Friday, May 12, 2017

Chesspub

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.

Brabo

Tuesday, May 9, 2017

Annotations part 2

After one of my last games my opponent polity refused my invitation for a post-mortem. He didn't consider it useful and preferred to drink a beer quietly at the bar. Engines are today much stronger than any player so why wasting time at some lousy analysis. There is definitely some truth in this as with some simple clicks you can generate automatically an analysis which is many times more accurate. Besides in part 1 I advertised a method of annotating completely based upon evaluations of the engines.

The recent Penrose Chess Institute Puzzle demonstrates clearly the dangers of blindly trusting these evaluations. Engines show a winning score for black while any experienced player easily sees it is just a draw. One and a half year ago I wrote on this blog about computers achieve autonomy but this doesn't mean that we can't play any role anymore. The doom-scenario described in the recent article at chess.com "is this the future of chess" is just ballyhoo.

1 example of some fabricated position not looking close to any normal position in standard play doesn't refute the absolute dominance of the engines. Therefore some only consider positions from serious games relevant to judge about the supremacy of the computer. Do such positions exist which we as human can access quicker and more accurate than the current engines? If yes which ones?

In my article fortresses I already covered some positions of which we can prove that the computer-evaluations are inaccurate or even plainly wrong. However humans won't do necessarily better without using any tools. Nonetheless there exist some exceptions where we are stronger than engines. 1 group of endgames, opposite bishops stands out. An experienced player can often very quickly access correctly such position. In below position the engine is not eager to exchange the queens but Robert correctly values the endgame as harmless.
[Event "Klubkampioenschap Deurne r5"] [Date "2015"] [White "Schuermans, R."] [Black "Brabo"] [Result "1/2-1/2"] [ECO "C56"] [WhiteElo "2140"] [BlackElo "2330"] [SetUp "1"] [FEN "6k1/p1pq2pp/1b6/2p5/2Pp3B/1P6/P5PP/4Q1K1 w - - 0 31"] [PlyCount "24"] 31. Qe7 {(Well judged by Robert as the endgame is an easy draw despite the pawn less. It is strange that my both top-engines still show a small advantage for black.)} Qxe7 32. Bxe7 Kf7 33. Bh4 Ke6 34. Kf2 h5 { (I also looked at g5 sacrificing a pawn, to get the king to e4 but I rightly rejected the move as it is far too risky.)} 35. Kf3 Ba5 36. Ke4 c6 37. h3 g6 38. g4 hxg4 39. hxg4 Bd2 40. Bd8 Bb4 41. Bh4 Bd2 42. Bd8 Bb4 {(The engines still do not want the draw but I had seen enough. There is nothing sensible to try.)} 1/2-1/2
In the final position the engine still gives a small edge for me but I was already for a long time convinced this is a dead draw. Another recent example is shown below. Again the engine calculates the position as better for white as black loses the c7 pawn. White still could continue instead of repeating moves but the draw is not hard to achieve of course.
[Event "Klubkampioenschap Deurne r3"] [Date "2016"] [White "Daenen, S."] [Black "Brabo"] [Result "1/2-1/2"] [ECO "C56"] [WhiteElo "1750"] [BlackElo "2314"] [SetUp "1"] [FEN "8/2pb1kp1/1bp4p/3pP3/pP1N1K1P/2P1B3/P5P1/8 b - b3 0 26"] [PlyCount "17"] 26... Bxd4 {(This forces the draw. Playing something else would only give white extra chances so I decided to stop all play.)} 27. cxd4 h5 28. Bc1 g6 29. Kg5 Bf5 30. a3 Bc8 31. Bf4 Bf5 32. Bd2 Bc8 33. Bf4 Bf5 34. Bd2 Bc8 { (White can still win the c7 pawn but black has no trouble to make a fortress. Serge did not want to prolong the game any further so agreed with the repetition of moves. Remarkably Serge barely used any time for the game while I was down to my last 5 minutes.)} 1/2-1/2
In both examples I consider it stupid to stick meticulously to my method of annotating. I exceptionally deviated from the evaluation of the engines and replaced them by my personal more accurate judgement.

