Shakmaty Bereolos - The Official Chess Site of Peter Bereolos

1/29/20 - BCE-302a, Neumann-Steinitz, Baden Baden 1870

Many consider the tournament in Baden Baden to be the first supertournament. Looking at Sonas' rankings at the start of the tournament, it comprised of 9 of the top 11 players in the world. Sonas' #1, Adolph Anderssen emerged on top with 11/16, a half point ahead of #3 Wilhelm Steinitz at 10.5, while #2 Gustav Neumann was another half point back at 10. In an unusual result for a double round robin, Anderssen swept his two games against Steinitz, who beat Neumann twice. Neumann completed the circle by beating Anderssen twice. Today's BCE game between Neumann and Steinitz was critical to the final standings.

I couldn't find any complete game analysis of this game. Perhaps the editors at the time did not want to waste the ink on a 124 move game. However, the move total in this game was artificially inflated. This was one of the first tournaments with clocks, but did not feature the 3-time repetition rule. I don't know if the players were trying to work a loophole in the time control, but for two stretches of the middle game they repeated the sequence Bf5-e4, Qf3-g4, Be4-f5, Qg4-f3. The first time they repeated the position 6 times before Neumann deviated, while the second time saw Steinitz deviate after 4 repetitions. All this sets the scene for the endgame after 85.Kd3

The game continued 85...Kg3 86.Rg1+ Kf4 87.Rf1 and here they repeated that sequence 7 more times before Steinitz finally deviated with 101...Re8 101...Rxc7 102.Rxf3+ Bxf3 103.Ne6+ Kf5 104.Nxc7 Bxc6 doesn't really offer any chances as White can establish his king on g1 from where it cannot be dislodged. However, I'm not sure what the point of Re8 is. 102.Kc4 Kg3? He should just have returned the rook with 102...Rc8 and settled for a draw 103.Nxf3! Bxf3 104.Rxf3+?? The rook does not need to be sacrificed immediately as the Black bishop cannot conver c8. White should win by 104.Kc5! intending to simply walk the king to b7. 104...Bg4 105.Rg1+ Kf4 106.Rxg4+ and White is a key tempo ahead of the game. 104...Kxf3! Finally reaching the starting point of BCE-302a 105.Kd5 Ra8 106.Ke5 Ke3 107.Kf6 Rc8? 107...Kf4 is the BCE correction 108.Kxg6! Kf4 109.Kf6! Rxc7 110.g6! Rxc6+ 111.Kf7! Kf5 112.g7! Fine stopped here indicating a clear draw, but Steinitz went on to win the game 112...Rc7+ 113.Kf8 Kf6 114.g8N+! Ke6 115.Nh6! Rh7

116.Ng4? The way to defend N vs. R is to keep the knight and king close to each other. So the correct move was 116.Ng8! Many endgame books use this game as an example of how to win when the knight and king become separated, but Fine chose to use only studies in that section of BCE. 116...Rh4 117.Ne3 Re4 118.Nd1 Rf4+ 119.Kg7 Rf3 120.Kg6 Ke5 121.Kg5 Kd4 122.Kg4 Rf1 123.Nb2 Rb1 124.Na4 Rb4 0-1

1/22/20 - BCE-231c, Judd-Mackenzie, 3rd match game 1881

Today, St. Louis is rightfully regarded as the chess capital of the US. The same was true in the late nineteenth century when a match was organized between top local player Max Judd and George Henry Mackenzie who was thought of as the US champion based on his victory in the 5th American Chess Congress. For perspective on their strength, at the end of the match Sonas ranks Mackenzie as #3 in the world and Judd as #11.

The match was somewhat strange in a couple of respects. The schedule does not seem to have been fixed, so while only 13 games were played, they stretched over a 3 month period. The lack of draws (only one) was not unusual for the time, but in a stat that must make Andras Adorjan smile, Black won the first 7 games. Overall, 6 of Mackenzie's 7 wins in his +7 =1 -5 victory came with the Black pieces, including this week's BCE position.

