Shakmaty Bereolos - The Official Chess Site of Peter Bereolos

10/30/19 - BCE-349, Kling & Horwitz 1851

Fine cites this week's BCE position as Horwitz 1881, but the position first appeared in the September 13, 1851 edition of Kling and Horwitz' journal The Chess Player as a study by the authors. However, their solution a couple of issues later was incorrect.

1...Ra6! 2.Kg6 Rb6 3.Kh5 Rb1? with a draw according to the authors. White wins with 4.g6 and the pawns are too far advanced to be stopped. Black had to pin the g-pawn with 3...Rb5! as demonstrated in the BCE main line.

Horwitz did reuse the position in 1881 for his column in The Chess-Monthly, reversing the colors. There, his solution did not include the incorrect line. Berger, who is Fine's primary reference, cites Horwitz in The Chess-Monthly down to the page number, but again reverses the colors and has the defending rook on the back rank. He also reintroduces the original mistake of meeting Kh5 with the rook moving to the first rank.

Thus, the line with 2...Rd6? looks like it was introduced by Fine. I think it is a good addition to the analysis and Benko probably should have just corrected the note instead of completely deleting it.

10/28/19 - Eswaran-Cervantes, 2019 US Masters

We often see cutting of king along a file in rook endings to prevent the defending king from defending against a passed pawn. The ending from the game between Ashritha Eswaran and Thalia Cervantes at this year's US Masters showed a twist on that theme. Black had been battling a piece down for quite some time, but White let the Black pawns get out of control and Black should be able to hold after 52.Bd5

Black only had a couple of minutes left on her clock, plus 30 seconds per move. She likely saw that she could get the piece back a reach what looks like a drawable rook ending a pawn down. However, the particular arrangement of pieces allows for a study-like win. 52...c2? Better is 52...Rb2 when Black should be able to win the bishop for only one of the pawns. 53.Bxb3 Rb2 54.Bxc2 Rxc2

55.Ra1? The key to victory is cutting the Black king off from defense of the g-pawn 55.Kg4! Rc5 56.Rf5! and White can creep forward and will eventually win the g-pawn to reach Lucena's position 55...Rc4 56.Ra6 Rb4 57.e4 Rc4? Black needed to keep the g-pawn in the crosshairs 57...Rb3+ 58.Kf4 g5+! 59.Kg4 Re3 60.Ra4 Kf6 and White can't make progress 58.Kf4! Rc5 59.e5 Rc1 60.Ra7+ Ke6 2 pawns down is hopeless, but Black is too passive after 60...Kf8 61.Kf5 61.Rxg7 Rf1+ 62.Kg5 Rf5+ 63.Kg6 Rxe5 64.Rf7 Re3 65.g4 Rg3 66.Rf4 Ke5 67.Rf5+ Ke6 68.g5 Rg1 69.Rf2 Ke7 70.Kh6 Ke8 71.g6 Ke7 72.g7 Rh1+ 73.Kg6 Rg1+ 74.Kh7 Rh1+ 75.Kg8 Ra1 76.Rh2 Kf6 77.Kh8 1-0

10/25/19 - Bereolos-Pojman, Highland 1979

I found another pawn ending with some warts. This one comes from my first year of tournament chess. Ed Vano was an Indiana state champion who did a lot to promote junior chess in northern Indiana. This game was from the first of a number of small 6-player invitationals that he held in 1979. I won my first two games, and played the highest rated player, Paul Pojman, in the last round. After 26...Kg8

White should play Ne2 with the idea of Nd4 blockading the isolated pawn and keeping the bad Be6 on the board. Instead, I simplified with 27.Nxd5 Rxd5 28.Rxd5 Rxd5 29.Rxd5 Bxd5 30.Qxd5 Qb6+ This move is better than taking the c-pawn, but even after 30...Qxc2 31.Qxb7 Black can simply deliver perpetual check. 31...Qd1+ 32.Kf2 Qd2+ 33.Kg3 Qg5+ 34.Kh3 Qh5+ 31.Kf1 Qxb2 32.Qd8+ Kh7 33.Qd3+ Kg8 34.Qd8+ Kh7 35.Qd3+ g6 36.Qb3 Qxb3 Black could keep the queens on with 36...Qa1+ 37.Kf2 Qd4+ 38.Kf1 Qd7 when it is still equal, but Black has a symbolic advantage because of the disconnected White pawns. 37.cxb3!

