two 
     branches:

     Ia.  15. fxg7 Bd5!  16. gxh8=Q+ Kxh8 and Black has very fine 
     compensation for the exchange.

     Ib.  15. Rxe6 gxf6  16. Qf3 (16. f4 Nd8 or 16...Qg7) 
     16...Kg7  17. Bf4 (or 17. h4 h5  18. Bf4 Bd6) 17...Bd6 with 
     extremely complicated play.  Black is under pressure, but he 
     has good chances of beating off the attack and exploiting his 
     queenside pawn majority (or white's exposed kingside).   

     II.  13...Kg6 is a dubious, but possibly playable, alternative:

     IIa.  The refutation, according to ECO-II (C55, note 132), is 
     14. fxg7 (14. Nxe6 Re8) 14...Bxg7  15. Rxe6+ Bf6  16. g4 Qd5  
     17. Nh3! Kf7  18. Nf4 Qc5  19. Qf3 Rhf8  20. b4!? (20 Rxf6 is 
     unsound, and 20. Qe4! Kg8 21. Nh5 Be5!  22. f4 is unclear
     according to Zagorovsky, but I think it's better for white) 
     20...Nxb4  21. Qe4 Kg8  22. Nh5.  But this analysis, attributed 
     to Lepeshkin, overlooks that 22...Qd5 holds the fort - 
     with advantage.

     IIb.  Soltis recommends 14. Nxe6 Re8 (better than 14...gxf6)  
     15. Nf4+.  However, it is difficult to see advantage for 
     white after 15...Kf7  16. Rxe8 Kxe8  17. fxg7 Bxg7.  For 
     example, 18. Qh5+ Qxh5  19. Nxh5 Be4 (or even 19...Kf7)  20. 
     f4 Bd6.

12. ...Qxg4+! 13. Qxg4 Bxg4  14. fxg7 Bxg7  15. Nf6+ Kf8  
16. Ngxh7+ ...
--------------------------------------------------------
Favoring black is 16. Nxg4 h5.

16. ...Rxh7  17. Nxh7+ Kg8  18. Ng5 Nb4  19. Re7 Nxc2  20. Rb1...
-----------------------------------------------------------------
Black now has, at least, sufficient compensation for the exchange.  
The problem is to try to win.    

20. ...f6
---------
Trying to eliminate white's rook from the seventh rank.  I'm not 
really sure how best to handle this difficult position.

21. Ne6 ...
-----------
This freely gives up the seventh rank.  21. Ne4 is better.  Black 
should probably then play 21...Bf6, with apparent advantage.

21. ...Bxe6  22. Rxe6 Kf7  23. Re2 ...
--------------------------------------
23. Re4 may be better.

23. ...d6  24. Re4 b5  25. Bd2...
---------------------------------
The bishop is very loose here.  Possibly 25. Be3 is better.

25. ...f5  26. Rh4 ...
----------------------
Not 26. Rf4 Bh6.  

26. ...Nd5  27. Rh3 Rh8  28. Kg2 ...
------------------------------------
Not 28. Rxh8 Nf3+  29. Kg2 Nxd2  30. Rd1 Bxh8  31. Rxd2 Bxb2.  

28. ...Rxh3  29. Kxh3 Nc2  30. a3 c5  31. Kg2 b4  32. axb4 cxb4
33. Kf3 Bxb2!  34. Rxb2 c3  35. Bxc3 ...
---------------------------------------------------------------
I also considered 35. Rxb4 cxd2  36. Rb1 a5  37. Rd1 a4  38. Rxd2 
a3  39. Rxd3 a2  40. Rd1 a1=Q  41. Rxa1 Nxa1 and though the knight 
is far-flung, the ending is won for black.  White's pawns are 
unable to exchange black's last pawn.

35. ... bxc3  36. Rb7+ ...
--------------------------
A pointless move, but white is lost in any case.

36. ...Ke6  37. Rb1 Nd5+  38. Ke6 c2  39. Re1 Ne2  40. Kd2 a5
41. h4  c1=Q+  42. Rxc1 Nxc1  43. Kxc1 a4
-------------------------------------------------------------
and white resigned.                                                  











