Last spring, I decided to try playing some correspondence chess. As has often been written, it is a great way to delve into new openings and study positions to great depth. I had tried it by mail a few times in the 1980s, but found it too slow going. However, with the rise of internet chess, it is very convenient to play on a server, which tracks the time. So I entered the ICC's Spring 2006 Correspondence Championship and was placed in a 6-player preliminary group.

Chesswise, I had a very good result, winning my section with a perfect 5-0 score to move on to the semi-finals. This got me a provisional 2165 correspondence rating on the server. That doesn't sound too high, but there aren't a whole lot of correspondence players and the highest established rating is only 2226.

As far as the correspondence experience went, that was more of a mixed bag. My primary complaint about the tournament was the enforcement of the time control. Correspondence events on ICC have a time control of 10 moves in 30 days, 20 in 60, etc. However, unlike every other form of chess on ICC there is no autoflag feature for correspondence. Instead, if a player oversteps, you first must complain to the TD, who issues a warning to the offending player and sets a new time control. If the player oversteps that, then he finally loses. On top of all that, a player is allowed an additional 30 days vacation time. This makes it very hard to lose on time, yet it still happened in 3 of my games, and really it should have been 4.

Some of these games are harder to annotate since I was digging deeply into variations a lot of time, most of which didn't see the surface. Please forgive me if it gets to dense without enough words at some points, especially the final game.

My first game to finish was with White versus "El-Tornado". I had a comfortable position out of the opening after **14. Bf5**

In a normal game you would say it ended abruptly after **14...Nfe4? 15. Bxe7 Qxe7 16. Nxe6 [1:0]** He's losing material after 16... fxe6 17. Nxd5 exd5 18. Bxc8 Rxc8 19. b4, but I never got another reply for him and thus had to wait a couple of months before the result was official.

It looked like my game with Black against "flupo" was going to take a similar turn. After
**13...0-0** I didn't hear from him for a long time.

When the 60 day limit for 20-moves passed, I put in a complaint. I heard back from the TD that my opponent was on vacation for 2 weeks. After a further 2 weeks, I submitted another complaint. This time I heard nothing. When a further month had passed, I complained again and got admonished by the TD not to submit additional complaints until the first one had been resolved. That was a bit frustrating. Either warn him or tell me why he is being given so much time. I didn't know what to do when finally after over 2 months, he sent **14. Qe2**. Over 120 days used for 14 moves, with no warning! We played a few more moves before the long-awaited warning came, and then he disappeared again before reaching move 20 and was eventually forfeited.

I had more interesting games with my other 3 opponents. With White against "Expert-Level", he played a novelty in the opening which I thought was close to equalizing for him until he didn't play a zwitchenzug and allowed his kingside to become crippled. He overstepped the 20 move time limit and was quickly warned by the TD. I thought the game was still interesting when he overstepped again before ** 24...Bg5**

Here, my original impulse was 25. Qg3 threatening both Nf6+ and Ne7+, but I couldn't seem to crack 25... f4. For example, 26. Nf6+ Kg7 27. Nh5+ (27. Qxg5+ hxg5 28. Nxd7 Rxc4 29. Rd1 Rc6) 27... Kg8 28. Qc3 (28. Qd3 Qg4) 28... f6 29. Qd4 Qf7 and Black is better. So finally I settled on the imbalanced ending after
**25. Ncxb6** and my time claim was accepted ** [1:0]**
After the further 25...axb6 26. Nxb6 Qxd6 27. Qxd6 Nxd6 28. Nxc8 Nxc8 I think the combination of the connected passed pawns with Black's inability to generate any kingside counterplay should end the game in White's favor.

