Tuesday, December 3, 2013

How to win from a stronger player?

During the game, a player has to make judgments often based on very little information. It makes that an uncertainty is created on the quality and eventually also the result. The elo-system takes into account this uncertainty as it makes a prognosis for each game based on an extensive statistical research depending on the difference in playing-strength between the 2 players. In 2011 there was a competition to develop a rating-system which provides a more precise prognosis, see the deloitte fide chess rating challenge. A lot of teams succeeded in presenting a clearly better performing mechanism as you can notice on the leaderboard. Still an adaption of the existing elo-system didn't happen by fide. I was not involved in the internal discussions but I can imagine some good reasons: 
  • The better performing mechanisms are based on complex formulas which have to rely heavily on computers.
  • To replace the current rating-system by a new one signifies extra costs. 
  • The rating-system has as primary function to define the playing-strength of a player. To prepare a prognosis for a single game is of minor interest. It is why the current rating-system is sufficient.

The disadvantage of defining a prognosis is that they can often be intimidating for the lower rated player. If you have 200 points less than your opponent that the expectancy score is only 24%. Intimidation is not a good adviser. Often you see players playing passively against stronger players to avoid big errors while it is exactly this policy that makes them lose without a chance. As there is little mentioned in literature about maximizing your chances as underdog, I thought it would be interesting to write an article about it.

First i have to admit that as lower rated player that you can't do much if the difference is 300 points or more. The opponent is then so much stronger on every domain that each of your plans will be sabotaged long before you even thought about it. An example of such scenario can be found in my blog-article met een kanon op een mug schieten. If the difference is smaller than 200 points then there are more chances to create resistance as weaker player. The strategy mentioned e.g. on the blog of the Ukrainian grandmaster Igor Smirnov is probably the best known one. If you play against a stronger player then play as bold as possible and complicate even if it costs material. The principle is based on the earlier mentioned uncertainties which pop up during a game. If we would have a game of 100% uncertainties then we can assume that the result will be random so the prognosis will be 50%. The bigger the complexity, the more uncertainties, the more interesting for the lower rated player as his expectancy score will improve. Somebody you don't need to tell him twice to complicate, is my clubpresident and teamcaptain Robert Schuermans. In 2006 Robert made a sensational exploit by beating with black in a sharp game the Ukrainian grandmaster Stanislav Savchenko.
[Event "Le Touquet op"] [Site "Le Touquet"] [Date "2006.11.01"] [Round "7"] [White "Savchenko, Stanislav"] [Black "Schuermans, Robert"] [Result "0-1"] [ECO "D07"] [WhiteElo "2532"] [BlackElo "2239"] [PlyCount "48"] 1. Nf3 Nc6 2. d4 d5 3. c4 Bg4 4. Nc3 e6 5. cxd5 exd5 6. Bf4 Bd6 7. Bg3 Nge7 8. e3 O-O 9. Be2 Nf5 {(The complications start after which the players quickly lose the control. As chaos is the theme of the game, it would not be correct to analyze the moves objectively.)} 10. Qc2 Nh6 11. a3 Re8 12. Nb5 Bf5 13. Bd3 Bb4 14. Ke2 Be4 15. Bxe4 dxe4 16. Ne5 Nxe5 17. Bxe5 Bd6 18. Qxe4 Bxe5 19. dxe5 c6 20. Nd4 Qa5 21. f4 c5 22. Nc2 Qb5 23. Kf3 Rad8 24. Rhd1 f5 {(This must have been a hard blow for the grandmaster as most likely he assumed that the worst complications were behind.)} 0-1'/>
The game is a nice example of how a grandmaster completely lost control over the game and eventually made the first big error. Despite that I see clear rewards on this chaos-strategy, I also have some critics if this method is really the best for everybody. Exists there no risk that you as weaker player won't create complications which mainly will backfire so leading to just extra mistakes solely for yourself? In my blogarticle tactic I wrote that I don't like to take risks so I doubt playing against your own style is the most clever strategy. Therefore I don't find it redundant to look if other strategies exist which can improve your chances as the weaker player.

