Friday, June 30, 2017

2017 Summer Party Round 3 Anton Taylor, 2026 - CM Billy Woodward, 1990 Sicilian: Moscow Variation (B51)

2017 Summer Party Round 3
Anton Taylor, 2026 - CM Billy Woodward, 1990
Sicilian: Moscow Variation (B51)

1. e4 c5 2. Nf3 d6 3. Bb5+ Nc6 4. Bxc6+ This trade is not usually taken until the capture is forced but here the bishop does at least a bit of structural damage on its way out. 4. ... bxc6 5. O-O Nf6? This is an innaccuracy that leads to a slight advantage for white. black should play e5 here to keep his center strong. 6. e5 White wins 70% of the games in the database after this move order. (34 games in chessbase) 6. ... Ng4 There are no white losses after this move. Nd5 is far more common. 7. exd6 Qxd6 8. Nc3 Here we come to the end of the database games. the two moves in the collection are g3 and d3. Neither of those moves further White's overall goals in my opinion to the text. Black's pawn weaknesses are chronic and so White needs to quickly develop to press his advantage. It shouldn't actually matter the move orders. The advantage is pretty wide in my mind. 8. ... Bf5 9. d3 g6? This move does not address the black positional problems but it is hard to recommend an alternative. White is winning. 10. h3 Re1 and Bg5 are great alternatives that all follow the same theme. Using white pieces to control e-file squares. 10. ... Nf6 11. Be3 this move is inaccurate but still uyseful development. I missed the e-file theme with the moves Re1 and Bg5 thrown in and instead opt for a plan to try and target the double pawns. 11. ... Bg7 12. Nd2 Nd7? Nd5 amazingly hold things together because of the active bishop on g7 but that's not easy to see from here. 13. Nc4 Qc7 14. Na4 ? Qc1! is spotted by komodo fairly quickly but it was never even one of my candidates. The plans of that move are straightforward. harass the queen on c7 by Bf4 and trade away the stronger bishop on g7 by Bh6. Also, the small sidestep guards the b-pawn to free up the c3 knight for useful action (clearing the pin). 14. ... Be6 Nb6 more or less forces the trade of the c5 pawn for white's on b2. Which would be a major victory for black. All white's play has been trying to win the weak pawns. If black can liquidate the weaknesses what is white's plan afterwards? He will have to take extra time to find his way to some new plan. 15. Nxc5 Bxc4 16. dxc4 Nxc5 17. Bxc5 Bxb2 18. Rb1 Bf6 19. Qf3 O-O 20. Rfd1 Qa5 21. Qa3 up to here the game is drawish. Black will not be satisfied with the queen trade falling into the draw and so concedes a small advantage by mistake. 21. ... Qc7? 22. Rb3 Rfd8 23. Rbd3 Rxd3 24. cxd3 a5?? The ending was still drawish until this little gem of a time-waster. Black assumes the e-pawn is untouchable. He is wrong. 25. Re1 h5 26. Bxe7 Re8 27. Bb4 Ra8 28. Bc3 Qf4 29. Bxf6 Qxf6 30. Re4 This is the last move I recorded because the time trouble was real and I mange to throw things away even after this winning position. The text move Re4 is an inaccuracy. Moving the rook off of my back rank gives black counter chances that weren't necessary. Better is the simple g3 removing all mating nets with rook+queen or the even more active Qc5 which I would have probably played with more time to think. 0-1 I'm not altogether dissatisfied with the moves I made in this game but the real issue I have is with the finishing technique. By the text I should have been able to close this out with little difficulty but I failed to convert a nice ending. Winning the won game is a big challenge for all chessplayers.