It is hard to imagine anything of interest could happen in this ending, yet the game continued. 37...Kg7 38.Ke2 Kf6 39.Kd3 Ke5 40.Ke3 g5 41.g4 b5 42.b4 f6 43.Kd3 Kf4 44.Ke2! f5 45.gxf5 Kxf5 46.Ke3 h5 47.Kd3 Kf4 48.Ke2! Ke5 49.Kd3 Kf5 50.Kd2 g4 51.Ke3 Ke5 52.fxg4 hxg4! 53.Kd3 Kf4?

Even though it has been 40 years since this game was played, I am sure my thought process was Kd4-c5--b6xa6xb5-Pa4-a5-a6-a7-a8 is 10 moves to queen,while Black queens in only 6 with Kf3-Kg2xh2-Pg3-g2-g1. Therefore, White must retreat. 54.Ke2? White wins with 54.Kd4! Kf3 (54...Kf5 55.Kd5! Kf4 (55...Kf6 56.Ke4) 56.Ke6 leads to the same result) 55.Ke5! and the kingside pawns will be traded after which White wins the race back to the queenside. 55...Kg2 56.Kf4 Kh3 57.Kg5! This is similar to the king and pawn ending that could have arisen in the game Newsom-Bereolos 54...Ke4 55.Kf2 Kf5 56.Kg3 Kg5 57.Kf2 Kf4 58.Kg2 Kg5 59.Kf2 1/2-1/2

Lesson from this ending: When calculating a race, make sure it is really a race and there are not other alternatives.

10/23/19 - BCE-226a, Horwitz 1880

This week begins another round of BCE corrections based on stuides by Bernhard Horwitz. Today's battle of knight plus two connected passed pawns probably should be attributed to Kling and Horwitz who published the following position as Black to play and draw in the November 29, 1851 edition of their journal The Chess Player

This is BCE-226a with colors reversed and the position mirrored on the queenside. Their solution in the December 6th edition was without error except for a typo 1...Bg4!They correctly noted this as the only drawing move 2.Nc6 Be2! 3.Ne5 Bb5 4.Nc4 Bd7 5.Nd6 This is given as 5.Nd4, which is illegal 5...Bg4=.

The version Fine used appeared in the End-games by B. Horwitz section of the April 1880 edition of The Chess-Monthly with the more conventional White to play and draw stipulation. Horwitz kept the same main line, but added some subvariations. This seems to be where 2.Bd7? was first examined, but Horwitz' line had a lot of flaws.

1.Bb5! Ng4 2.Bd7? Nf2+! 3.Kg2 Ke3? 3...Nd3 as in BCE, wins 4.Be6? 4.Bf5 or 4.Bc8 draws since White can meet 4...Nd3 with 5.Kh3! 4...h3+? 4...Nd3! wins as the fork threat gives Black a key tempo 5.Bd7 Nf4+ 6.Kh2 Kf3 7.Bc6+ Kg4 8.Bd7+ Kh5 9.Be8+ Ng6 and Black is ready for g4 5.Kg3! with a draw.

In practice, positions with knight plus two connected passed pawns versus a bishop should be won by avoiding positions like BCE-223a. The pawns need to be carefully advanced and the king and knight should be maximized before pushing the pawns. Here is an example from the 1963 Brazilian championship game between Jose Thiago Mangini and Antonio Rocha, both of whom were two time national champions. After, 39.Kxh2 White can only dream of a draw and Black should win by systematically advancing the pawns.