I remember last year that the Dutch expert Danny De Ruiter defeated in 2 weeks time, the well-known grandmasters Ivan Sokolov and Jan Timman.  This article explains how Danny spent 3 weeks to prepare different variations for his game against Sokolov and in the end was fully rewarded for it. The game-preparation is an important weapon for the lower rated player. As described in my blog-article de sterktelijst the available materials for studying your opponent increase seriously with his rating. Myself I remember 1 victory on the Bulgarian grandmaster Ventzislav Inkiov based on a successful game-preparation.
[Event "Interclub Lille EDN- Mingé Auxances"] [Date "2007"] [White "Brabo"] [Black "Inkiov, V."] [Result "1-0"] [ECO "B65"] [WhiteElo "2247"] [BlackElo "2510"] [PlyCount "165"] 1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 Nc6 6. Bg5 e6 7. Qd2 Be7 { (I only had 1 earlier game played with a6 in open Leuven. Fortunately I prepared myself in the hotelroom for my opponent and noticed that he had this opening on his repertoire.)} 8. O-O-O O-O 9. f4 Nxd4 10. Qxd4 Qa5 11. Bc4 Bd7 12. e5 dxe5 13. fxe5 Bc6 14. Bd2 Nd7 15. Nd5 Qd8 16. Nxe7 Qxe7 17. Rhe1 Nb6 18. Bf1 Rfc8 { (Until this position we both blitzed the moves. I knew that my opponent already in 1998 played this rare move but I was not able to study deeply this idea during the 1 hour in my hotelroom. )} 19. Qg4 $5 {(Kb1 is in this position more popular but my choice is maybe even a little bit stronger. )} Qc5 $5 $146 {(A novelty which Inkiov probably has improvised at the board. Still known was Bd5. )} (19... Bd5 $5 20. Kb1 Nc4 21. Bh6 g6 22. c3 $1 $146 {(Last year the young Dutch FM Roeland Pruijssers played the blunder Qg5 and was harshly punished by FM Blokhuis Jeroen with Na3. The game-continuation prevents this idea and gives white somehow the better chances on the kingside due to the big holes on the black squares.)}) 20. Qb4 $5 {(Qb4 is the safe positional approach. Probably Bd3 is slightly stronger but against a much higher rated player I do not find my choice less good.)} (20. Bd3 $5 Bb5 21. Bb4 Qc6 22. Be4 Nd5 $14) 20... Qxb4 $5 {(After long reflection played. Bb5 is maybe stronger but the prospect of having an inferior but likely still tenable endgame right after the opening, was no option for the grandmaster. )} (20... Bb5 $5 21. Qxc5 Rxc5 22. Bb4 (22. Bxb5 Rxb5 23. b3 Rc8 24. Re4 Rd5 25. c4 Nd7 26. Bc3 Rxd1 27. Kxd1 $14) 22... Rd5 23. Bxb5 $1 Rxb5 24. Bd6 Ra5 25. a3 Rc8 26. b3 Rb5 27. Re3 Rd5 28. Rdd3 $14 {(The majority of pawns on the queenside gives white in both variations the better endgame. )}) 21. Bxb4 Bd5 22. b3 a5 23. Bd2 Nd7 24. Kb2 Nc5 25. c4 Be4 $6 {(Black searches desperately for counterplay or at least complications. However re-positioning the bishop to g6 drastically weakens the queenside while black can not get enough in return.)