In a recently played endgame I took it a step further in my annotations. Only a handful pawns are on one side of the board. The computer makes a complete mess when evaluating the played moves. Some moves are considered weak while there is nothing wrong. Others aren't annotated while there are clearly better ones. The original annotations linked to the evaluations of the engine can be found below.
[Event "Interclub Brasschaat - Deurne"] [Date "2017"] [White "Vrolijk, L."] [Black "Brabo"] [Result "1-0"] [ECO "A04"] [WhiteElo "2340"] [BlackElo "2304"] [SetUp "1"] [FEN "8/6k1/1B2n2p/3R2p1/8/6P1/r4P1P/6K1 w - - 0 34"] [PlyCount "151"] 34. Kg2 $6 {(Te5 of Le3 /-)} Ra4 35. Be3 Kg6 36. Rb5 Rc4 37. Rb6 Kf5 $6 { (Kf7 /=)} 38. Kf3 Ra4 39. Ke2 Ra2 40. Kf1 Ra4 41. Rb8 Rc4 42. Kg2 Ra4 43. Bc1 $6 {(h3 /-)} Rc4 44. Bb2 Rc6 $6 {(h5 /=)} 45. Rb7 Kg6 46. Rb4 Rd6 47. Kf3 Nc5 48. Bd4 Ne6 49. Be3 Ng7 50. Ke4 $6 {(h3 /-)} Nf5 $6 {(g4 /=)} 51. Ke5 Ra6 52. Bc5 Rc6 53. Kd5 Rf6 54. Ra4 Rf7 55. Ra8 Rf6 56. Rc8 Ra6 57. Ke5 Ng7 58. Rd8 $6 {(h3 /-)} Rc6 $6 {(g4 /=)} 59. Be3 Re6 60. Kd5 Ra6 61. Rc8 $6 {(h3 /-)} Nf5 $6 {(g4 /=)} 62. Bc5 Ng7 63. g4 Ne6 64. Bd6 Ng7 65. Bg3 Ra5 66. Ke4 $6 { (Tc5 /-)} Ra4 67. Kf3 Ra6 $6 {(h5 /=)} 68. Be5 Ne6 69. Kg2 $6 {(Tg8 /-)} Ra5 70. Bg3 Ra6 71. Rb8 $6 {(Th8 /=)} Rc6 $2 {(h5 =)} 72. Rb5 $2 {(Tg8 /-)} Nd4 73. Rd5 Ne6 74. Ra5 Rb6 $2 {(h5 =)} 75. Be5 Rc6 76. Rb5 Rc5 77. Rxc5 Nxc5 78. Kf3 h5 79. h3 Ne6 80. Bd6 Nd8 81. Ke4 hxg4 82. hxg4 Kf6 83. Kd5 Ne6 84. Be5 Kf7 85. Kd6 Nf8 86. Bd4 Ng6 87. Bb6 Nf8 $6 88. Bd8 Kg6 89. Be7 Nh7 90. Ke6 Kg7 91. Kf5 Kh6 92. f3 Kg7 93. Bxg5 Kf7 94. Be3 Nf8 95. f4 Ng6 96. Kg5 Ne7 97. Bd4 Nd5 98. f5 Ne7 99. Kf4 Nd5 100. Ke4 Ne7 101. g5 Nc6 102. g6 Kg8 103. Bb2 Ne7 104. Ba3 Nc6 105. f6 Nd8 106. Kf5 Kh8 107. f7 Nxf7 108. gxf7 Kh7 109. f8=R 1-0
After swapping all this with my personal more accurate evaluations we get a very different image of the endgame. I assume this also much better matches our intuition of such type of endgame.
[Event "Interclub Brasschaat - Deurne"] [Date "2017"] [White "Vrolijk, L."] [Black "Brabo"] [Result "1-0"] [ECO "A04"] [WhiteElo "2340"] [BlackElo "2304"] [SetUp "1"] [FEN "8/6k1/1B2n2p/3R2p1/8/6P1/r4P1P/6K1 w - - 0 34"] [PlyCount "151"] 34. Kg2 $5 {(In the remaining analysis I will very often ignore the evaluations of the engine as they do not make sense. Here Re5 and Be3 are recommended but no win can be found.)} Ra4 35. Be3 Kg6 36. Rb5 Rc4 37. Rb6 Kf5 $5 {(Kf7 is probably a bit easier.)} 38. Kf3 Ra4 $6 { (Weird but if I just look at the traditional method of annotating then this is not a mistake. However I do not agree. H5 to prevent g4 is much stronger.)} 39. Ke2 $6 {(Again the engines do not look properly at this position. I worried during the game about g4 followed up by h4. Hours of analysis did not bring a final verdict. Some lines end in a tablebase in which a win in 60 moves is shown. I do not know if that win complies to the 50 moves-rule.) } (39. g4 $1 Ke5 40. Kg3 Ra1 41. h3 Rd1 $5 42. Rb4 $5 Kf6 43. h4 $5 Rg1 $1 44. Kh2 Ra1 45. Kg2 $5 Rd1 $5 $16) 39... Ra2 40. Kf1 Ra4 41. Rb8 Rc4 42. Kg2 Ra4 43. Bc1 $5 {(White tries something new but there is no progress. H3 is also possible but seems not very promising.)