by the 10th World Correspondence Champion Dr. Vytas (Victor) Palciauskas

GM Peter Hertel of Germany

The "Game of the Month" returns after a lengthy absence, with a fascinating battle between GM Peter Hertel of Germany and the venerable GM Erik Bang of Denmark. Peter has recently (1999) gained his GM title and has a current ICCF rating of 2652, which is the 15th highest in the ICCF rating list!

Peter Hertel was born in 1958 in Cuxhaven, a town near the North Sea, where he still lives today. Peter is a postman in part-time employment; one week of work, one week free. This he finds as an excellent situation for pursuing other interests such as Correspondence Chess. He is still a bachelor.

In 1972 he was motivated by the remarkable Spassky-Fischer match, and he has been very seriously involved with chess since that time. He began Correspondence Chess in 1977 and has achieved many outstanding results in national and international competitions. Several of these are: 1st place in the 3rd National Junior CC Championship, 2nd place in the 4th Championship (1981-84), IM title with 8/14 in the "Bertl-von-Massow-Memorial" (1988-1995), 1st place in the 23rd National CC Championship in Germany (1991-95), 1st place in the 4th European Team Championship with Germany and had the best result on board 2 with 8,5/11, tied for 2nd place in the "BdF-50-Invitational". This is also the tournament where he achieved the full GM norm with 11,5 out of 14.

His most recent achievement was the best result on board 3 with 10/11 in the preliminaries of the 13th CC Olympiads.

Peter has also been active over- the-board. His biggest success in OTB chess has been an IM tournament in his hometown in 2001 where with 7 out of 9 he achieved more points than necessary to achieve an IM norm (elo performance of 2549). Since 1995 Peter has been a FIDE master. He also plays in the German team league (Verbandsliga = 5th division in Germany) on board 1 for his club "SG Niederelbe".

The game between Erik Bang and Peter Hertel is a classic King's Indian that illustrates beautifully the value of strong points (squares). In the King's Indian the strong squares are often colour related; Black being weak on the light squares, White being weak on the back squares. One question that Peter raises in this game is "Do present computer chess programs adequately evaluate the importance of weak or strong colour related fields (squares)? His discussion focuses on the move 26….Nh5, as he played in the game, as compared to computer analysis of that position. In addition the beauty of 36….Ne7!! which takes away key squares for white's queen, should be fully appreciated.

The game analysis is by Peter Hertel. Chris Lueers provided the English translation, and I made minor editorial.

Bang, E (2660) - Hertel, P [E99]
Bertl-von-Massow Memorial Tournament, 1989
[Peter Hertel]

This CC game was played during the years 1988 - 1989 when I had no elo rating. Even though at that time I played without the help of a computer, I cannot find a mistake in my original analysis even today. In my view the outstanding characteristic of this game is white's weakness on the king's wing black squares for which black had to sacrifice several pawns. A very common event in the classical King's Indian. In my view I believe it is a beautiful and aesthetic game.

In the forum at www.chessy.de GM Arno Nickel analyzed this game on the aspect of computer-based analysis. This is the introductory text of GM Nickel: "Counter play on black squares - and what Computers can say about it." His discussion focuses on the position after 26.Nf2xg4. Deep Fritz and Junior 7 for example show little or no sympathy for the move 26...Nf6-h5 that I played in the game. Instead they prefer the "mundane" Nf6xg4 as giving the best practical chances for black to equalize the position.

GM Nickel continued. Comparing the computer analysis to the analysis of Peter Hertel leads to several conclusions:

1. PC-Programs do not understand the long-term strategical character of counterplay on color-related field complexes. This leads to systematic misjudgments of the strength of variations. Both, Deep Fritz and Junior 7, constantly regard black's position as almost hopeless (-1.00 to -2.00). Yet, going deeper into the variations in no situation a forced win for white can be found. This judgment of the computer seems to be based on purely the material advantage for white (one and later two extra pawns), which is no surprise. We know about the problems computers have with opposite color bishop endgames. In principle we can see this occurring here, in perhaps an even more complex position in the middle game. The game at hand shows that the black squares in the white king's position paralyze white's attacking plans. On the other hand white has some control or even advantage on the black squares on the queen's wing. Clearly the term "color" has no meaning for a PC program, but neither has "strategy", "plan, etc.
   