} 26. Be3 Bg6 27. Rd6 h6 28. g4 Na6 $6 { (Black tries to re-position the knight to c6 but probably underestimates my next move. After Na6 black loses forced material. On top white keeps the initiative which means black is here already lost. That I still need almost 60 moves to actually win the game, should be linked to the fact that I am only an amateur and my opponent a grandmaster. Stronger continuations are Kf8 or a4 but also then white keeps a big advantage.)} (28... a4 $5 29. Bxc5 Rxc5 30. b4 Rc7 31. c5 h5 32. g5 Bf5 33. Be2 $16) 29. Bg2 Rc7 30. Bb6 Re7 31. Red1 Nb4 32. a3 Nc6 33. Bxc6 bxc6 34. Rxc6 Kh7 35. Rc5 a4 36. b4 {(The 2 pawns on the queenside are obviously sufficient for the win but black anyway manages to put up a stiff resistance.)} Rb7 37. Rc6 Be4 38. Rcd6 Bf3 39. R1d4 Kg6 40. Rf4 Be2 41. Be3 Rc8 42. Kc3 Rbc7 43. Bc5 Kg5 44. Rdd4 f6 45. exf6 {(I spent my remaining extra minutes to find the best continuation. Herein I succeeded but with the consequence that I had to play the rest of my moves solely on the increment of only 30 seconds. )} gxf6 46. Rde4 Bd1 47. h4 Kxh4 48. Rxf6 Rh7 49. Rf1 Bb3 50. Rxe6 Kg3 51. Rg1 Kf4 52. Rf6 Kg5 53. Rf5 Kg6 54. Re1 Rd7 55. Re6 Kg7 56. Rff6 Rd1 57. Rg6 Kh7 58. Rxh6 Kg8 59. Rhg6 Kh7 60. Rgf6 Rc1 61. Kd2 Rxc4 62. Re7 Kg8 63. Rg6 Kh8 64. Ke3 Rc3 65. Kf4 Rc4 66. Kg3 Bc2 67. Rd6 Re4 68. Rxe4 {(With only 30 seconds remaining on the clock, I do not think that somebody can blame me to liquidate to a won endgame instead of Rh6 which wins spectacular.)} (68. Rh6 $1 Kg8 69. Rg6 Kh8 70. Rxe4 Bxe4 71. Bd4 Kh7 72. Rg7 Kh6 73. Kh4 {(And mate on g5 can only be postponed with 1 more move.)}) 68... Bxe4 69. Kf4 Bc6 70. Ke5 Kg7 71. Re6 {(I remember from my endgame-sessions with Eddy Verledens that endgames with opposite bishops are always won if there are more than 2 columns between the 2 pawns. This is here the case so I did not have to be afraid of black exchanging rooks. )} Be8 72. Kd6 Kf7 73. Re7 Kf6 74. Bd4 Kg6 75. Ke5 Bc6 76. Be3 Bb5 77. Re6 Kf7 78. Bc5 Kg7 79. g5 Bc4 80. Rb6 Rd8 81. Bd4 Bf7 82. Ke4 Kh7 {(Black was playing quickly not to give me any extra time to think but here he blunders into mate although after Kf8 it is also lost e.g with g6.)} 83. Rh6 {(A little bit silly to resign one move before mate but naturally it is never nice to lose against an almost 300 points lower rated player.)} 1-0'/>
Now it is a delusion to think that it is always that easy to get an advantage out of the opening against a grandmaster. I believe my Bulgarian opponent likely never suspected that I could prepare myself sufficiently in 1 hour( time between announcing the pairings and the start of the game) on a system which I had little to no experience (I had no games in the database) with. As described in another blogarticle of Igor Smirnov most grandmasters will deviate from their standard repertoire from the moment they sense some danger. A fast playing opponent in a for him unfamiliar opening is surely a warning signal which the professional won't ignore.