39...Kg5 40.Kg3 h5 41.Bb7 h4+? Black should first maneuver his knight to f5. Then, he would be ready for Kf6 and g5. From f5, the knight can come to g7 if necessary to defend the h5 pawn. 42.Kh3 Nh5 43.Bc8? Pardoxically, putting the bishop in position for a blockade is the losing move. The reason for this is that the king can't maintain its position on h3. So, White should anticipate ...Nf4+ with 43.Kg2! Kf4 (43...Nf4+ 44.Kf3) 44.Kh3 g5 45.Bc8 with a draw as in the variations of the Horwitz study. 45...Ng3 46.Bg4 Ne4 47.Bc8! Nf2+ 48.Kg2! Ke3 49.Bf5! (not 49.Be6? a la Horwitz) 43...Nf4+! 44.Kh2 Kh5 45.Bd7 g5 46.Be8+ Ng6 This is the position at the end of the variation with 4...Nd3! above. The rest is smooth sailing for Black 47.Kg2 g4 48.Kh2 Kg5 49.Bc6 Nf4 50.Be4 g3+ 51.Kh1 h3 52.Kg1 Kh4 53.Kh1 h2 54.Bf5 Kg5 55.Be4 Nh3 56.Kg2 Kh4 57.Kf1 Nf2 58.Ba8 h1Q+ 59.Bxh1 Nxh1! 60.Kg2 Nf2 61.Kf1 Kh3 62.Kg1 g2 0-1

10/20/19 - Timman-Karpov, Belfort 1988

A rook ending with two rook pawns also occurred in the game between Jan Timman and Anatoly Karpov at the World Cup tournament at Belfort in 1988. I previously looked at the very end of this game and wondered if the game score was correct as it appeared that a stalemate trick was missed twice. I have a few osources that I didn't have at the time of my original post. What should likely be the definitive source is Timman's annotations in Informant 45, which is the version I'll use below. Surprisingly, Timman's notes on the rook ending are somewhat light and he doesn't mention either of the stalemates. Karsten Muller and Yakov Konoval also noted the variety of game scores that exist, but also chose the Informant version when they looked at it using the 7-piece tablebases in their book Understanding Rook Endgames. Finally, Kavalek's book on the World Cup includes the unannotated game score. This one is two moves shorter and only has one instance of the missed stalemate. One other book that I thought might have this game was Endgame Virtuoso Anatoly Karpov by Karolyi. However, the game is not included perhaps because the maestro hit quite a few sour notes in this performance.