There are several interesting points prior to the BCE position, so I'll pick up the game after 35...Nf5

36.Ra5 Surprisingly, this move was criticized in both The Chess Monthly and Deutsche Schachzeitung. While a knight is generally superior to a bishop with all the pawns on same side, White should have no problems holding the draw due to the limited material plus the doubled Black g-pawns. I think it would be more risky for White to leave the rooks on. 36...Rxa5 White also holds easily in the rook ending 36...Rb2 37.Rxf5 Rxd2+ 38.Kf3 Rd3+ 39.Kf2 Rg3 40.h4 37.Bxa5! Kf7 38.Bd2 Kg6 39.Ba5 Ng3 40.Bc7 Kf5 41.Kf3 Nh5 42.g4+?! White naturally wants to exchange pawns, but after this move, Black gains a passed pawn, so White should have just waited. For example, The Chess Monthly gives the line 42.Ba5 Nf6 43.Bc3 Ne4 44.Be1 (and of course there is nothing wrong with 44.Bxg7) 42...fxg3 43.Bxg3 Nf4 White has to make a few precise moves in the pawn ending after 43...Nxg3 44.Kxg3! The key is to wait for Black to play Kg6 before playing Kg4 44...Kf6 45.Kf3! (45.Kg4? Kg6! 46.Kf3 Kh5 47.Kg3 g6 followed by ...Kh4 and ...g4 winning) 45...Kf7 46.Kg3! Kg6 47.Kg4! Kh6 48.h4! 44.h4 g4+ 45.Ke3 Nd5+ 46.Kf2 Ke4 The Chess Monthly gives a win for Black by collecting the h-pawn, but White draws with accurate play as his king is not as confined as it was in the game 46...Kg6 47.Be5 Nf6 48.Kg3 Kh5 49.Bd4 Ne8 50.Bc3 Nd6 51.Kf4 g6 52.Be1 (instead of 52.Kg3? Nf5+ 53.Kg2 Nxh4+ 54.Kg3 Nf5+! 55.Kg2 Kh4 56.Kh2 g3+ 57.Kg2 Kg4 58.Bf6 Nh4+ -+) 52...Nf5 53.Bg3! Nxh4 54.Bf2! Nf5 (54...Nf3 55.Kg3!) 55.Be1! g3 56.Kf3! Kh4 57.Kf4! Kh3 58.Bxg3! Nxg3 59.Kg5! The great number of only moves that White has to play in this line show that Black would have great practical chances. All of this risk for White is the consequence of 42.g4?! 47.Bd6 Ne3 48.h5 Finally reaching the starting point of BCE-231c 48...Nf5 49.Bf8 Fine's 49.Bc7 Nd4 50.Bd6 Kf5 51.Bf8 Ne6 52.h6 is also good enough 49...g3+ 50.Kg1? Also losing was 50.Kg2? Kf4! 51.Bxg7 Nxg7! 52.h6 Nf5! 53.h7 Nh4+! and Black stops the pawn with 54...Ng6; but the BCE correction 50.Ke2!= would have saved the half point. 50...Kf3! 51.Bxg7 continuing with his plan, but it was already too late to back up with 51.Bc5 Kg4 52.Kg2 Nh4+ 53.Kg1 Kxh5 54.Bd4 Kg4 55.Bc5 Kh3 51...Nxg7! 52.h6 Nf5 Fine gives the route to e2 as 52...Ne6 53.h7 Nf4 but all the other sources I found give the text, including the on the scene newspaper coverage in the St. Louis Globe-Democrat. 53.h7 Nd4! 54.h8Q Ne2+! 55.Kf1 g2+! 56.Ke1 g1Q+! 57.Kd2 Qc1+ 58.Kd3 Nf4+! 59.Kd4 Qa1+ 0-1