An approach which I like a lot, is to find positions which maybe don't guarantee an advantage but are easily playable. Besides studying openings also psychology plays herein a role. The higher rated player will feel obliged to play for the win but the type of positions will require disproportional risks. Already in the first round of the new Belgian interclubseason I showed the merits of this approach.
[Event "Interclub Deurne - Rochade"] [Date "2013"] [White "Brabo"] [Black "Feygin, M."] [Result "1-0"] [ECO "C18"] [WhiteElo "2340"] [BlackElo "2512"] [PlyCount "109"] 1. e4 e6 2. d4 d5 3. Nc3 Bb4 4. e5 c5 5. a3 Bxc3 6. bxc3 Qa5 7. Bd2 Qa4 8. Qb1 c4 9. Ne2 {(Against Hovhanisian I chose Nf3 in 2006. 2 years ago I tried Be2. This time I chose for a bigger variation which scores pretty well for white in practice although today h4 is the most popular try.)} Nc6 10. g3 {(Nf4 is more critical and more popular but I had before the game decided to play g3 which leads to more simple play. In practice white also seems to score better with g3.)} Bd7 11. Bg2 {(Also now Nf3 is more popular but again not better performing in practice.)} O-O-O 12. O-O f5 $6 $146 {(An interesting and psychological novelty to destabilize me but objectively I find it less exact than the known f6 as it gives white extra interesting possibilities.)} 13. exf6 $6 {(I transpose to the known variation with f6 but leaving the center closed is probably strategical stronger. After h4 white keeps somehow the better prospects on the kingside although it is certainly not without risks to play on the wing where your own king is placed. )} Nxf6 14. Re1 Ne4 $146 {(A second novelty which came as a surprise. Afterwards I discovered that the move is a standard-idea in this type of positions so clearly my opponent was more familiar than I about the system. Black more or less equalizes with this move but the position remains easier playing for white. Therefore the opening can be considered a mini-success for white.)} 15. Be3 Be8 16. f3 Nf6 17. Nf4 Bf7 18. Bd2 Rhe8 19. h4 Qa5 20. Re2 e5 21. dxe5 Rxe5 22. Rxe5 Nxe5 23. Qb4 Nc6 $6 {(Black wants to release the pressure by exchanging queens but now whites bishops get more scope to play. Stronger was Qc7 and black seems able to defend if we trust the engines. Anyway it is evident that white keeps the nicer looking position.)} 24. Qxa5 Nxa5 25. Bh3 Kb8 26. Re1 Nc6 27. Ne6 Bxe6 28. Bf4 Ka8 29. Bxe6 b5 30. Kf2 Kb7 31. Bh3 d4 32. cxd4 Rxd4 33. Be3 $6 {(After the game I already indicated that I somehow could play more exact in this phase and indeed engines recommend the subtle Bf5 with a more clear advantage for white.)} ( 33. Bf5 $1 Rd8 34. g4 g6 $1 35. Re6 Nd5 36. Bg5 Rc8 $14) 33... Rd8 34. Bg2 a5 35. g4 Re8 $4 {(Black has little time and makes a serious judgement-error by relinquishing the d-file. Kc7 as well as Nd5 look playable for black.)} 36. g5 Nh5 37. Rd1 $6 {(This guarantees white a big advantage but even more powerful was Rb1.)} (37. Rb1 $1 b4 38. axb4 axb4 39. f4 Kc7 $18) 37... Kc7 38. Bh3 Ne5 $6 {(The endgame with Rd8 is of course no pleasure but maybe more persistent than the game-continuation after which I had few technical problems to solve. )} (38... Rd8 $1 39. Rxd8 Nxd8 40. f4 Kd6 41. Bg4 {(F5 immediately shall normally transpose.)} g6 42. f5 b4 43. axb4 axb4 44. Be2 $1 Kd5 45. Bd2 $16 {(This should very likely win for white but of course still requires a good technique.)}) 39. Rd5 b4 40. axb4 axb4 41. f4 Ng6 42. Rc5 Kd6 43. Rxc4 Ng3 44. Rd4 Kc7 45. Rd7 Kc6 46. Rxg7 Nh1 47. Kf3 Nxh4 48. Kg4 Rxe3 49. Kxh4 Ng3 50. Rxh7 Ne2 51. Rf7 Kd6 52. g6 Re8 53. g7 Rg8 54. Kg5 Nd4 55. Bf5 1-0'/>
My teammate Daniel Sadkowski summarized well afterwards by stating that I chose the easier playing side of the board. Loyal readers will surely still remember my blog-article green moves in which I discussed the usage of openingbooks for engines. Well if you would check the opening with a recent openingbook then you would quickly discover that I regularly didn't follow the most popular/ critical continuation but instead chose the best scoring in practice continuation. We all know that statistics have their limitations but concerning practical chances for a boardgame they are pretty useful.

Purely playing the man without taking into account your own strong points, seems to me wrong. Above examples show that we can create optimal chances by starting from our own strengths: chaos for the sharp tactical player, opening-knowledge for the player willing to spend lots of time in the preparation and study, positional play for the more positional player,... Therefore I recommend to choose a strategy against stronger players based on self-confidence and your own trumps instead of fear (to lose).

Brabo

No comments:

Post a Comment