   
2. The alternative 26...Nxg4, which is the computer's favorite move, could in fact be a choice based on making life easier for black, than it is after 26...Nh5. It is the practical approach. This is my personal theory and probably not in line with Peter's thoughts. GM Hertel constantly had to search for "only possible moves" to finally get equality. It's still not clear whether 26. ... Nxg4 in fact leads to a balanced position."

1.Nf3 Nf6 2.c4 g6 3.Nc3 Bg7 4.e4 d6 5.d4 0-0 6.Be2 e5 7.0-0 Nc6 8.d5 Ne7 9.Ne1 Nd7 10.Nd3 f5
11.Bd2 Nf6 12.f3 f4 13.c5

Both players are uncompromising in following the classical plans.

13...g5 14.Rc1 Ng6 15.cxd6 cxd6 16.Nb5 Rf7 17.Qc2 Ne8 18.a4 h5 19.Nf2 Bf8 20.h3

If 20.Nxa7 Bd7 (Rc7 21.Ba5 is unclear) Black sacrifices a pawn, but keeps the important bishop for an attack on the king's wing.

20...Rg7 21.Qb3 Nh4

After 21...Nf6 the control of c7 is fatally weakened. 22.Rc2 a6 23.Nc7! Rxc7 24.Ba5± Ne8 25.Rxc7 Nxc7 26.Rc1 and black faces serious problems.

22.Rc2 a6 23.Na3 Nf6 24.Be1

The diagonal e1-h4 is very important. 24.Nc4 g4! 25.fxg4 hxg4 26.Nb6 (26.Ba5 Qe8 27.Nb6 gxh3 ended badly for white) 26...gxh3! 27.Nxa8 Bd7 28.Nb6 f3 29.g3 Rxg3+ 30.Kh1 Nh5 31.Rg1 Rxg1+ 32.Kxg1 Qf6-+ with no hope for white. Now the typical break through in this position follows.

24...g4 25.hxg4

If 25.Nxg4? hxg4 26.Bxh4 gxh3 27.Rf2 (27.Bd3 Rh7 28.Bxf6 Qxf6 29.Nc4 Kh8 with an attack) 27...h2+! 28.Kh1 what now? (28.Kxh2 Rh7 29.g3 fxg3+ 30.Kxg3 Nxe4+ 31.fxe4 Qxh4+ 32.Kg2 Qh2+ 33.Kf1 Qh1mate) 28...Nh5!! wonderful solution;

25...hxg4 26.Nxg4

Now 26.fxg4 would be answered with Nh5! The position is quite unclear with many possibilities, all of which cannot be shown here.

26...Nh5!

26...Nxg4?! The following remarks are again by Arno Nickel: "Let's discuss whether the computer-move 26...Nxg4 is a serious alternative to Hertel's move 26...Nh5. After my computer-based analysis I tend to answer this question with a "yes". This move presents problems for both sides and leads to a dynamic balance, which is really on the edge. Nevertheless, I could not make out a final proof. A large degree of uncertainty remains and it might be a matter of taste. It's remarkable that computer programs, while not understanding the strategy of pressure on a field color (and misjudging the variant 26...Nh5 totally), nevertheless discover a tough alternative way of playing this position." That's Arno Nickel's comment.