1/18/20 - Wang-Yu, 2019 CCCSA Fall GM

In the last couple of years, the Charlotte Chess Center & Scholastic Academy has been putting on a number of norm events. An endgame from the top section of their fall event caught my eye as another example of the side defense gone wrong. This one comes from the game between Kevin Wang and US Women's Champion Jennifer Yu. After 52.Rxh4

This is the same ending we looked at in the He-Becerra game, except there are some extra pawns on the queenside. These pawns make some significant differences in the assessment of this ending. The first point is that they provide some shelter for the White king. Without the pawns on the board, Black would win here with Rb7+! cutting off the White king and soon reaching the Lucena position. However, the presence of the queenside pawns are also an additional target for Black. Additionally, the extra pawns mean that White will not be able to rely on exchanging rooks to reach a drawn pawn ending. 52...d3? This push looks natural, but should have cost the half point. The pawn on d4 was serving a purpose of taking away the square c3 from the White king. Black should have first manuevered her rook into a position where a3 could be targeted before making this advance 52...Ke3 53.Rh3+ Rf3 54.Rh6 Rf2+ 55.Kc1 Rf1+! 56.Kb2 Re1 57.Rh3+ Kd2 58.Rh2+ Re2 59.Rh1 d3 60.Kb3 Re3 61.Kb2 Ke2 62.Kc3 d2+! 63.Kc2 Rh3! Black has to throw this in. White can hold after (63...Rxa3? 64.Rh2+! Kf1 65.Rh6 Ra2+ 66.Kd1! Kf2 67.Rh3! cutting off the Black king and setting up stalemate tricks 67...Rb2 68.Rf3+ Kg2 69.Rd3 Rxb4 70.Rxd2+ Kf3 71.Ra2!) 64.Rg1 Rxa3

The point of pushing the rook to the g-file is seen in the variation 65.Rg2+ Kf1 66.Rg6 Ra2+ 67.Kd1 Kf2! and here g3 is not available to the White rook, so the Black king escapes and it just takes a little technique to finish things off 68.Re6 Kf3! 69.Re8 Kf4 70.Re6 Kf5 71.Re8 Rb2 72.Ra8 Ke4 73.Rxa6 Kd3 74.Rd6+ Kc4 -+ 53.Rh6? White has to latch on to the pawn with his king in order to tie the Black rook to the c-file 53.Rh2+! Ke3 54.Kc3! Rc7+ (54...Rf6 55.Rh3+) 55.Kb2 d2 56.Rh3+! Ke4 57.Rh4+! Kf3 58.Rh3+! Kg2 59.Rd3! 53...Ke2? again rook maneuvers were called for 53...Rf2 54.Rh1 Re2 -+ 54.Re6+? The same mistake we saw in He-Becerra. In that game White had two ways to draw, but here Black blocks a side check 54.Rh2+? Rf2!, but White still draws with the other method by attacking the pawn with the king 54.Kc3!= 54...Kd1 0-1

1/15/20 - BCE-221, Kosek 1910

The Czech endgame composer Vojtech Kosek published a number of studies involving knight and pawn versus bishop. Fine used a couple of them in BCE. Walter Korn looked at BCE-221 in the August 20, 1955 edition of Chess Life . The critical position arises after move 8.

Korn repeats Fine's incorrect line starting with 9.Kb7, which is addressed on the correction link. He also points out another way to draw against Fine's 9.Na3? with 9...Bg1 10.Kd7 Bh2! 11.Nb5 Bb8!. Finally, Korn gives another winning line which he states is Kosek's intended win with 9.Nd6 Ke2Korn's K-K2 is a typo. Benko changed the main line to 9.Nd6 is the revised edition, but continued 9...Kf4 10.Nc4 and wins immediately since the Black king blocks the h2-b8 diagonal 10.Nc4 Bg1 11.Kd7 Bh2 12.Nd6!