Thursday, June 29, 2017

2017 Summer Party Round 2 CM Jerry Baker, 2067 - Anton Taylor, 2026 Nimzowitsch-Larsen: Indian (A01)

2017 Summer Party Round 2 
CM Jerry Baker, 2067 - Anton Taylor, 2026 
Nimzowitsch-Larsen: Indian (A01)

1. b3 Nf6 2. Bb2 e6 3. g3 Jerry is always trying to shake things up a bit. This is probably objectively better for black (it is normally bad to fianchetto both bishops. It creates disharmony in your own position rather than controlling the center like you might expect. 3. ... d5 4. Bg2 c5 5. Nf3 Nc6 6. O-O Be7 7. d4 cxd4 8. Nxd4 O-O 9. Nd2 Nxd4 10. Bxd4 b6 11. c4 Bb7 12. Rc1 Qd6? 13. Bb2?? 13. Bxf6 Bxf6 14. cxd5 Bxd5 15. e4 Bb7 16. e5! wins immediately but both me and my opponent missed this continuation in the game. 13. ... Rfd8 14. Qc2 Rac8 15. Qb1 Ba6 16. Rfd1 Qb8 17. Bf3 Bc5? Again there is this theme of Bxf6 creating a win but it is hard to trade away a good bishop and my opponent again misses the forcing moves that clearly win in the end. 18. Qa1 d4 When I played this move I felt it was nearly winning. Somehow it is actually a very bad move and requires some investigation. Essentially Nf1 followed by e3 leads to black getting either a handful of holes n his position or going a pawn down ... both losing choices. 19. Ne4 Nxe4 20. Bxe4 e5 21. Qb1? This throiws away the win. Correct is e3! leading to a dominant position in the center for white. I saw none of that during the game. 21. ... g6 22. Qd3 Bb7 23. g4 Qa8 24. f3 Be7 25. Rf1 Bg5 26. Rcd1 Bxe4 27. fxe4 Bf4? This makes the win harder but isn't losing. Better is 27... Be3+ 28. Kh1 b5 29. cxb5? Rd7 30. Rf3 Rdc7 0-1 28. h4 f6 29. Rxf4 exf4 30. Kg2 Rd7 31. Kf3 g5? worrying about the pawn is foolish. White taking the pawn as black piles up on the e4 pawn is suicidal and mate is sure to follow closely. 32. hxg5 fxg5 0-1 The rest of the game score is lost in blitzing low on time. However, black is fairly far ahead and should win from here.


Wednesday, June 28, 2017

2017 Summer Party Round 1 Anton Taylor, 2026 - Andrew Orr, 1693 "Marshall Gambit" (C70)

2017 Summer Party Round 1 
Anton Taylor, 2026 - Andrew Orr, 1693 
"Marshall Gambit" (C70)

1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Be7 6. Re1 b5 7. Bb3 O-O 8. c3 d5 9. exd5 Nxd5 10. Nxe5 Nxe5 11. Rxe5 c6 12. Bxd5 cxd5 13. d4 Bd6 14. Re1 Qh4 15. g3 Qh3 16. Qf3 Be6 17. Be3 Rad8 18. Nd2  I have quite become a bit of a specialist in the white side of the marshall. After encountering it a few times in blitz this year and twice in tournaments I have studied even grandmaster games in the variation (specifically MVL playing as white and Aronian as black). 18. ... Rfe8 This move is a bit useful but the mainline that leads to the "GM draw" is 18. ... Bg4. 19. Qg2 This is the standard idea for white after Bg4 so there is no reason not to play it here. The silicon monsters like Bf4 and Bg5 but that seems a bit over-ambitious. After all, white only needs to stabilize because he is a pawn ahead. 19. ... Qh5 20. f4? I can see a master level commentator saying "that's a horrible move. who plays like this?" and objectively that is probably correct. The bishop on e3 is terrible and made so much worse by this move. However, in my experience if black can get a pawn on f4 in most situations even if it is a legitimate sacrifice black gets huge compensation. consolidating f4 in that case is paramount to me, Also, since we are a pawn ahead we can actually give a pawn back to improve the bishop if it becomes necessary to free the "big pawn". 20. ... Bh3 21. Qh1? Qf3 or Qf2 are much better. It doesn't matter. 21. ... Re6 22. Bf2?? 22. ... Re2! would win on the spot. Bf2 is pretty bad but it is made much worse in most variation by the queen trapped on h1. 22. ... Rg6? see the previous note. 23. Be3 The bishop has to go right back to the old square. white has obviously done something wrong. 23. ... Qf5?? Black is a little better with moves like Re8 but white is starting to unravel as he is allowed. 24. Nf3 Bxf4 25. Bxf4 I did not see that Nh4 worked because I thought that Bxg6. what I missed is this variation ... 25. Nh4 Bxg3 26. Nxf5 Bxe1+ 27. Ng3!  the text is still equal25. ... Qxf4 26. Kf2? Yet another chess sin. Re2 is the drawing move and an "only" move. 26. ... Rf6?? 27. gxf4 White gives up as everything loses. 1-0 

It seems a bit of poor taste to call myself a "specialist" and then show you a game in which I begin to go wrong immediately after the book moves end. I recant of that indulgence in phrasing. However, as with my previous games in this line this game will go into my mental data bank and serve to make me better in the line as white. 