My game with Black against "torrent" took an interesting turn
after **27. Ra2 **

I think the two bishops should give Black the edge. Probably best is 27...Bc3 which gets out of the way of both the e5 push and attempts to trade the bishop with Bg5. It also stops Nd2 from getting active immediately. Instead, I decided **27...Nb7** trying to do something with the awkwardly placed knight, but this gave him some interesting sacrificial possibilities.
**28. Rxa8** Black is slightly better in the ending after 28. e5 dxe5 29. Bxe5 (29. Nxe5 Rxa2 30. Qxa2 Bxe5 31. Rxe5 (31. Bxe5 f6) 31... Rxe5 32. Bxe5 f6 33. Bf4 Qxd5) 29... Rxa2 30. Qxa2 Bxe5 31. Rxe5 Rxe5 32. Nxe5 h6 33. Nc6 Bxc6 34. dxc6 Na5 35. Qa4 Qxd2 36. Qxa5 Qd1+ 37. Kh2 Qd6+ 38. g3 Qxc6
**28... Qxa8 29. Bxd6!?** Sacrificing a piece for a bunch of pawns. 29. Nc4 is more solid. I also analyzed 29. e5 dxe5 30. Bxe5? Bxe5 31. Nxe5 Rxe5 32. Rxe5 Qa1+ and 30. Nxe5? Rxe5 31. Rxe5 Qa1+ 32. Qb1 Bxe5, but 30. Ne4!? gives some attacking possibilities.
**29... Nxd6 30. Qxc5**
Not 30. e5? Bxe5 31. Nxe5 Rxe5 32. Rxe5 Qa1+
**30... Qa6 31. e5?** This fails to the same trick as before. Instead, I expected 31. Qxb4 when the three pawns should give some compensation. After 31... Re8 32. Kh2 I was still debating between 32...Be2 and 32... Be7
**31... Bxe5** and I consolidated the extra piece.

My final game to finish was with Black against "Morad-Magdy". I really enjoyed this game as I dug very deeply into it and discovered many beautiful mating patterns. I initiated an attack on his king with ** 19...h4 **