The rook ending began after an exchange of minor pieces on e7 37...Kxe7

38.Rxg4 Rf8 39.Rg1? The pawns on c5 and h5 allow White to first push the king away with 39.Re4+! Kd7 40.Re1! 39...Rf4? Black should activate his king first 39...Kf6 40.Kc2 41.Rf1 Kg5 42.b5 Kg4 43.Kc3 Kf3 and Black should win without trouble 40.Rf1 Ke6 41.Kc2 Rxc4+ 42.Kd3 Rxb4 Another try is 42...Rf4 43.Ke3 Rxb4 44.Kxf3 Kd5 (44...Rh4 45.Rb1 Rxh5 46.Rxb7 Rxc5 is similar to the game) 45.Rg1 Kxc5 46.Rg6 Rb6 (46...Ra4 47.Rxh6! Rxa2 48.Rg6 (48.Rh7? b5! 49.Ra7 b4 50.h6 b3 51.h7 Rh2!) 48...b5 49.h6 b4 50.Kg3 Ra1 51.Kg2 Ra2+! 52.Kg3=) 47.Rxb6 Kxb6 48.Ke4 Kc5 49.Kf5! Kd5 (both sides get queens in the race 49...b5 50.Kg6! a5 51.Kxh6!) 50.Kg6! Ke6 51.Kxh6 Kf6 52.Kh7 Kf7 53.h6 b5 54.a3! a5 55.a4! bxa4 56.Kh8 a3 57.h7! and Black loses if he doesn't deliver stalemate 43.Rxf3 Rh4 44.Rf1 Rxh5 45.Rb1 Rxc5 46.Rxb7 Kd5 47.Rd7+ Ke5 48.Re7+ Kf5 49.Rf7+ Kg6 50.Rf4? 50.Rf2 to maintain checking distance while guarding a2 holds. 50...Kg5? an obvious move, advancing the king with tempo, but it gives back the draw. Strange as it seems Black wins only with 50...Re5! 51.Kd4 (51.Ra4 a5 is different than the game as White does not have Ke2) 51...Re2 52.a4 Kg5 53.Rf1 Ra2! and Black picks up the a-pawn while White is nowhere near Vancura's defensive position. 51.Ra4 a5 52.Ke2 Rf5 Muller starts from this position, which Timman evaluated as -+. White can still hold, I'll only give light notes from here, see Muller's book for more details. 53.Ra3? [53.Rc4=] 53...Kg4 54.Rc3 h5 55.Rc8 h4 56.Rg8+ Rg5 57.Ra8 Kg3 58.Kf1 Kf3? 59.Rc8! Ke3 60.a4? [60.Rc4=] 60...Rg4! 61.Rc5 h3! 62.Re5+ Kf3 63.Rh5 Kg3! 64.Kg1 Rxa4! 65.Rg5+ Kh4 66.Rc5 Rg4+ 67.Kh2 a4 68.Rc3 Rg2+! 69.Kh1 Rg4? 70.Kh2 [70.Rxh3+] 70...Rg2+! 71.Kh1 Rg3 72.Rc4+ Rg4! 73.Rc3 Rb4 74.Ra3 Rg4? 75.Rc3 [75.Rxh3+=] 75...Kg5 76.Kh2 Rh4 0-1

10/18/19 - Smith-Bereolos, 2019 US Masters

I had mentioned that GM Bryan Smith seemed to be a once per decade opponent for me. This cycle was broken at this year's US Masters where we were paired for the second time in less than a year. I got a quite decent position out of the opening, but made a tactical oversight and did not follow up with strong resistance.

10/16/19 - BCE-386a, Euwe-Bogoljubow, Zurich 1934

We complete the set of BCE positions from Bogoljubow's games with a very interesting rook ending against Max Euwe in the Zurich 1934 tournament. This tournament, played shortly after the Alekhine-Bogoljubow match saw a mix of top players along with the best Swiss players. The cream rose to the top with World Champion Alekhine a full point clear of Euwe and Flohr with Bogoljubow another half point back. Alekhine didn't fare too well against his nearest rivals, losing to Euwe and drawing with Flohr and Bogoljubow, but scored a clean sweep of the other 12 players.

I didn't find any analysis outside of BCE to compare my notes with. I don't have the tournament book with Alekhine's annotations. It would seem that the original edition should be in the public domain, but I didn't find it digitized anywhere. Other major endgame books seem to have overlooked this game, which is a pity as it is quite instructive. The players entered the rook ending after an exchange of minor pieces on e5 38...fxe5!