Wednesday, June 14, 2017

2017 Kentucky Open Round 5 Andy Porter, 2106 - Anton Taylor, 2030 English: Anglo-Indian (A15)

2017 Kentucky Open Round 5
Andy Porter, 2106 - Anton Taylor, 2030
English: Anglo-Indian (A15)

1. c4 Nf6 2. g3 e6 3. Bg2 d5 4. d4 c6 5. Nf3 Nbd7 6. Qc2 Be7 7. O-O O-O 8. Rd1 Even at the GM level this move is fairly common. It makes only a little sense to me. It is just a way to clear the f1 square of the rook. That's pretty much it. There is no hope for this rook to use the d-file. 8. ... Ne4 9. Nfd2? A strange looking move that doesn't really help white extricate the black knight. In fact allowing himself to play f3 makes the knight's journey even more frutiful than it might otherwise be. f3 would be a permanent weakness in the white camp. 9. ... f5 10. f3? A blunder as stated but this was the idea of moving the knight from f3 on the previous move. 10. ... Nef6 Just going back. No need to make the knight awkward by Nd6 even if it is better according to Komodo. Why complicate matters when you're positionally better? 11. c5? This move makes it even easier for black to control d5 and gives black targets on d4 and c5 so his better development gives him something to play for now. 11. ... b6 12. Nb3 bxc5 13. Nxc5?? This further concession gives black a classical center. Black is clearly winning. 13. ... Nxc5 14. dxc5 e5 15. b4 Qc7 There is no need to get complicated just guard everything and white should eventuially collapse. 16. Nd2 a5 17. bxa5 Rxa5 18. Nb3 Ra7 19. e3 Be6 20. Bb2 Nd7 21. Bf1 f4? At the time I felt this move was a lever peeling open the white king's hiding place but  realistically this opens lines for the White bishop pair that needed not be opened. The simple Rfa8 or the powerful (if a bit complicated) Rb8 are much better alternatives. 22. gxf4? exf4 opening the file for the rook and not compromising white's king position is better. 22. ...  exf4 23. e4 dxe4 I felt I was completely winning here and Komodo agrees. However I missed an elementary tactic in a straightforward variation as you will see. 24. Qxe4 Bxb3?? The losing continuation. Nxc5 is really the only move and promises better than equality after something like 24... Nxc5 25. Nxc5 Bxc5+ 26. Kh1 Qf7 27. Qxc6 Be3 25. Qxe7! When I calculated this after dxe4 I thought 25. ... Bxd1 netted the exchange. Unfortunately, with the position on the board I discover to my horror that the bishop capture is met by Qxg7 checkmate! I was mortified. I had just voluntarily given up a piece in a completely winning position. 25. ... Rf7 26. Qe8+ Rf8 27. Qxd7 Qa5 28. Qd2 Bxd1 29. Qxa5 Rxa5 30. Bc4+ Kh8 31. Rxd1 Rxc5 32. Ba3 Rg5+ 33. Kf2 c5 34. Bb2 h5 35. Bc3 Ra8 36. h4 Rg3 37. Be5 Ra3 38. Rd8+ 1-0