The van der Heijden endgame database, generally gives the original authors line as the main line and shows subsequent cooks as variantions. When I looked up this position, it gave the same line as Fine had. This got me curious if I could find the original source online. I was not optimistic because the original is in Czech, but amazingly I did manage to find scans of old issues of Časopis českých šachistů. Volume 4 has the original problem on page 2 and the solution on page 117. As the proposed solution matches exactly what appears in BCE, it is not clear where Korn got the variation with 9.Nd6. HHDBV credits the move to Tokarev and Mamiev in a 1981 issue of Shakmaty v SSSR, but that is several decades after Korn's analysis.

1/14/20 - Kaidanov-Bereolos, 1996 Kings Island Open

I've added my game against Gregory Kaidanov in the final round of the 1996 Kings Island Open to the GM games section. This was a very one sided game. I chose a meek retreat in the opening rather than a bold piece sacrifice that would have given me many interesting attacking possibilities. Initially, it doesn't look like Black should have enough pieces for such speculation, but somehow the Black rooks can find ways to quickly enter the attack. The engines think the best choice for White is to allow a perpetual check.

1/11/20- He-Becerra, 2019 US Masters

When I discussed BCE-324b, I mentioned that in R+P vs. R positions where the defending king is not in front of the pawn, the defense can be tricky. In last year's game from the US Masters between Anthony He and Julio Becerra, the veteran GM kept trying against his young opponent and was eventually awarded with a mistake. At first glance, it looks like Black should have a big advantage in the rook ending after 33...Rxb2

Black has an extra pawn and White has blocked doubled g-pawns, but it is not so simple. 34.a4? 34.Rh7+!? is a worthwhile alternative 34...Kd6 35.Rg7 Rxa2 36.Rxg6 and the g-pawns make their utility felt. The lead pawn is further advanced than Black's queenside pawns and the back pawn shields the file from attack from behind. It looks like White holds after 36...b5 37.Rg8 Ra4+ 38.Kf3 b4 39.Rb8 Ke5 40.Kg4 a5 41.g6 Kf6 42.Kh5 Ra3 43.Rf8+ Kg7 44.Rf7+ Kg8 45.Ra7 However, it is always a difficult decision to allow your opponent connected passed pawns. 34...b5 35.axb5 Black is far ahead of the previous variation after 35.a5 Ra2 36.Rh7+ Kd6! 37.Rg7 Rxa5 38.Rxg6 b4 35...Rb4+! Driving the White king back before recapturing. Black can't really make progress on 35...Rxb5? 36.Rh7+ Kd6 37.Ra7 a5 38.Ra6+ 36.Ke3 Rxb5 37.Rh7+ Kd6 38.Ra7 a5 39.Kf4 Rb4+ 40.Ke3 Rb5 41.Kf4 Rb4+ 42.Ke3 a4 43.Ra5 Rb3+ 44.Kf4 e5+ 45.Ke4 Rb4+! 46.Kf3 Rb3+? Becerra probably just wanted to repeat moves here to gain time on the clock, but this move gives White a saving opportunity 47.Ke4? 47.Kf2 a3 48.Ra6+! Kc5 49.Rxg6! Kb5 50.Rg8 and the g-pawn gives White enough counterplay 47...Rb4+! 48.Kf3