} Rc4 44. Bb2 Rc6 $5 { (I decide to wait. Active play with h5 is also playable.)} 45. Rb7 Kg6 46. Rb4 Rd6 47. Kf3 Nc5 48. Bd4 Ne6 49. Be3 Ng7 50. Ke4 $5 {(The bishop returned to e3 so white tries this time something with the king. Again h3 is also possible.)} Nf5 $5 {(Playing more active with g4 is an alternative but I prefer to wait.)} 51. Ke5 Ra6 52. Bc5 Rc6 53. Kd5 Rf6 54. Ra4 Rf7 55. Ra8 Rf6 56. Rc8 Ra6 57. Ke5 Ng7 58. Rd8 $5 {(Stalling time before making decisions is the best practical chance. H3 is still recommended by the engines.)} Rc6 $5 {(I also have time. I let white make the difficult decisions. G4 is possible but I was looking at the 50 moves-rule.)} 59. Be3 Re6 60. Kd5 Ra6 61. Rc8 $5 {(H3!?)} Nf5 $5 {(G4!?)} 62. Bc5 Ng7 63. g4 {(It has been 30 moves any pawn was played or any piece was taken. White decides to put the counter back to 0 before it must be done in a less attractive situation.)} Ne6 64. Bd6 Ng7 65. Bg3 Ra5 66. Ke4 $5 {(White still wants to maintain the rooks on the board as I still have a lot of time on my clock.)} Ra4 67. Kf3 Ra6 $5 {(I keep waiting although h5 is very interesting here.)} 68. Be5 Ne6 69. Kg2 $5 {(The more active Rg8 does not bring any dividends.)} Ra5 70. Bg3 Ra6 71. Rb8 $5 {(Th8!?)} Rc6 $5 {(H5!?)} 72. Rb5 $5 {(Tg8!?)} Nd4 73. Rd5 Ne6 74. Ra5 Rb6 $5 {(H5!?)} 75. Be5 Rc6 76. Rb5 {(Finally white allows the exchange of the rooks. I do not reject this offer as white can always force this. Besides I was getting low of time on the clock so liked to resolve the position quickly.)} Rc5 77. Rxc5 Nxc5 78. Kf3 h5 79. h3 Ne6 80. Bd6 Nd8 81. Ke4 hxg4 82. hxg4 {(This is a tablebase-draw. However white still wants to try as mistakes occur easily when there is not much time left.)} Kf6 83. Kd5 Ne6 84. Be5 Kf7 85. Kd6 Nf8 86. Bd4 Ng6 87. Bb6 Nf8 $2 {(I saw the drawing-mechanism with the knight at f7 which can be achieved via h8 but I was not sure. I feared some zugzwang. I had not enough time to study the position seriously and chose at the very last moment for a different setup. That is a big mistake.)} 88. Bd8 {(I missed this move and I immediately realized that I am lost. I still continue to play as I was too disappointed to resign already.)} Kg6 89. Be7 Nh7 90. Ke6 Kg7 91. Kf5 Kh6 92. f3 Kg7 93. Bxg5 Kf7 94. Be3 Nf8 95. f4 Ng6 96. Kg5 Ne7 97. Bd4 Nd5 98. f5 Ne7 99. Kf4 Nd5 100. Ke4 Ne7 101. g5 Nc6 102. g6 Kg8 103. Bb2 (103. f6 Ne7 104. fxe7 {(This was another cute stalemate which I saw during the game.)}) 103... Ne7 104. Ba3 Nc6 105. f6 Nd8 106. Kf5 Kh8 107. f7 Nxf7 108. gxf7 Kh7 109. f8=R {(I suspect my opponent did not mind to finalize this game in such funny way.)} 1-0
This endgame shows there is also often a clear difference between understanding and actual play. We are still much more prone to blunders especially when we are running out of time.