Tuesday, June 13, 2017

2017 Kentucky Open Round 4 NM John Marcsik, 2149 - Anton Taylor, 2030 Caro-Kann: Exchange (B13)

2017 Kentucky Open Round 4
NM John Marcsik, 2149 - Anton Taylor, 2030
Caro-Kann: Exchange (B13)

1. e4 c6 2. d4 d5 3. exd5 cxd5 4. Nf3 Nc6 5. Bd3 A lackluster move but solid. I have no problems with it but the sharpest lines (i.e. those experimented with by the 2700 club) are the 5.Ne5 lines. 5. ... Nf6 Bg4 is more common. 6. c3 e6 7. O-O Be7 8. Nbd2 O-O 9. Ne5 Nxe5 10. dxe5 Nd7 11. f4 f6?? This direct method of handling the threat of f5 is a mistake. It just creates more black weaknesses. Qb6+ and Nc5 are better options. I especially like Qb6+ because it takes over the e3 square which makes white's bishop on c1 seem rather silly and makes its development awkward. 12. Qh5 f5 White has chosen an overly-aggressive continuation and things are back in equilibrium. Black then must be better as the White pieces need to reorganize and Black gains a tempo with his next move. 13. Nf3 Nc5 14. Bc2 Bd7 15. Nd4 Ne4 16. Be3 Bc5 17. Rf3 Bxd4? This move is a very human decision. I want the b5 square and I want to play Be8 and transfer the other bishop to my king's defense. This capture perpetuates both of these ideas simultaneously but Komodo absolutely hates the idea. The problem with the plan is simple. A bishop cannot be on the king and queensiude simultaneously. Therefore, this is not a winning idea. The exchange is poor. 18. Bxd4 Be8 19. Qh3 Qe7? This move is a waste of time. Looking back on this game playing g6 earlier and controlling the a6-f1 diagonal makes a lot more sense. 20. Kh1 White finally gives up on his premature mating ideas with pieces and decides on a different method of mating attack using the g-pawn. 20. ... Bg6? Bb5 makes more sense ... countering white's kingside aggression by opening the game and controlling the mating squares while making my own threats on the queenside. That could be a winning plan if white persists too far. 21. Rg1 b5 22. Rff1 b4 23. Qd3 Rfc8 24. Bd1 a5? After a series of very strong moves and the right idea (attacking on the queenside to rob white of his time for attacking my king) I fall into this time wasting "gem". It is better to simply exchange the b-pawn and make white's lone c3 pawn a target of attack. Komodo correctly sees this as winning. Black is better even after a5 but it is much harder to convert to a win. 25. g4?? This is a howler of a mistake and I completely miss the continuation. One sample variation is 25... fxg4! 26. Qe3 (Black threatened to win the queen by capturing with the bishop after the knight checks on f2) bxc3 27. Bxg4? Nd2! 28. Rf2 Be4+. That variation is only eight half-moves deep but from the starting position moves like Nd2 are particularly hard to see as the squares are completely undefended at the beginning of the variation. 25. ... bxc3 26. bxc3 Rab8 27. gxf5 Bxf5 28. Bg4?? I was so hopelessly taken in by the idea that White had a strong attack that I failed to make a concrete observation that there was an attack to be discovered against the queen with once again this key check by the knight on f2 winning white's queen. 28. ... Qh4?? This hands the game over to white in all logical variations. A shame. 29. Bxf5 exf5 30. e6 g6 31. Rb1 Qe7? Qxf4 is better as my passer will be faster and the strong Nf2 is still in the air. 32. Be5?? Rxb1?? (Nf2+! for the third time is missed. This time winning the queen outright.)33. Rxb1 d4?? A sacrifice out of desperation and missing the saving and completely winning fork for the fourth time. 34. Qxd4 Nxc3?? A losing blunder. Komodo recommends Qxe6 as drawing but giving up the seventh rank seems a hard decision for black to make. 35. Qxc3 this wins but Qd7 is much faster. Qxe6 36. Qxc8+ Qxc8 37. Rb8 Qxb8 38. Bxb8 Kf7 39. Kg2 Ke6 40. a4 Kd5 41. Bc7 Kc4 42. Bxa5 Kb3 43. Kh3 Kxa4 44. Be1 Kb5 45. Kh4 h6 46. Bc3 Kc5 47. Bg7 Kd5 48. Bxh6 Ke6 49. Kg5 Kf7 50. h4 1-0