48...Ke6? With a second crack at it, Black still doesn't find the key. The Black king should head towards the queenside. Pardoxically, in comparison to the previous variation with the pawn on a3, Black should be winning here because he can build a bridge with his king and rook. 48...Kc6 49.Ra6+ (49.Rxe5 a3 50.Ra5 Rb3+ 51.Kf4 Kb6 52.Ra8 Kc5 53.Ke5 Rb5 54.Kf6 Rb6+ 55.Kg7 Kb4-+) 49...Kb5 50.Rxg6 a3 51.Rg8 Ra4! 52.Rb8+ Kc5 53.g6 a2 54.g7 a1Q! 55.g8Q and the White king is too exposed, a sample winning line is 55...Qh1+ 56.Kf2 Ra2+ 57.Ke3 Qe1+ 58.Kd3 e4# 49.g4! Rb3+ 50.Ke4 Rb4+ 51.Kf3! Kd6 52.Ra6+ Kd5 53.Rxg6 e4+ 54.Ke2 Rb2+ 55.Ke3! Rb3+ 56.Ke2 Ke5 57.Ra6 Kf4 58.g6 Rb2+ 59.Kd1 I think 59.Kf1 is more precise, denying the Black rook access to the g-file 59...Rg2 60.g7 Rxg4 61.Ke2 Rg2+ 62.Kf1! 62.Ke1? Rxg7! 63.Rxa4 Kf3! 64.Kf1 Rh7 65.Kg1 e3 62...Rxg7 63.Rxa4! Rg3 64.Ra8 Rb3 65.Rf8+ Ke3 66.Ra8 Rc3 67.Ra7 Rc1+ 68.Kg2 Rd1 69.Re7 Rd2+ 70.Kf1 Rf2+ 71.Ke1 Rh2 72.Kf1 It is good practice to go to the short side, but White could even hold this position with 72.Kd1?! Ra2 73.Re8 Ra1+ 74.Kc2 Kf3 (74...Re1 75.Rh8 Kf2 76.Rh2+ Kg3 77.Rh8 e3 78.Kd3) 75.Kd2! 72...Rh1+ 73.Kg2 Re1 74.Ra7 Rd1 75.Re7 Rd8 76.Kf1 Rf8+ 77.Kg2 Kd3 78.Ra7 Rc8 79.Ra6 e3

80.Rd6+? White draws with either 80.Ra3+ Rc3 81.Ra8 or probably simplest 80.Kf3 threatening Rd6+ 80...Rf8+ 81.Kg2 and White is again prepared for check from the side 80...Ke2! now White has no side checks and his king is cut off. 81.Ra6 Rg8+! 82.Kh2 Kf2 83.Rf6+ Ke1 84.Re6 e2 85.Re7 Rg5 86.Kh3 Kf2 87.Kh4 Rg1 88.Rf7+ Ke3 89.Re7+ Kd3 0-1

1/8/20 - BCE-487a, Simpson-Bereolos, USCL 2008

This week's BCE position is a very practical endgame fortress. Fine did not cite a game for this position, but it arose in my game against Ron Simpson in the opening match of the 2008 US Chess League. He had shuffled around for a few moves and after 64...Qe5 he went for the queen trade.

65.Qd4?! Objectively, the position is a draw (although your chess engine likely shows a winning advantage to White), but White would still maintain practical chances by probing with the queens on for awhile longer. I knew the position without queens was a draw as it is one of the blue positions in Dvoretsky's Endgame Manual. However, I did not remember all the ins and outs and gave my opponent an opportunity to win. 65...Qxd4 66.Rxd4 Bf6 67.Rc4 Kf8 68.Rc7 Kg8 69.Kg2 Be5 70.Re7 Bf6 71.Rb7 Be5 72.Kf3 Bf6 73.Ke4 Bc3 74.Kd5 Bf6 75.Ke6 Bc3 76.Rb3 Bd4?

I thought this was a flexible move that gave me the option to target White's kingside pawns, but with the White king on e6, the d4 square is poisoned for the bishop. 76...Ba1! is the only move to hold, 77.Rb4 Kg7! 78.g4 hxg4 79.Rxg4 and as in the game, Black is in time to counterattack with his king 79...Kh6! 80.Kf7 Kh5 77.g4? The right idea, but the wrong execution. It isn't easy, but White wins with 77.Rb4 Bc3 78.Rc4! Bb2 (78...Be1 79.Kf6! Kh7 80.Rc6 Bd2 (80...Bxg3 81.Kg5!) 81.Kf7 g5 82.hxg5 Bxg5 83.Re6 Bd2 84.Kf6! Kh6 85.Kf5+ Kg7 86.Re2 Bc3 87.Rh2 Be1 88.Rxh5) 79.g4 hxg4 80.Rxg4! Kh7 81.Kf7! Kh6 82.Rxg6+! Kh5 83.Rg2! Bc3 84.Rh2!

