Shakmaty Bereolos - The Official Chess Site of Peter Bereolos

Games versus GMs

Welcome to my chess page. This is mostly random thoughts and analysis in the form of a chess diary with other sections of the site slowly developing. A lot of the content will come from my own experience. There are two reasons for this. One, so I can use this site as a self-improvement tool. Two, so you the readers will have content that is not found on other chess sites. Follow the link to the left to reach my annotated games against grandmasters. Send me comments and ideas

Pete

Corrections to Basic Chess Endings
Links
Archives

4/22/18 - Burnett-Bereolos, 1986 US Junior Open

The Kasparov vs. The World ending showed that this type of queen ending is so complex that a strong team with several days to analyze and use of the best engines at the time could still make an error. Indeed, it is somewhat remarkable that there was only one move in the entire endgame (54...b4?) that changed the assessment of the position. In a tournament game with the clock ticking it is expected that there would be several more errors. This is illustrated by my game against Ron Burnett in the 1986 US Junior Open. The 7-piece ending was reached after 48...Qxc5

This position should be drawn and eventually was, but not without serious mistakes by both sides 49.Qe6+ Kh5 50.Qg6+? [An unnecessary check. The White queen was well-positioned stopping both Black pawns from advancing. White should get his pawn moving with 50.e5= so he is ready for e6 with serious counterplay after the queen moves. It isn't clear how Black can even test White here.; Instead 50.Qd5? loses Qf2+! (50...Qxd5+ 51.exd5 h3 52.d6 h2 53.d7 h1Q 54.d8Q is a draw) 51.Kg7 h3 52.e5 h2 53.e6 Qe1 54.Qf3+ Kh4! and White has no good check] 50...Kg4-+ 51.Qe6+ Kg3? [51...Kf4!-+ is pretty simple. Black collects the e-pawn with an easy win.] 52.e5? Now this is too slow. The difference from move 50 is the position of the Black king which can support the pawns. Ironically, the only drawing move is 52.Qd5!= because the attack on the g-pawn gains White a crucial tempo. 52...Qf2+ (52...Qxd5+ 53.exd5 h3 54.d6 h2 55.d7 h1Q 56.d8Q= is a draw, but Black certainly has practical chances after 56...Qe4) 53.Kg6 g4 54.e5 h3 55.e6 h2 56.e7= 52...h3!-+ 53.Qb3+ Kg2? [53...Kh4!-+] 54.Qb7+ Kg3 55.Qb3+? [When defending it is natural to try to repeat the position, but this let's Black correct his mistake. 55.Qe4!= again a centralization with the queen supporting both offense and defense 55...Qf2+ 56.Kg6 h2 57.e6! Qf3 58.Qxf3+ Kxf3 59.e7 h1Q 60.e8Q and the g-pawn will soon be collected since 60...g4? 61.Qa8+ turns the tables] 55...Kh4!-+ correcting the error from move 52 56.e6 Qf5+ 57.Kg7 h2 [57...Qe4 might be a little simpler] 58.Qb8 Qf4! 59.Qxf4+ gxf4! 60.e7 h1Q! 61.e8Q Black should win this ending, but it was drawn after further misadventures as I have previously discussed.

This ending further emphasizes some of the lessons from the Kasparov vs. The World ending. In particular, centralization of the queen (55. Qe4! and 57...Qe4) and taking advantage of your trumps when defending (50. e5). Some other lessons are to not always be quick to give check when defending (50. Qg6+? and 55. Qb3+?) and perhaps the hardest one is to not necessarily repeat a position for a second time (55.Qb3?). This last one is particularly hard to resist, but if the position has changed (the first time the queen started at e6 and the second time it was on b7) the differences need to be appreciated. Finally, another observation on how subtle and difficult queen endings are. The positions after Black's 49th and 51st differ only slightly by the position of the Black king. However, in these positions the moves e5 and Qd5 give opposite results.


4/20/18 - Finegold-Bereolos, 2009 Kings Island Open

I've added my game against Ben Finegold at the 2009 Kings Island Open to the GM games section. I've played many games against Ben dating all the way back to 1979, but this was so far our only game after he achieved his final GM norm. Last year, Ben and his wife Karen Boyd opened the Chess Club & Scholastic Center of Atlanta. I finally got down there for a tournament a few weeks ago. It's a very nice, large facility and the turnout for the tournament was great. Hopefully, they will have continued success there.

There wasn't much to add to the game notes from my original tournament report other than a few opening notes. I had a rather decent game until I tried to stir up a kingside attack and ended up losing quickly thereafter.

4/13/18 - Just Like Starting Over

I've been looking at the 7-piece tablebases lately. It's like a bottomless mine for extracting chess gems. I've found a lot of interesting results showing many errors in both my games and analysis. Trying to decipher the reasons for some of the evaluations has been very educational, so I thought I'd restart the blog to share some of what I've found.

My posts will still be fairly infrequent. In today's world of hot takes and tweets I think there is still a place for longform writing. So instead of instant analysis and tournament reports, I'll take my time and hopefully produce polished, worthwhile posts.