I personally think 26...Nxg4 is a doubtful move. It's too much to present all my analysis here, but here are some general possibilities:

27.fxg4 Qg5 (27...Bxg4 28.Bxg4 Rxg4 29.Qh3 with a slight advantage to white) 28.Qh3 Rh7 29.Nc4 Ng6 30.Qf3 and now Black has difficulties in finding a good continuation

a. 30...Nh4 31.Bxh4 Rxh4 32.Ne3 Qh6 33.Nf5 Bxf5 34.gxf5 (34.exf5? e4 35.Qxe4 Rh1+ 36.Kf2 Qh4+ 37.Kf3 Qg3 #) 34...Be7 35.Kf2±;

b. 30...Qh6 31.Bb4 Be7 (31...Nh4 32.Qh3) 32.Qh2 Qg7 33.Qd3 Qh8 (33...Bh4 34.Rfc1 Bg3 35.Bf3 Qh6 36.Kf1 Bxg4 37.Ke2±) 34.Bf3 (34...Bh4 35.Rfc1; 34...Rh2 35.Kf2 Qh4+ 36.Ke2 Bxg4 37.Kd1 Rd8 38.Kc1) Bxg4 35.Bxg4 Rh1+ 36.Kf2 Qh4+ 37.Ke2 Qxg4+ 38.Qf3±;

c. 30...Qh6 31.Bb4 Qh2+ 32.Kf2 (32...Qh4+ 33.g3 fxg3+ 34.Qxg3 Rf7+ 35.Kg2 Nf4+ 36.Rxf4 Qxg3+ 37.Kxg3 exf4+ 38.Kf3+-) Rh4 33.Ke1 Bxg4 (33...Rxg4 34.Qh3 Rxg2 35.Qxh2 Rxh2 36.Nb6+-) 34.Qf2 Bxe2 35.Qxe2 Qg3+ 36.Kd1 Qb3 37.Bxd6± (37.Be1; 37.Nb6);

27.Nc4!

A new move at that time. Fritz7, which did not exist at the time of Arno Nickel's comment, here finds the right moves. All other engines are more or less without a clue at what is best in this position. The game Ftacnik-Z.Polgar, 1985, continued 27.a5 Ng3 28.Bxg3 fxg3 29.Qb6 Qe7 30.Rfc1 Bxg4 31.fxg4 and was drawn in 62 moves, but black might have chances to improve.

27...Ng3 28.Bxg3 fxg3 29.Qb6! Qe7

A mistake would have been 29...Qg5? 30.Qe3 (30.Ncxe5 dxe5 31.Rxc8 Rxc8 32.Qe6+ Rf7 33.Qxc8 Qd2 34.Qc4 b5! 35.axb5 axb5 36.Qd3 Bc5+ 37.Kh1 Qxd3 38.Bxd3 Nf5!-+) 30...Qd8 31.Rfc1+-

30.Nce3 Rh7 31.Rfc1 Bxg4 32.Nxg4

32.fxg4 Qg5 (32...Qe8) 33.Nf1 is unclear

32...Ng6 33.Rc7 Qg5

34.Rxh7

Other moves are not better:

34.R1c2 Be7 35.Rc8+ (35.Rxb7 Kg7 36.Rxe7+ Nxe7 37.Bd3 Qh4 38.Kf1 Qh1+ 39.Qg1 Ng6 40.Rc7+ Kh8; 35.Qe3 Nf4 36.Rxb7 Qh4 37.Kf1 Qg5 38.Qg1 looks dangerous for white) Rxc8 36.Rxc8+ Kf7 37.Rc7 Nf4 38.Bc4 Kg6 39.Rc8 (39.Rxb7 Rh2!-+) 39...Qh4 40.Kf1 Nxg2!-+;

34.Qe3 Nf4 35.Rxh7 Kxh7 36.Rc7+ Be7 (36...Kg6) 37.Bc4 b5 38.axb5 axb5 39.Bxb5 Qh4 and Black will have the choice of moving his rook to a1 or h8.

34...Qxc1+ 35.Bf1 Kxh7 36.Qxb7+

A very difficult position, that even Fritz7 could not evaluate! Try to find the best move.