White is winning, but it is still difficult. Zugzwang plays a key role. A sample variation 84...Be1 85.Kf6 Bg3 (85...Bxh4+ 86.Kf5! is immediate zugzwang) 86.Rh1 Bf2 87.Kf5 Be3 (87...Bg3 88.Rd1 Bf2 89.Kf4 Bb6 90.Kg3) 88.Rh2 Bg1 89.Rg2 Bb6 90.Rb2 Bc5 91.Rc2 Be3 92.Ke4 Bh6 93.Rh2 Bg7 94.Kf4 Bf8 95.Kf3 Ba3 96.Ra2 Be7 97.Re2 Bf6 (97...Bxh4 98.Rh2! Kg5 99.Rh1! taking away e1 from the bishop 99...Kh5 100.Kf4 is again zugzwang) 98.Kf4 Bd8 99.Rc2! Be7 100.Rd2 Bf6 101.Rd7 Bb2 102.Rh7+ Kg6 103.Rb7 followed by Kg4, White has finally broken the blockade on h5 and should win without further problems. 77...hxg4! 78.Rb4 Bc3 79.Rxg4 79.Rc4 Be1 79...Kh7! 80.Kf7 Kh6! 81.Rxg6+ Kh5! 82.Rc6 Be1! 83.Kf6 Kxh4! 84.Kf5 Kg3 85.Ke4 Kg2 86.Rg6+ Kf2 87.Kd3 Bb4 88.Rf6+ Kg2 89.Rb6 Bc5 90.Rb5 1/2-1/2

1/7/20 - George-Bereolos, 1983 Western Open

My game versus Richard George in the 1983 Western Open featured a very similar ending to the one seen in the Kasparov-Karpov game. After 39...Rxd6

40.Rxf7? White should just accept defending the drawn 3 vs. 2 ending with 40.Rb7! Rf6 (40...Rd5 41.Rxf7 and the Black rook doesn't get behind the pawn) 41.Rxb5 Rxf4 Instead, restores material equality, but risks losing. 40...Rb6! 41.Rc7 b4 42.Rc2 b3 43.Rb2

43...Kg6 44.g4 44.Kf2 Kf5 45.Ke3 h5! 46.g3 Rb4 47.Kf3 h4 48.g4+ Ke6 49.Ke3 Kd5 is similar to the next variation, but the Black pawn being on h4 instead of h6 does make a difference in the technique, but not the result. A key variation where the difference is seen is 50.Kd2 Rd4+ (here 50...Ke4? is a mistake since 51.Kc3 Rb7 52.f5 Kf4 53.Rxb3 and the pawn ending is drawn after a rook exchange) 51.Ke3 (51.Kc1 Rd3 52.Re2 Rxh3 53.Re7 Rf3) 51...Re4+ 52.Kf3 Rb4 53.Ke3 Kc4-+ 44...Kf6? 44...Rb4! preventing Kf2 appears to give Black the decisive tempo he needs to win 45.Kg2 Kf6 46.Kf3 Ke6 47.Ke3 Kd5

