Friday, May 12, 2017


After the Belgian youth-championship I asked my students to analyze a couple of their games so we could discuss them together. Only 1 of them did an effort and that was just a print of an automatically generated engine-analysis. Pretty sad if you know that I work with some of the most talented youngsters in Belgium. 2 of them are reigning Flemish champions and got both 4th place in their respective age-categories of the latest Belgium youth-championship. This mentality explains of course why there is such big gap with the level of top-players in neighboring countries.

Deep analyzing of your own games is crucial to develop yourself maximally as a player. 1 of the first to propagate this was world-champion Mikhail Botvinnik and any trainer will still repeat it even today. However Mikhail Botvinnik continued by stating that we should also publish our analysis. This allows to control the quality of the analysis by the eyes of dozens of players.

This last piece of advice is nowadays doubtful. I agree with John Hartman in his article at us-chess that our current best engines are sufficient to easily detect our mistakes in the homemade analysis. On the other hand engines still don't give answers always. Feedback from other players can still be very useful. A computer only spits moves with an evaluation and can't talk to us.

However any player with some experience in publishing analysis will surely have noticed, how rarely you still get reactions of readers nowadays. The frustration in the article "getting attention on my analysed games" is very clear. I don't expect any reactions for a long time anymore on the many analysis which I publish on this blog. I write because I like to share something otherwise I would've stopped already.

Maybe the best place around to get analysis commented, is the forum of chesspub. The site was in the first place created to promote their paying services at chesspublishing but it also has its own identity. Many members visit the forum daily already for many years (I do in the meanwhile for more than a decade) and post regularly without being obliged to take a subscription. An important element in the success are the French GM Tony Kosten and a bunch of moderators, managing to keep away any trolls. Many forums die quickly due to a lack of monitoring.

I often post analysis myself. On the other hand with 750 posts I am looking quite inactive compared to the undisputed number 1. The counter of Mark Nieuweboer having posted some articles here too before has crossed the milestone of 10.000 ! Of course all of this is not always very serious. I prefer to write only when I feel a connection with the topic. Last an opening was popping up in a discussion which I covered here on this blog see king's gambit with Nf3. Buddo encountered a problem for white which he could not solve see chesspub: John Shaw King's Gambit. I spent a couple of hours analyzing the position at home and found an answer.
[Event "Koningsgambiet Fischervariant 5.g3"] [Date "2017"] [White "?"] [Black "?"] [Result "1/2-1/2"] [ECO "C34"] [PlyCount "42"] 1. e4 e5 2. f4 exf4 3. Nf3 d6 4. d4 g5 5. g3 g4 6. Nh4 Bh6 {(I found no game with this move in the databases but I did analyze it in 2014.)} 7. Nc3 {(In my analysis of 2014 I only discussed Bb5. In the last 3 years engines made considerable progress and show now different evaluations.)} Bg5 8. Qd2 { (My mainline in 2014 continued with Ng2 but this is not enough for equality. Today I consider Qd2 or Qe2 with the same idea as better.)} Bxh4 9. gxh4 Qxh4 10. Qf2 Qxf2 {(Weak engines/ computers have troubles to evaluate g3 correctly.)} ( 10... g3 11. Qxf4 g2 12. Qxh4 gxh1=Q 13. Kf2 $18 {(White has a rook less but black will have to give up the queen.)}) 11. Kxf2 Nc6 {(This is a very complex position even without the queens. Probably Ne7 is a more critical test but even after Ne7 white has enough compensation.)} 12. Nd5 Kd8 13. Bxf4 Nxd4 14. Nxc7 { (H3 and Rd1 are also interesting.)} Kxc7 15. Rd1 Nc6 {(G3 is playable here or a move earlier. In both cases white can answer simply with Kg2.)} 16. Rxd6 Kb6 17. Rd5 a6 18. Be3 Kc7 19. Bf4 Kb6 20. Be3 Kc7 21. Bf4 Kb6 1/2-1/2
Buddho used also his engine to analyze the position but could not discover this. In my article computers achieve autonomy I tell that in the ongoing world-champion final of correspondence-chess we see a drawing rate of almost 100%. However it is serious mistake to deduct that anybody can analyze like in that top-tournament. I even dare to state that making never (detectable) mistakes in analysis is something very few are capable of. Please remember my comments upon the fantastic game Navara-Wojtaszek in the article g4 in the najdorf  or more recently in the game Wojtaszek - Mamedyarov after which black contributed his loss to an error made by his helper in the preparation of the game see chess24, a site quickly gaining popularity.