36...Ne7!!

The point of this knight move is to take away the squares a6 and b6 from White's Queen, after moving to c8. Without Nc8 it would be impossible to bring my bishop to e3. It is such subtle moves that determine the outcome of this game. Other moves would not save the game for me:

36...Be7? 37.Qxa8 Qc5+ (37...Nf4? 38.Qxa6+-; 37...Bg5 38.Qa7+ Kg8 39.a5±) 38.Kh1 a5 39.Qc6 Qa7 40.Qc1 Nf4 41.Bb5±;

36...Kh8? 37.Qxa8 Qc5+ 38.Kh1 a5 39.Qc6 Qa7 40.Bb5 Qh7+ 41.Kg1 Qa7+ 42.Kf1 Qh7 43.Qb6+-;

36...Bg7? 37.Qxa8 Nf4 38.Qxa6 Qxb2 39.Qb5 Qa1 40.a5 Bf8 (40...Bh6) 41.a6 Kg7 42.Kh1 Kg8 43.Ne3 Qd4 44.Nf5+- .

37.Qxa8 Bh6 38.Qa7

[38.Qxa6?? Be3+ 39.Kh1 (39.Nxe3 Qxe3+ 40.Kh1 Qh6+ 41.Kg1 Qh2#) 39...Bd4-+]

38...Kg6 39.Qd7

Other moves also lead to a draw:

39.a5 Nc8 Now the knight drives away the Queen from the diagonal g1-a7. 40.Qd7 (40.Qxa6 Be3+ 41.Kh1 Bd4 42.Qb5 Qg5 43.Qe8+ Kg7 44.Qd7+ Ne7 45.Nh2 Kf8 46.Bb5 Qh4 47.Qh3 Qxh3 48.gxh3 gxh2 49.Kxh2 Bxb2=) 40...Be3+ 41.Nxe3 Qxe3+ 42.Kh1 Qe1=;

39.Qxa6 Be3+ 40.Kh1 Bc5 41.b4 (41.Nxe5+? Kg7 42.Ng4 Qg5-+; 41.Qb5? Qg5-+) 41...Bxb4 42.Qc4 Qd2= two extra pawns cannot help white;

39.Qb6 leads to an equal endgame. 39...Nc8 40.Qb3 (Be3+ had to be prevented) 40...a5! 41.Qc3 (41.Qc4?! Be3+ 42.Kh1 Qxc4 43.Bxc4 Bc5 and although white has two extra pawns, he cannot win this position because his king and knight cannot join the game.) 41...Qxc3 42.bxc3 Bd2 43.c4 (43.Ba6 Nb6 44.Bb5 Bxc3 and I do not believe white can win.) 43...Bc3 (43…Bf4!?) And how can white play for a win?

39...Be3+

and a draw was agreed. My opponent commented: "Unbelievable!"
.

White cannot escape the threats on the black squares: 40.Nxe3 Qxe3+ 41.Kh1 Kf7 (41...Qh6+ 42.Qh3 a5 43.Qxh6+ Kxh6 is probably also sufficient, but Kf7 is more beautiful.) 42.Qe6+ (42.Qxd6 Qc1 43.Qe6+ Ke8 44.Qxa6 Qg5 45.Bb5+ Kf7 46.a5 Qc1+ 47.Bf1 Ng6=) 42...Kf8 43.Qxd6 Qc1 44.Qxa6 Qg5 Black continues to set threats. 45.Qe6 Qc1 46.Qf6+ Kg8 47.Qa6 Kf8 48.a5 Qg5 49.Qe6 Qc1 50.Qf6+ Ke8 51.Qa6 Qg5 52.Bb5+ Kf8 53.b4 Qh4+ 54.Kg1 Qh2+ 55.Kf1 Qh1+ 56.Ke2 Qxg2+ 57.Kd1 Qxf3+ 58.Kc2=]

½-½