Friday, June 9, 2017

2017 Kentucky Open Round 3 Anton Taylor, 2030 - NM Davis Whaley, 2357 Sicilian Canal/Moscow Variation (B51)

2017 Kentucky Open Round 3 
Anton Taylor, 2030 - NM Davis Whaley, 2357 
Sicilian Canal/Moscow Variation (B51)

1. e4 c5 2. Nf3 d6 3. Bb5+ Nd7 4. a4 I had recently seen a game or two from Magnus Carlsen in this variation last year where this odd looking move had been thrown in. It's interesting. It should be pointed out that Black scores rather well in this variation at top level but I'm a risk taker. 4. ... Nf6 5. Nc3 g6 6. O-O Bg7 This is where theory currently sits on this variation and where several different moves have been tried for white's 7th. The move I choose is not in the database. 7. d4 cxd4 8. Nxd4 O-O 9. Be3 Nc5 10. f3 a6 11. Bc4 Bd7 12. Qd2 Rc8 I was regretting playing Bc4 here after this move that forces me to make some kind of concession. There are three main moves here for white ... Ba2, b3, and Bd3. I didn't like Ba2 Because controlling the a2-g8 diagonal isn't really my priority. I want to solve the tension on the queenside from the g7 bishop and rook on c8 in the most efficient way. 13. Bd3 Komodo gives immediately that Nxd3 leads to a black advantage but giving up the c5 knight is difficult to human eyes. I wagered that after Nxd3 I could keep my grip on c3 and maybe even expand safely in the center leading to a small edge for white. Davis must have come to the same conclusion. 13. ... Qc7 14. Nde2 Bc6? I think this is the first real mistake. Allowing me to block up the c-file will make black's queen and rook look stupid. Sure, I have to give up the "good" bishop I have controlling the dark squares but with the file closed the bishop on d3 can come back to c4 with a more solid grip than previously. 15. Bxc5 dxc5 16. Qe3 Nd7 17. f4 White has a classic center but he has to be really careful. 17. ... Qb6 18. a5! I really liked this move. White wants to play Bc4 and e5 when he will be comfortable but in most move orders a knight ends up on b6 to ruin white's plans. This move thrown in fixes that problem. 18. ... Qb4 19. e5 c4 20. Be4 Qc5 21. Qxc5 Nxc5 22. Bxc6 Rxc6 23. Rfd1 At the time this seemed the most principled reply. Develop the pieces. 23. ... e6?? Ne6 was the only move that didn't lose on the spot. Once again my opponent and I labor under the same errors. I thought e6 was forced. 24. Rd4 f6 25. exf6 Bxf6 26. Rxc4 Rd8 27. Rd1 Rxd1+ 28. Nxd1 Kf7 29. b4 Na4 30. Rxc6 bxc6 31. Kf2? This move is still completely winning but the immediate c4! stops all the counterplay. 31. ...  c5 32. Nc1 a humble retreat to reorganize the piece. komodo sees that b5! is a winning sacrifice but I couldn't see that far. It seems obvious now that you offer a piece sac on c3 and black either "counter-sacrifices" by declining white's and losing a piece or the a-pawn marches to queening unimpeded. Fascinating. 32. ... cxb4 33. Nd3 Be7 34. Ke3 Nc3 35. Nxc3 bxc3 36. Kd4 Bd8 37. Nc5 Bxa5 38. Nxa6 Kf6 39. g4 h5 40. h3 hxg4 41. hxg4 g5 42. Ke4 gxf4 43. Kxf4 Bd8 44. Nc5 Bc7+ 45. Ke4 Bd6 46. Nd3 Kg5 47. Ne5 The last ten moves are so have been played with black having second on his clock and playing with the time delay to stay alive. The position is a draw as my win has slipped through my fingers but ... 47. ... Bxe5?!? Black takes the trade and that runs into a lost king and pawn. 48. Kxe5 Kxg4 49. Kxe6 Kf4 1-0 I don't remember the exact moves from here but it is completely winning for white. 