39.Rc6 BCE-386a starts from here 39...Ke6 40.Ra6 h5 41.a4 Rxd6 42.Rxa7 Rb6 43.Kd3 Rd6+ 44.Ke3 Rb6 45.a5 Rb3+! 46.Kf2 46.Kd2 Rg3 (in the revised edition of BCE, Benko gives a worse version for Black with 46...Rb2+ 47.Kc3 Rxg2 48.Ra8 concluding and wins, but Black is drawing here as well with 48...Kd6) 47.a6 Kd6 also gives Black sufficient counterplay to draw. For example, 48.Ra8 Rxg2+ 49.Kc3 Kc5 50.a7 Rg3+ 51.Kb2 Kb6 52.Re8 Kxa7! 53.Rxe5 Rg4 46...Rb2+? As shown in the correction link, with the king on f2, it is the optimal time for Black to play 46...Rb4! so that after 47.Ra8 Rxe4 48.a6 the rook returns to the defense with 48...Rf4+! 49.Ke3 Rf7! which also has the benefit of sheltering the f-file allowing the Black king to enter White's position 47.Kg3 Rb3+ 48.Kh2 Ra3 49.a6 Kd6 the best try. Passive defense is hopeless because of the weakness of the e5 pawn. 49...Kf6 50.Ra8 Kg7 51.a7 Kh7 52.Kg1 Ra2 53.Kf1 Kg7 54.Ke1 Kh7 55.Kd1 Kg7 56.Kc1 Kh7 57.Kb1 Ra4 58.Kb2 Kg7 59.Kb3 Ra6 60.Kb4 Ra1 61.Kb5 Rb1+ 62.Kc6 Ra1 63.Rd8 Rxa7 64.Rd7+ Rxd7 65.Kxd7 Kf7 66.Kd6 Kf6 67.Kd5 50.Rg7? Since the defense hinges on a number of only moves, it is not surprising that first distracting the king with 50.Ra8! is the route to victory 50...Kc7 51.a7 Kb7 52.Rg8 50...Kc5! 51.Rxg6 51.Rd7 Rxa6 52.Rd5+ Kc4 53.Rxe5 Kd4 54.Re8 Ra5 doesn't give White anything 51...Kd4! 52.g4 hxg4? Black must first check the White king away from the kingside beginning with 52...Ra2+ 53.h5 Kxe4 53...Rh3+ 54.Kg2 Rxh5 fails to 55.a7! Rh8 56.Ra6! Ra8 57.Ra4+! 54.h6! Kf5

55.Rb6! a very subtle winning move, White needs to cover b2 in a key line 55.Rc6? e4! 56.Kg2 Ra2+! 57.Kf1 Ra1+! 58.Ke2 g3 59.h7 g2! 60.h8Q Ra2+! here the difference is revealed, White does not have 61.Rb2 so a drawn ending is reached after 61.Rc2! Rxc2+! 62.Kd1! g1Q+ 63.Kxc2 55...e4 56.Kg2 Ra2+ 57.Kf1 Ra1+ 58.Ke2 Ra2+ 58...g3 59.h7! g2 60.h8Q! Ra2+ 61.Rb2! is the point of 55. Rb6 61...Rxb2+ 62.Qxb2! g1Q 63.Qb5+!+- 59.Ke3 Ra3+ 60.Kd2 g3 61.h7! e3+ 62.Ke2 g2 63.h8Q g1N+ 64.Kf1 e2+ 65.Ke1 1-0

10/9/19 - BCE-361, Bogoljubow-Thomas, Hastings 1922

This weeks BCE position comes from the Hastings Tournament of 1922. That year's Hastings event was an exception to the normal round robin taking place across the new year. Instead, it was played in September and featured 4 top continental players Alekhine, Rubenstein, Bogoljobow, and Tarrasch along with 2 top English players Thomas and Yates. Alekhine and Rubenstein completely dominated scoring 7.5 and 7 out of 10 respectively, leaving all other players with negative scores.

Today's featured players tied for 3rd with 4.5 points each. Thomas was trying to bring home the point with an extra pawn and bishop versus knight after 60.c5