** 20. Nc4** He had overstepped sometime before here, but he wasn't warned until about a month later. On 20. g4 I had prepared a sacrifice 20...Bxg4 21. hxg4 Nxg4+ 22. Kg1 h3 23. Bh1 (23. Bf1 h2+ 24. Kg2 Nxf2 25. Kxh2 Nxd1 26. Rxd1 Be5+ 27. Kg2 Qd7; 23. Bxe4 Nxe4 24. Nxe4 Qxe4 25. Qxe4 Rxe4 26. f3 h2+) 23... Qc7 24. Nf1 (24. Nxe4 Qh2+ 25. Kf1 Qxh1+ 26. Ke2 Qxe4 27. Qxe4 Nxe4 28. f3 Ng3+) 24... Be5 25. f4 (25. Bxe4 Nxe4) 25... exf3 26. Nxf3 (26. Bxf3 Nxe3 27. Nxe3 Bxd4) 26... Bxb2 27. Qxb2 Nxf3+ 28. Bxf3 Nxe3 29. Rd6 Nf5 with 3 pawns for the piece and the White king completely exposed.
**20... hxg3+ 21. fxg3 Nh5 22. Nd6** One of the keys to Black's attack is a variation like 22. Ne2 Nf3+ 23. Kh1 (23. Bxf3 exf3 24. Nf4 Nxf4) 23... Qg5 24. Rxd7 Qxg3! threatening mate on h2 and giving mate with two knights after 25. Nxg3 Nxg3#. Black also wins on 25. Bxf3 Qxh3+ picking up Rd7.
**22... Be5 23. Ne2**
23. Nxe4 loses to 23...Nxe4 24. Bxe4 Bxg3+ and Black has full compensation for the exchange after 23. Nxe8 Bxg3+ 24. Kh1 Rxe8
**23... Nf3+ 24. Kh1** The pawn is too strong after 24. Bxf3 exf3
**24... Bxb2** Going for the Qxg3 idea, but White has a resource, so 24...Nxg3+ may be best
** 25. Qxb2 Qg5 26. Nxe8?**
26. Bxf3 exf3 27. Nxe8 Rxe8 28. Rxd7 fxe2 is again good for Black
and 26. Nxe4 Rxe4 27. Bxf3 (27. Rxd7 Qxg3) 27... Rxe3 28. Bxh5 Bxh3 gives Black a nice edge
but White seems to hold with 26. Qh8+! Kxh8 27. Nxf7+ Kg7 28. Nxg5 Nf6
**26... Rxe8**
Black can't rush in yet because of 26... Qxg3? 27. Nf6+ (or the more spectacular 27. Qg7+) 27...Kf8 28. Nxd7+ and wins. After the text I think White is busted
**27. g4** After this the Black pieces swarm in, but I didn't find anything better for him, here's a sampling of the lines I analyzed
A) 27. Bxf3 exf3 28. Rxd7 (28. Nf4 Nxg3+ 29. Kh2 (29. Kg1 Ne2+ 30. Kh1 (30. Kh2 Qg3+ 31. Kh1 Bxh3 32. Rg1 Bg2+ 33. Nxg2 (33. Rxg2 fxg2+ 34. Nxg2 Qh3#) 33... Qh3#; 30. Kf2 Qg3+ 31. Kf1 Qg1#; 30. Kf1 Qg1#) 30... Bxh3 31. Rg1 Qh4… Bg2#) 29... Rxe3 30. Rxd7 (30. Qd4 Re2+ 31. Nxe2 (31. Kg1 Nf5+) 31... Nf1+ 32. Rxf1 Qg2#; 30. Rg1 Qxf4) 30... Re2+ 31. Nxe2 (31. Qxe2 Nxe2; 31. Kg1 Rxb2) 31... Nf1+ 32. Rxf1 Qg2#; 28. h4 Qg4; 28. Ng1 Nxg3+ 29. Kh2 Ne2 30. Nxf3 Qg3+ 31. Kh1 Qxf3+ 32. Kh2 Qg3+ 33. Kh1 Qxh3#) 28... fxe2 29. g4 Qxe3;
B) 27. Rf1 Qxe3 with the idea Qxe2 (27... Qxg3? 28. Rxf3) 28. Bxf3 (28. Rab1 Qxe2 29. Bxf3 Nxg3+ 30. Kg1 Qxb2 31. Rxb2 Nxf1) 28... exf3;
C) 27. Bf1 Qxe3… Nxg3+ 28. Kg2 (28. Rxd7 Nxg3+ 29. Kg2 Nh4+ 30. Kh2 Qf2+ 31. Bg2 Qxg2#) was a variation that took me awhile to crack, but I eventually found the nice king hunt 28... Nxg3 with the idea Nh4+ 29. Kxg3 (29. Nxg3 Qg1#; 29. Qf6 Nxe2 threatening Qg1# 30. Bxe2 Qxe2+ 31. Kg3 Qh2#) 29... Ne1+ 30. Kh4 (30. Kh2 Qf2+ 31. Kh1 Nf3 32. Bg2 Bxh3 33. Rg1 Bxg2+ 34. Rxg2 Qh4+ 35. Rh2 Qxh2#) 30... Qf2+ 31. Kg5 Nf3+ 32. Kf4 Nd4+ 33. Kg5 Re5+ 34. Kh6 Qh4#;