Objectivity/ searching the truth still get absolute priority in my analysis. Engine-evaluations are used intensively but it is still good not to ignore your own chess-knowledge.

Brabo

Wednesday, May 3, 2017

Young and old

Our sport needs advertising to grow or just even to survive but rarely its biggest asset is used. Chess can be played a whole life. It keeps intriguing and enjoying us. I am addicted to the game for more than 25 years and still every day I discover new things. Chess is an endless source of exciting stories on and off the board. Countless examples are shown on this blog.

An important role in all this are the absence of strict age-categories. Any physical sport must take into account the age of the person but a game of chess can be played between a very young player and a very old experienced player. Chess acts as a bridge builder between generations. It not only is a lot of joy but it is also a very easy topic of discussion.

Besides very young can be considered literally. At the age of 3 years children are already capable to learn the rules and adopt them in a game. My son Hugo wasn't yet 4 years old when he was together with his sister Evelien introduced to chess see cheating. On the picture below we see Hugo playing a game at home just before his fourth birthday.
No, I am definitely not a parent pushing my children to play chess. My daughter quickly dropped the game while Hugo slowly learned more and more about chess. Nothing was mandatory so it took 2 years to finish step 1. I believe there is a big difference with the performance of the 3 years old Misha Osipov in a Russian TV show. A genius or just a kid? I am not the only one asking if this is appropriate.

On the other hand support of the parents is absolutely necessary for very young children. Clubs must warn the parents if this isn't happening. I even dare to state that some minimum conditions should be demanded from the parents when they subscribe their child in a chessclub. Too often I see parents dropping their 7/8 years old child in the class without any further commitments. Sorry but this sounds more like an elite day care. Children can/ should play youth-tournaments once they are at step 2. I don't recommend earlier as often the rules aren't well mastered yet and the child risks otherwise to lose a long demotivating string of games. Hugo was already 6,5 years old when he started to join me to the youth-tournaments (see basic).