This position is reached in the game with the White king already on d3. Here, White has to give ground. 48.Rb1 Alternatives do not help a) 48.f5 Kc4; b) 48.Kd3 Rxf4; c) 48.h4 Kc4; d) 48.Kd2 Ke4 (with the pawn on h6, 48...Rd4+ is less effective as the h-pawn is not passed after ...Rxh3 49.Kc1 Rd3 50.Re2 Rxh3 51.Re7=) 49.f5 Kf4; 48...b2 49.Kd3 (49.Kd2 Rxf4) 49...Rb3+ 50.Kc2 Rxh3 51.Kxb2 Ke4 and wins 45.Kf2! Ke6 46.Ke3 Kd5 47.Kd3! Rb4 47...Rb7 48.Rb1! (48.f5 Ke5! 49.Kd2 Kf4 50.Kc1 Kg3) 48...b2 49.h4! Rb3+ (49...Rb4 50.h5! Rxf4 51.Rxb2 Rxg4 52.Ke3 Rg5 53.Kf4 Ke6 54.Rh2) 50.Kc2 Rb4 (50...Rh3) 51.h5! Rxf4 52.Rg1 Rb4 53.Kb1! Ke6 (53...Ke5 54.g5) 54.Rf1! Rxg4 55.Kxb2!

with the draw we saw in the analysis of Kasparov-Karpov. Black will only be able to win the White h-pawn at the expense of his g-pawn. 48.f5 Rb6 48...Ke5 allows White to swap the defensive duties of the rook with the king 49.Kc3 Rb7 50.Re2+ Kf4 51.Kb2 Black could keep g7 covered 48...Rb7 but again White seems to hold the balance as eventually the black rook has to help in harvesting the White kingside pawns allowing White to activate his rook 49.Kd2 (49.Kc3 Ke4) 49...Ke4 50.Kc1 Kf3 (50...Rc7+ 51.Kb1 c3 52.Rd2 Rxh3 53.Rd7) 51.Rd2 Re7 52.Kb2 Re3 53.Rd7 49.Kc3 Ke4 50.Re2+ Kf3 51.Re6 Rb7 52.Kb2 Kg3 53.Rg6 Kxh3 54.f6 gxf6 55.Rxh6+ Kxg4 56.Rxf6! Rb4 57.Rc6 Kf4 58.Rc3 Ke4 59.Rxb3 Rxb3+ 60.Kxb3! 1/2-1/2

1/3/20 - Kasparov-Karpov, 1984/85 World Championship, Game 6

In the Bogoljubow-Fine game, we saw Fine take his chances in a pawn ending rather than staying in a rook ending where the attacking rook could support its outside passed pawn from behind. A famous example of such an ending occured in the sixth game of the first K-K match after 42...Ra8

This ending has been heavily analyzed, but perhaps there is still more of the story to tell. 43.Ra5 Kb6 44.Ra2 a5 45.Kf1 a4 46.Ke2 Kc5 47.Kd2 a3 48.Kc1 Jan Timman recently revisted the K-K clashes and declares this the losing move, suggesting instead 48.Kc3 However, I think White can still hold after the text, and the real losing moment comes one move later. 48...Kd4 49.f4? The line beginning 49.Kb1 Rb8+ 50.Ka1 Rb2 51.Rxa3 Rxf2 was first anlayzed by Yusupov in Informant 38 and repeated by Kirolyi in his book on Karpov's endings and by Kasparov in his book on the matches. Here, instead of Yusupov's 52.Ra6? why not 52.h4 keeping the pawn protected as well as maintaining the barrier on the third rank, which gives time for the White king to assist in the defense. I don't find a win for Black. For example, 52...Ke4 53.Kb1 Rf3 54.Ra4+ Kf5 55.g4+ Ke5 56.h5 Rf4 57.Ra5+ Kf6 58.Kc2 Rxg4 59.Ra1!

and strangely enough it doesn't appear that Black can win the h-pawn without giving up his g-pawn. The 7-piece tablebases confirm that this is a draw. A. 59...Rh4 60.Ra5; B. 59...Rg5 60.Rh1 Kf5 61.Kd3 Kg4 62.Ke4; or C. 59...Kg5 60.Rh1! Rh4 61.Rg1+! Kf6 62.Rf1+! Ke7 63.Rf5 Alternatively, Black could advance the g-pawn first, as he did in the game, but White has enough time to bring his king to the defense 53...g4 54.Kc1! g4 55.Kd1! Rf3 56.Ra6 Rxg3