D) 27. Rab1 Qxg3 28. Bxf3 Qxh3+ 29. Kg1 exf3 30. Nd4 (30. Nf4 Nxf4 31. exf4 (31. Qh2 Ne2+ 32. Kh1 Ng3+ 33. Kg1 f2+ 34. Qxf2 Qh1#) 31... Re2) 30... Qg3+ (30... Qg4+ 31. Kf1 (31. Kf2 Qg2+ 32. Ke1 Rxe3+; 31. Kh2 Qg3+ 32. Kh1 f2; 31. Kh1 Ng3+ 32. Kh2 Qh4+ 33. Kg1 Qh1+ 34. Kf2 Ne4#) 31... Qg3) 31. Kh1 f2;
E) 27. Rxd7 Qxg3 28. Bxf3 Qxh3+ 29. Kg1 exf3 30. Nd4 (30. Nf4 Qxd7 31. Nxh5 Qg4+) 30... Qxd7
**27... Bxg4 28. Rf1** Another slew of variations many leading to mate
28. hxg4 Qh4+ 29. Bh3 Qxh3#;
28. Qc1 Bxh3 29. Bxh3 Ng3+ 30. Nxg3 Qxg3;
28. Rg1 Qh4;
28. Bxf3 Bxf3+ 29. Kh2 Qg2#;
28. Bf1 Qh4 29. Ng1 Nxg1 30. Kxg1 Bxd1 31. Rxd1 Qg3+ 32. Bg2 Qxe3+ 33. Kh2;
28. Nf4 Ng3#;
28. Rac1 Qh4 29. Rc2 Bxh3 30. Bxf3 exf3 31. Nf4 Nxf4 32. exf4 Bg2+ 33. Kg1 Qh1+ 34. Kf2 Qxd1;
28. Qc3 Bxh3 (28... Qh4) 29. Bxh3 (29. Rg1 Bxg2+ 30. Rxg2 Qh4+ 31. Rh2 Qxh2#; 29. Bxf3 exf3; 29. Nf4 Ng3#) 29... Ng3+ 30. Nxg3 (30. Kg2 Nxe2+) 30... Qxg3]
**28... Qh4 29. Bxf3**
29. Rxf3 exf3 30. Kg1 (30. Qd4 fxg2+) 30... fxe2;
29. Rab1 Bxh3 30. Rxf3 (30. Bxf3 Bxf1+ 31. Kg1 exf3) 30... exf3 31. Bxf3 Bg4+ 32. Kg2 Qh3+;
29. Nf4 Ng3#]
**29... Qxh3+ 30. Kg1 ** He overstepped the time limit again here, but the game finished before my claim was ruled on**30... Bxf3 31. Rxf3**
31. Nd4 (or Nc3 or Nc1) 31... Qg4+ (better than winning the queen with 31... Qh1+ 32. Kf2 Qg2+ 33. Ke1 Qxb2) 32. Kf2 (32. Kh2 Qg3#) 32... Qg3#;
31. Rf2 Qh1#;
31. Nf4 Qh1+ (31... Nxf4 32. Qh2 (32. exf4 Qh1+ 33. Kf2 Qg2+ 34. Ke3 Qxb2; 32. Rab1 Ne2+) 32... Ne2+) 32. Kf2 Qh2+ 33. Ke1 Qxb2;
31. Kf2 Qh4+ 32. Kg1 (32. Ng3 Qh2+) 32... Ng3 (32... Re5 33. Qxe5 (33. Rxf3 exf3 34. Qxe5 Qg4+ 35. Ng3) 33... Qh1+ 34. Kf2 Qg2+ 35. Ke1 Qxe2#) 33. Rxf3 (33. Rfe1 Qh1+ 34. Kf2 Qh2#; 33. Rfd1 Qh1+ 34. Kf2 Qg2+ 35. Ke1 Qf1+ 36. Kd2 Qxe2+ 37. Kc3 Qxe3+) 33... Qh1+ 34. Kf2 Qh2+ (34... Qxf3+ 35. Ke1 (35. Kg1 Nxe2+ 36. Kh2 Qg3+ 37. Kh1 Qh3#) 35... Qf1+ 36. Kd2 Qxe2+ 37. Kc3) 35. Ke1 exf3]
**31... exf3 32. Nd4**
[32. Nc3 Rxe3 33. Rd1 (33. Qh2 f2+ 34. Kh1 Re1+; 33. Qf2 Rxc3) 33... f2+ 34. Qxf2 (34. Kxf2 Rf3+ 35. Kg1 Rg3+ 36. Kf2 Rg2+ 37. Ke1 Rxb2) 34... Rg3+;
32. Nc1 Rxe3 (32... Ng3 33. Nd3 (33. Qf2 Qh1#; 33. Qh2 Qf1#; 33. e4 Qh1+ 34. Kf2 Qg2+) 33... Qh1+ 34. Kf2 Ne4#) 33. Qf2;
32. Nf4 Nxf4 33. exf4 Re2]
**32... Rxe3 33. Rd1**
33. Qh2 f2+ 34. Kxf2 (34. Qxf2 Rg3+ 35. Qg2 Qxg2#; 34. Kh1 Re1+ 35. Rxe1 fxe1=Q#) 34... Qxh2+;
33. Rf1 Re2;
33. Qf2 Qg4+ 34. Kh2 Qxd4 35. Re1 Qf4+
**33... Re2 [0:1]**
I'll apologize once more for the jungle of variations. I actually cut out some of
what I analyzed in attempt to make it smoother, but it is still quite dense.
I guess this is a case where the readers can analyze a position themselves
and then compare their findings to my annotations.

Archives