Beginning of 2017 Hugo achieved step 3 so time to try his luck in standard games. However here I encountered a problem as there is no club in the neighborhood arranging a standard competition in the daytime. No club had experienced issues with players going normally to bed between 8 and 9 PM. I toyed with the idea to contact my old club de Torrewachters in Roeselare. I know they have a very nice championship on Saturdays at 2 PM. But I didn't like the prospect of each week to drive more than 300km for just 1 game. In the end KMSK offered an emergency solution. Hugo could play a couple of games in 1 of their 10 teams of the interclub. 1 day after his 8th birthday Hugo played his first official game. Coincidence or not but there were 75 years difference with his 83 years old team-mate Walter Huyck.
Both players are clearly enjoying side by side a game of chess. The huge age-difference was no obstacle. Besides Walter remembered a cute anecdote which he shared with my son. Walter played 17 years ago once against me. Playing chess at a very advanced age seems to train the memory as indeed I do have a game in my personal database which I played against Walter.
[Event "Zilveren Toren Deurne - Mechelen2"] [Date "2000"] [White "Brabo"] [Black "Huyck, W."] [Result "1-0"] [ECO "C82"] [WhiteElo "2261"] [BlackElo "1990"] [PlyCount "81"] 1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Nxe4 6. d4 b5 7. Bb3 d5 8. dxe5 Be6 9. c3 Bc5 {(It has been since my very first official game, 5 years ago that I met the open Spanish. Herman Ottevaere played at that time Be7.)} 10. Nbd2 O-O 11. Bc2 Bf5 12. Nb3 Ba7 13. Nfd4 Bd7 $146 {(I only found 1 correspondence-game with Nxd4 leading to complex play. My move is also shown by Fritz so looks interesting.)} 14. f3 Nc5 $5 { (Ng5 is answered by Be3 again with somewhat the better position for white.)} 15. Nxc5 Bxc5 16. f4 f5 $6 {(Black is not comfortable. F6 is risky because of Qh5. Qh4 avoids Qh5 but after Be3 white has also a clear advantage.)} (16... f6 $5 17. Qh5 h6 (17... g6 18. Bxg6 hxg6 19. Qxg6 Kh8 20. e6 Bxe6 21. Qh6 $1 Kg8 22. f5 Qe7 23. fxe6 $16) 18. Be3 Qe8 19. Qf3 Bxd4 20. cxd4 fxe5 21. dxe5 d4 22. Bd2 $14) 17. Be3 Nxd4 18. Bxd4 $5 {(Cxd4 is recommended by the engine but I prefer to exchange my bad bishop. Black proposes a draw with his next move which I reject.)} Qe7 $6 {(Slightly stronger must be Be7.)} 19. b4 Bxd4 20. Qxd4 Be6 {(Black hopes to build a fortress.)} 21. a4 c6 22. Ra3 Qb7 23. Bd3 Rfd8 24. Kf2 h6 25. Ra2 Kh8 26. Ke3 {(White can afford to bring his king to the center. This frees both lowest lines so the rooks can quickly switch between the wings.)} Qe7 27. g3 Rdc8 28. Rfa1 Qb7 29. Rg2 Rd8 $6 {(Rf8 or Qf7 to hamper the push g4, was better. A5 recommended by Fritz can be countered by Rga2 which shows why freeing both lines was useful.)} (29... a5 30. Rga2 axb4 31. cxb4 bxa4 32. Rxa4 Rxa4 33. Rxa4 $16 {(White has a big endgame advantage.)}) 30. g4 Qd7 $6 {(Again inaccurate as Qf7 was better to keep an eye at h5.)} 31. g5 Rg8 $5 {(Black realizes too late that white is not only playing at the queen-side. Fritz wants to play h5 but then g6 becomes a real headache.)} (31... h5 $5 32. axb5 $1 {(G6 can be answered by a5. White still has a big advantage but there is no clear win immediately.)} axb5 {(After cxb5 there is g6 and black has no counterplay.)} 33. Rga2 Rxa2 34. Rxa2 Qe8 35. Ra7 Ra8 36. Qb6 Rxa7 37. Qxa7 {(Black can not defend this any longer.)} Kg8 38. Qc7 g6 39. Kd4 Bd7 40. Kc5 Qe7 41. Qd6 Kf7 42. h4 Be8 43. Kb6 Qxd6 (43... Bd7 44. Qf6 Qxf6 45. gxf6 Ke6 46. Be2 Be8 47. Kc7 Bd7 48. f7 $18) 44. exd6 Ke6 45. Kc7 Bd7 46. Bxf5 $3 $18) 32. Qc5 $5 {(Gxh6 recommended by Fritz is probably a bit quicker but my move sets a nasty trap which is very difficult to avoid when you do not have much time.)} Qc7 $2 {(Black just waits. However black should have played here something like Rge8 although it would just prolong the game a bit.)} 33. axb5 $5 {(This wins but even stronger is gxh6 with the idea of exchanging rooks and next infiltrate with Ra7 after axb5.)} axb5 34. Rxa8 Rxa8 35. gxh6 gxh6 36. Rg6 Qf7 $6 {(Stronger was Qd7 as now also c6 drops.)} 37. Rxh6 Kg8 $6 38. Qxc6 Qa7 $6 39. Kf3 Bf7 $2 40. Qf6 Qa1 41. Rh8# 1-0
Walters playing-strength has decreased over the latest years but I don't think he really cares. Korchnoi or Strong Jan are/ were big exceptions. Jan playing our first board this season got 70 years old. Extraordinary if you know that last year he was laying for some time in the hospital for a cerebral infarct. In the past some players already died during a game due to cardiac failure see e.g. deaths at the chess olympiad. I know several (older) players whom don't play competitive chess anymore after their doctor warned them.