57.Ke2 (not 57.Rxh6? Ke3! and the White king gets cut off) 57...Rg2+ 58.Kf1! Rh2 59.Rxh6 with a draw After the text, Karpov's execution was flawless 49...Ke4! 50.Kb1 Rb8+! 51.Ka1 Rb2! 52.Rxa3 Rxh2! 53.Kb1 Rd2! 54.Ra6 Kf5! 55.Ra7 g5! 56.Ra6 g4 57.Rxh6 Rg2 58.Rh5+ Ke4! 59.f5 Rf2 60.Kc1 Kf3 61.Kd1 Kxg3 62.Ke1 Kg2! 63.Rg5 g3 64.Rh5 Rf4 65.Ke2 Re4+! 66.Kd3 Kf3! 67.Rh1 g2 68.Rh3+ Kg4 69.Rh8 Rf4 70.Ke2 Rxf5 0-1

1/1/20 - BCE-72, Teichmann-Blackburne, Berlin 1897

We start 2020 with a classic maneuvering king and pawn ending. This example has been a staple in endgame literature, although my research found that most analysts don't start from a position that occurred during the quoted game between Richard Teichmann and Joseph Henry Blackburne. Fine even claims it was adjudicated as a draw, but this is not the case, as Blackburne won.

Berlin 1897 was a massive 19-player round robin (it began with 20 players, but von Bardeleben withdrew after one round) to celebrate the 70th anniversary of the Berlin chess club. Several notable players took part, but the event was missing Lasker, Steinitz, and Tarrasch. Blackburne took clear 3rd with 12/18, trailing Charousek (13.5) and Walbrodt (13). His opponent in the opening round, Teichmann, finished well off the pace in 16th with 6.5. After 48...Nb5

49.Nc3? The pawn ending is lost for White because Black has two advantages, a better king position and the spare tempo c6-c5. White should try to take one of those advantages away by activating his king with 49.Kf3!? when White has drawing chances as illustrated by the following variations A. 49...f5 50.Nc3 Nxc3 51.bxc3 Kf6 52.Ke3 Ke5 53.f4+ Kd5 54.Kf3=; B. 49...Ke6 50.Ke4 h5 51.Nc3 Nxc3+ 52.bxc3 g5 53.hxg5 fxg5 54.f4 h4 55.g4 Kf6 56.f5= 49...Nxc3 50.bxc3 Ke6 51.f3 51.Kf3 Kf5 is the difference 51...Kf5 52.Kf2 h5 53.Kg2 g5 54.Kh3 Ke5 55.hxg5 fxg5! 56.Kg2 Euwe and Hopper start here, except they have White to play 56...Kf5 This is BCE-72 except it is White to play. The British magazine The Field also showed this diagram with Black to move in their September 17, 1897 edition. Benko corrected this by just presenting it as a position with Black to move and not quoting any game source. 57.Kh2 Znosko-Borovsky starts from here except that he has the White king on f2. 57...Kf6 Black has to time the ...h4 advance accurately 57...h4? 58.Kg2! Ke5 59.f4+!= 58.Kg2 Kg6 58...h4? 59.f4! 59.Kh259.Kh3 is a harder nut to crack, Black has to revert to the plan shown in the BCE correction to put the king on e5 and then play h4 59...h4 Now everything runs smoothly for Black 60.Kh3 60.f4 doesn't work anymore 60...gxf4! 61.gxh4 Kh5 62.Kh3 and the spare move comes into play 62...c5! 60...hxg3! 61.Kxg3 Kf5! 62.Kf2 On 62.Kg2 Blackburne gives the following winning variation 62...Kf4 63.Kf2 c5 64.Ke2 Kg3! 65.Ke3 Kh3 66.Kd2 Kh2 62...Kf4 63.Ke2 Kg3 64.Ke3 c5 65.Ke2 Kg2! Most sources end the game here, but Blackburne's book gives one more move for each side 66.Ke3 Kf1 0-1