Thursday, June 8, 2017

2017 Kentucky Open Round 2 Mike Thomas, 1977 - Anton Taylor, 2030 Richter-Veresov System (A45)

2017 Kentucky Open Round 2 
Mike Thomas, 1977 - Anton Taylor, 2030 
Richter-Veresov System (A45)

1. d4 Nf6 2. Nc3 I took a moment here. I have never played a tournament game against the Veresov. I have seen it many times in blitz games over the internet but I never bothered to study any of the openings's variations. As it turns out I follow a sideline but it works out. The only problem with my approach was that time was spent on the earlier moves where it might have been saved had I done my homework. 2. ... d5 3. Bg5 Nbd7 4. Qd3 c6 5. Nf3 h6 6. Bxf6 Nxf6 7. e4 dxe4 8. Nxe4 Nxe4 9. Qxe4 Qa5+ 10. c3 Bf5 There are a couple of master games in this position but in both of those white plays Qe5 and the games eventually dwindle into a draw. 11. Qf4 e6 12. Bc4 g5 Rd8 followed by Bd6 gives black a comfortable lead in development. The text was played in order to make kingside castling somewhat distasteful for white. However, it does not turn out so useful. The weakness of the black king leads to a drawn ending where there might otherwise have been a win. 13. Qe3 Bd6 14. a4 b5 15. Bb3 O-O 16. Ne5 Qc7? this is actually a mistake. b4 was better fixing the a-pawn on an awkward square and restricting white's counterattack options. 17. h4 g4 18. axb5 cxb5 19. O-O a5 20. Bd1 Bxe5 21. dxe5 Rfd8 22. Qxh6? It would be better to solidify the e-pawn and then force black to defend his h-pawn. Qxe5 23. Qg5+ Qg7 24. Qf4 a4 As is often the case in these types of positional problems I move the "wrong pawn". But truthfully the symmetry is suggestive of a draw. 25. Re1 Rd5? Useless. It is better to go ahead and play a3 to liquidate the pawn islands. The current state of things will result in white building up an attack on both the b and g pawns. 26. Be2 Rad8 27. Red1 Qf6? At a depth of 20 or so Komodo claims that the only move that isn't losing here is Kh7. 28. g3 Rxd1+ 29. Rxd1 Rxd1+ 30. Bxd1 a3 31. bxa3 Qxc3 32. Bxg4 Bd3? At the time I thought this was a saving move. My opponent apparently suffered under the same delusion. The correct white reply is 33.Bh5! when all black replies fail and he is lost. 33. Qg5+ Kh8 34. Qh6+ Kg8 35. Qg5+ 1/2-1/2 and we agree to a draw.

Wednesday, June 7, 2017

2017 Kentucky Open Round 1 Anton Taylor, 2030 - Kevin French, 1800 Sicilian Moscow (B52)

2017 Kentucky Open Round 1 
Anton Taylor, 2030 - Kevin French, 1800 
Sicilian Moscow (B52)

1. e4 c5 2. Nf3 d6 3. Bb5+ Bd7 4. a4 Nf6 4. ... a6 is probably the most accurate response here. 5. Nc3 e5 6. d3 Be7 7. Bg5 This is the most logical move that develops the last minor piece. However, the position holds no real secrets. It is a pretty dead drawn position. Meaning that the only positive white can claim is that his development is slightly more useful but not enough to win. 7. ... O-O 8. Qd2 Nc6 9. Bxf6 Bxf6 10. Nd5 Rc8 I expected Bg4 but I saw that as not much of a threat. This move seems like much the same as the rest of black's development. good for equality and not much else. 11. h4 I decide to stir things up and I think this is the most interesting way to do that if not particularly principled. 11. ... Be7 12. g3 Be6 Kh8 preparing f5 is probably the best way to go. the f5 break has to be where black's advnatge lies. 13. Bc4 Nd4 14. Nxd4 cxd4 15. f3 f5 16. O-O Qd7? good alternatives to this move are f4 and Rc5. Play on the two bishop files are similar plans that give black a slight edge. 17. exf5 Rxf5 18. f4 Rcf8 19. g4 R5f7 20. f5 Bxd5 21. Bxd5 Bxh4 22. g5?? In my opinion this is the decisive error. When I had calculated this variation on move 17 I missed that 24. Bxg5 can be interjected. Had I re-evaluated g5 properly when it appeared on the board I would have seen the problem. however, I trusted the earlier calculation as is my worst current habit and failed to convert a nice position. 22. ... Kh8 23. Be6?? This is a worse sin. However my previous note applies to this move as I had looked at it on move 17 and failed to recheck here. Black's reply is technically winning. 23. ... Qe7 24. Bxf7 Bxg5 25. Qh2 Qxf7 26. Rae1 Qd5 27. Re4 Rf6 28. Qh5 Be3+ 29. Kg2 Qf7?? throwing away the win. Qc6 guarding the back rank and simultaneously threatening Qxc2+ is simple and strong. 30. Qxf7 Rxf7 31. c4 Bf4 32. Rfxf4! exf4 33. Rxf4?? Terrible. Without even thinking about it I miss a mate in 2 on the back rank! For all of black's sins White's are far worse in this game. That is how an Expert can draw a low class A player ...  commit worse sins. 33. ... Kg8 34. Kg3 Kf8 35. Kg4 Ke7 36. Rxd4 g6 37. fxg6 hxg6 38. Kg5 Rf2 39. b4 Rg2+ 40. Rg4 Rb2 41. c5 Ra2 42. cxd6+ Kxd6 43. b5 Ra3 44. Rd4+ Kc5 45. Rc4+ Kd5 46. Kxg6 Rxd3 47. Rc7 Ra3 48. Kf7 Rxa4 49. Rxb7 Kc5 50. Ke7 Ra5 1/2-1/2