So I help others but sometimes I learn something too. In January there was a brief thread about a very specific line in the Dutch Defense. I didn't want to spend much time at analyzing the line as I only got it once in a standard game on the board. That game was played more than 20 year ago, see below.
[Event "Open Geraardsbergen 1ste ronde"] [Date "1996"] [White "Hannecart, M."] [Black "Brabo"] [Result "1/2-1/2"] [ECO "A85"] [WhiteElo "1900"] [BlackElo "2290"] [PlyCount "23"] 1. d4 f5 2. c4 Nf6 3. Nc3 e6 4. Nf3 Bb4 5. g3 $5 {( This move has been played before and it is not so easy for black to get equality.)} O-O 6. Bg2 d6 7. Qb3 Bxc3 8. bxc3 Re8 9. O-O Ne4 $6 {( Black forgets to develop.)} (9... Nbd7 $5 10. Bg5 $1 c6 11. Rfd1 e5 12. Rab1 exd4 13. Nxd4 Nc5 $11 {(The best location for the knight which guarantees black acceptable play.)}) (9... b6 $5 {(This looks also playable.)} 10. Ne5 $1 Ne4 11. Nd3 $13) 10. Nd2 Nxd2 11. Bxd2 c6 12. e4 {(White proposed a draw as he feared my extra rating-points. I accepted as white is better.)} (12. e4 fxe4 13. Bxe4 Nd7 (13... b6 14. Rfe1 Ba6 15. Bg2 $14) 14. Rae1 Nf6 (14... e5 15. Bg2 Nf6 16. Bg5 h6 17. Bxf6 gxf6 18. f4 Qc7 19. Qd1 $14) 15. Bg2 $14) 1/2-1/2
The rating-difference allowed me to escape with this premature draw. I was a warned man so I did follow with the necessary interest the rest of the discussion on chesspub. MNb (Mark Nieuweboer) proposed an interesting anti-dote (5...Bxc3) which I checked at home with my engine.