In the spirit of starting over, before digging into my own games, I want to look at a famous 7-piece position that was the major topic when this blog started: the queen ending from Kasparov vs. The World 1999. We pick up the action at he start of the queen ending after 50...d1=Q

This is one of the most complicated 7-piece endings that is likely to turn up in a practical game. Queen endings are notoriously difficult, even more so here with all 3 pawns being passed. At the time of this game, there were only 5-piece tablebases available. Ironically, this drove some on the World team to advocate giving up the pawns to reach a solved position, since the Black pawns get in the way in some lines allowing the White King to hide. But in many of the lines, we will see that the pawns are beneficial to Black. I think the lure of the 5-piece ending may have been what eventually doomed Black. Garry now played the tricky 51. Qh7 setting up a discovered check and attacking the b-pawn. This position led to the first big debate among the World team between the natural looking 51...b5 (passed pawns must be pushed) versus 51...Ka1 stepping out of the line of check. The vote ended up going to 51...b5 The oracle now tells us that both moves were good enough to draw, but I wonder how many of those who voted for 51...Ka1 would have played that move over the board. It should be noted that those are the only two moves that draw showing just how tough queen endings can be. For example 51... b6? loses as will be shown below. 52. Kf6+ Kb2 53. Qh2+ Ka1 The move I voted for 53... Ka3? loses 54. Qg3+! Ka2 (54... Ka4 55. Qf4+! brings the queen to an even better position than f2) 55. Qf2+! stopping Black checks on d4 or the f-file 55...Ka3 (55... Ka1 56. g6 b4 57. g7 is similar) 56. g6 b4 57. g7 Qg4 (57... Qd5 58. Qg3+ and queens) 58. Qf5 Qd4+ 59. Kg6 Qc4 60. Qf3+ Kb2 (60... b3 61. Qa8+) 61. Kf6 (threatening Qg2+) 61...Qd4+ 62. Kf7 Qa7+ 63. Kg6 Qg1+ 64. Kh6 b3 65. Qd5 centralizing and controlling the queening square 65...Qe3+ 66. Kg6 Qg3+ 67. Kf7 Qf4+ 68. Ke8 Qa4+ 69. Kf8 Qf4+ 70. Qf7 here the Black d6 pawn is in the way 70...Qh6 71. Qf5 (preparing Kf7) 71...Kc1 72. Kf7 b2 73. g8=Q b1=Q 74. Qg1+ winning 54. Qf4 b4? The losing move, which nevertheless got nearly 60% of the vote. Black had two moves to hold 54... Qd3, which got my vote, guards both pawns, angles for perpetual check and makes use of the Black pawns 55. g6 Qc3+ (without the pawn on d6, this move could be met by 56. Qe5! transitioning to a winning pawn ending.) 56. Kf7 Qc7+ 57. Kg8 (57. Kf8 Qb8+! 58. Kg7 b4! was my line at the time, which also gives enough counterplay to hold) 57... Qc4+! this move is not possible if Black only plays his pawn to b6 on move 51. The other drawing move is 54... Qd5 controlling the queening square 55. g6 b4! using the b-pawn to create counterplay. Again, this is not possible with the pawn on b6. 56. Qxb4 (56. g7 b3!) 56... Qe5+! 57. Kf7 Qd5+ 58. Ke7 Qe5+ with perpetual check) 55. Qxb4 mate in 82. The World might have been misled by the fact that without pawn d6 here the 5-piece tablebase shows a draw. Indeed, most of the remaining positions in the game are drawn without the Black pawn. While the win is still complicated, I'm not going to add a whole lot to the remaining moves. I didn't see any glaring errors in my original postings other than the evaluations at the end of various lines are always still wins for White. The Boss did not allow any further chances and the 6-piece ending has already been much discussed in the years since the game ended including previously by me. Some of the first 6-piece tablebases were created just to get to the bottom of this ending. 55...Qf3+ 56. Kg7! Only here. 56. Ke7? Qe3+ 57. Kf6 Qe5+ and Black gets perpetual. 56. Kg6? blocks the pawn from advancing so Black will hold with 56...d5 56...d5 This was the move that Kasparov thought was the loser. My vote went to 56...Qe3 not allowing Qd4 and if 57. g6? Qe5+ Black achieves a perpetual, but White still wins by repositioning his queen before pushing the pawn 57. Qa5+ Kb2 58. Qb5+ Ka1 59. g6 57. Qd4+! Kb1 58.g6! Qe4 59. Qg1+ Kb2 60. Qf2+ Kc1 61. Kf6 d4 62. g7 the World did not play the most resisting defense, so now it is down to mate in 28. 1-0

Some of the lessons from this ending are: 1. Differences between actual positions and theoretical ones must be carefully considered. In this game, it is easy to give an incorrect evaluation of "draw" on the basis of "drawn even without the Black d-pawn" 2. Centralization of the queen can be very powerful because of the numerous squares it controls from the center of the board. 54...Qd5 and 57. Qd4+ are examples of this. 3. Always be on the lookout for counterplay. The main trump in the Black position was the passed b-pawn, Black shoudn't have been in a hurry to give it away. 4. Continually set problems for your opponent. Although the position was drawn, Kasparov managed to create difficult options that even a team of strong players using the best computers of the time were not able to overcome.


older material


Stats since June 1, 2006
www.000webhost.com