A lot depends of course of how a player looks at chess. Is winning still very important or can you quicker accept a loss? If I look at my son Hugo then I notice already a positive trend. It has been a while that he cried despite losing the first 3 games in the interclub. Nevertheless he is still motivated which his big smile betrayed when he won his very first game against the 55 years old Greek Nikolaos Zaimis.
[Event "Interclub Europchess - KMSK"] [Date "2017"] [White "Zaimis, N."] [Black "Hugo"] [Result "0-1"] [ECO "B13"] [PlyCount "109"] 1. e4 c6 2. Nf3 d5 3. exd5 cxd5 4. d4 Nc6 5. Bb5 Bd7 6. Nc3 Nf6 7. O-O e6 8. Re1 Bd6 9. Bg5 h6 $2 {(Too often Hugo keeps his king in the center.)} 10. Bxf6 Qxf6 11. Nxd5 Qf5 12. c4 O-O 13. Nc3 e5 14. d5 Nb4 15. Bxd7 Qxd7 16. Nxe5 {(White is of course winning with 2 pawns up.)} Qf5 {(This attacks the knight at e5 and simultaneously threatens a fork on c2.)} 17. Qe2 $4 {(Qb1 and Nf3 counter all threats but it is not simple.)} Rfe8 18. f4 Qxf4 19. Qf3 Bxe5 20. Qxf4 Bxf4 {(Now Hugo has a won position. Technically it is still very hard for him as the white center-pawns are not so easy to deal with.)} 21. Rxe8 Rxe8 22. Nb5 Nc2 23. Rf1 Bb8 24. Rc1 Nb4 25. a3 Na6 26. Rd1 Rd8 27. b4 b6 28. h3 Nc7 29. Nd4 Ne8 $4 30. Re1 $4 {(Hugo escapes as Nc6 was again winning.)} Bg3 31. Re2 Bh4 32. Nc6 Ra8 33. Nxa7 Nf6 34. Nb5 Bg5 35. Kf2 Bc1 36. Ra2 Ne4 37. Kf3 Nd2 38. Ke2 Nxc4 39. Rc2 Nxa3 {(Luck? I do not know.)} 40. Rxc1 Nxb5 41. Kd3 Rd8 42. Ke4 Nd6 43. Ke5 f6 44. Ke6 Ne4 45. Rc6 Ng5 46. Ke7 Rxd5 47. Rxb6 Rd2 48. Ke8 Rxg2 49. Rc6 Rb2 50. Rc4 Nxh3 51. Ke7 f5 52. Ke6 f4 53. Kf5 f3 54. Kg6 Kf8 { (Nice. Hugo is still attentive.)} 55. Rf4 {(White did not want to see Nxf4 so resigned. A wild but surely deserved first win in the interclub for Hugo.)} 0-1
It is still too early for Hugo to go to big international tournaments like Gent, Charleroi,...  I think step 4 is a minimum with a rating of approximately 1400. I don't want to run before walking as some other youth-players do. Some -10 players already are at step 5 or 6 while technically they are not mature at all.

A real danger is children advancing too fast and getting exhausted at some point. I also see adults with unrealistic goals. Fun should always be the most important. Only if we respect that then chess can be something unique of which young and old can enjoy at the same time.

Brabo