Only one and a half month later I was happily surprised to meet the line on the board. Almost 20 years it didn't happen and now suddenly it does. When you talk about the devil then you see his tail. The surprise prepared by my opponent, the Belgian FM Frederic Verduyn returned as a boomerang. I wasn't able to win the game which is at least partly due to Frederic's strong defensive skills.
[Event "Interclub KBSK - Deurne"] [Date "2017"] [White "Verduyn, F."] [Black "Brabo"] [Result "1/2-1/2"] [ECO "A84"] [WhiteElo "2200"] [BlackElo "2304"] [PlyCount "84"] 1. c4 f5 2. Nc3 Nf6 3. Nf3 e6 4. d4 Bb4 5. g3 {(2 months ago I wrote at chesspub that this line is very rarely played in practice and then it happens of course. The last time I got this line in a standard game dates already from 1996! )} Bxc3 {(Marc Nieuweboer recommended this move at chesspub and I liked it after checking it quickly with my engines. In 1996 I still played 0-0 against Marc Hannecart but I had no fun with black.)} 6. bxc3 b6 {(I was not able to remember fully what my engine at home showed me so I just had to figure it out at the board.)} 7. Bg2 Bb7 8. O-O O-O 9. a4 {(White has a number of alternatives. D5 is the most important one which we looked at both during the game. Black has also after d5 a comfortable position.)} Nc6 10. Ne5 $6 {(This has been played before in practice but it is not so good. Better is Ba3 maintaining the balance.)} (10. Ba3 $1 Re8 11. Bb4 Rb8 $5 12. a5 d6 13. Qa4 $13) 10... Na5 11. Bxb7 Nxb7 12. c5 d6 $146 {(I still found one game in the database with bxc5 but my engines prefer my novelty.)} 13. cxd6 cxd6 14. Nd3 Rc8 15. Qb3 Nd5 16. Bd2 Na5 17. Qa2 Nc4 $6 {(I miss a tactical trick. Stronger are Qe8 or Qe7 with some advantage for black.)} (17... Qe8 $5 18. Nf4 Nc4 $5 19. Nxd5 exd5 20. Rae1 $1 $15) 18. Rac1 $2 {(Also white miss the trick with Nc5 permitting to get back equality.)} Qd7 $2 {(Only after releasing my piece I finally saw the trick. Much stronger was Qe8.)} 19. Rfe1 $2 { (Frederic had noticed it too but misjudged the consequences.)} (19. Nc5 $1 dxc5 20. Qxc4 cxd4 21. Qxd4 e5 22. Qd3 Qe6 23. Rfd1 f4 24. c4 Nf6 $13) 19... Nxd2 $6 {(This harsh decision was rightly condemned after the game. I wanted to avoid the trick with Nc5 and could not see how to make progress in my position. Nevertheless quiet play with Qf7 looks better although the position remains very complicated.)} 20. Qxd2 Qxa4 21. e4 fxe4 22. Rxe4 Rfe8 23. Qe1 Qd7 24. c4 Nf6 25. Re2 Re7 $6 {(I prefer to play solidly but white becomes too active now. The direct a5, recommended by the engines is better.)} 26. Nf4 Rce8 27. c5 {(Frederic has lost quite some rating as FM in the last couple of years but he still understands the game very well as this move proves.) } bxc5 28. dxc5 e5 29. c6 Qc8 30. Nd3 Rc7 31. Rec2 Nd5 32. Qe4 {(Qa5 was sharper but also Qe4 is enough compensation for the pawn.)} Qe6 33. Qc4 Qf7 34. Qb5 Rec8 35. Rd2 e4 {(Frederic asked me after the game if I missed his reply. No I did not. I just wanted to simplify the position as my time was running out. Anyway the engines do not show any serious improvements.)} 36. Ne5 dxe5 37. Rxd5 e3 38. fxe3 Qf3 39. Qb3 {(After the game we wondered if white missed something in this phase. Rxe5 seems the only chance to try something although Stockfish evaluates the resulting endgame as equal.)} (39. Rxe5 $5 h6 40. Qc4 Rf7 41. Qd5 Qf2 42. Kh1 Qb2 43. Rf1 Qxe5 44. Qxf7 $11) 39... Rf7 40. Rd7 Qf2 41. Kh1 Qf3 42. Kg1 Qf2 {(Rcf8 was still playable but I had seen enough. Anyway we already won the match so there was no real need anymore to push.)} 1/2-1/2
I would like to tell you more successful stories about chesspub but the truth is that the best days are passed. We had 183312 posts in the last 15 years. That is averagely almost 34 posts each day. However lately we see regularly days without any post. The silent periods are becoming longer and longer.

Most posts are today about chess-books, DVDs and repertoires. What do you recommend and what not? Analyzing of positions has become seldom while originally it made the backbone of the forum. I suspect that we can point again to the engines as culprit. Their answers are for most amateurs sufficient. Besides that people are less open and are more fond of privacy. People prefer small closed groups eg. at Facebook. Initially the chesspub-founder Tony did not have the intention to keep the site running for 15 years but now it seems the chesspub-forum is slowly dying. Everything ends but after 10 years it hurts.


No comments:

Post a Comment