60...Ra8 In the tournament book Alekhine suggested 60...Ra7 in order to meet 61.Nd5+ with 61...Kf5 but after 62.Nb4 White will again have counterplay with his c-pawn. Best might be 60...Ke6 to stop the knight from coming to d5 61.Nd5+ Bxd5 62.Kxd5! The starting position for BCE-361 62...Ra6 62...g4 is the subject of the BCE correction 63.c6 Ke7 64.Kc5 Kd8 65.Kd6 g4 66.Rg3 a3 67.Rxg4 Ke8! 68.Re4+ Kd8 69.Rh4 Ke8! 70.Rh8+ Kf7 71.Rh7+ Ke8 72.Kc5 a2? Alekhine points out that Black draws with 72...Kd8! it might seem a little curious since on 73.Kd6 the king must go right back with 73...Ke8! effectively giving White 2 free moves. The point is that now the c-pawn is pinned, so White can't make further progress. Black probably should have found this since he had moved the king back and forth to e8 on several of the previous moves. 73.c7! Ra5+ 74.Kb6 Ra6+ 75.Kc5 Ra5+ 76.Kc6 Ra6+ 77.Kd5 Ra5+ 78.Ke6 78.Kc4 Ra4+ 79.Kb3 Ra3+ 80.Kc2 78...Ra6+ 1/2-1/2?

White wins in similar fashion to the previous variation 79.Kd5! (79.Kf5? Rc6 stops the pawn, while 79.Ke5?? lets Black queen with check) 79...Ra5+ 80.Kc4 Ra4+ 81.Kb3 Ra3+ 82.Kc2 I was a bit surprised by Alekhine's comment a win for White would have been a matter of luck, and a draw is a satisfactory conclusion.. I think White had made his own luck by continuing to put questions to Black and eventually forcing him to make several only moves. This seems to be the way Magnus Carlsen wins a lot of drawn endgames these days.

10/7/19 - Cezila-Zheng, 2019 US Masters

A recent rook ending featuring two rook pawns appeared in the third round game between Rubens Cezila Jr. and Michael Zheng at this year's US Masters. Black should be winning after 52.Rxf4

52...h3 53.Kf2 Rd2+? The path to victory is to cut off the White King from the g-file 53...Rg5 54.Rh4 Rh5! 55.Rg4+ Kh6 the king has to shelter behind his rook. Going to the f-file gives White the tempo he needs to bring his king to the defense (55...Kf5? 56.Ra4 or 55...Kf6? 56.Rf4+!) 56.Rg1 (56.Kg1 Rg5) 56...h2 57.Rh1 a4 and White is helpless against the advance of the a-pawn 54.Kg1! Ra2 55.Rh4 a4 56.Rxh3! a3 57.f4 Kf5

58.Rh5+? 58.Rb3 is a book draw by Vancura's method. White keeps the pawn under fire from the side tying the Black rook to defense in front of the pawn. If the Black king comes to the queenside White will check from the side along the f-file. If Black ever goes for Ra1+ and a2 only then does White go behind the pawn with Ra3 and now if the Black king approaches White can drive him off by checking vertically on the a and b files from the seventh or eighth rank. 58...Kg4! It appears that Black knew about the Vancura defense so he correctly avoids 58...Kxf4? When White can reestablish the defense with 59.Rh3! 59.Ra5 Kg3 60.Rg5+ Kh3 61.Rh5+ Kg4 62.Rg5+ Kxf4 63.Ra5 Ke4 64.Ra8 Kd4 65.Ra7 Kc3 66.Ra8 Ra1+ 67.Kf2 Kb2 68.Ke3 Rd1 69.Rb8+ Ka1 70.Ke2 Rb1! 71.Ra8 a2 72.Kd2 Kb2! 73.Rb8+ Ka3 74.Ra8+ Kb3 0-1

10/6/19 - Wengret-Bereolos, 1983 Illinois Junior Invitational

The blunder-filled finale of my game against Ted Wengret in the 1983 Illinois Junior Invitational could have led to a very instructive rook ending. The name of this tournament is somewhat strange as it was open to all juniors regardless of state. Unfortunately, the air conditioning at the community center where this event was held broke down, so the games were played in sweltering July heat. I'll offer that up as an excuse for the play of both players in this one. After 33.Bg4

Instead of the passive retreat 33...Nh6 I played for tactics 33..Nxh4?? There was a tactical solution with the sacrifice 33...Rc1 34.Qxc1 (34.Qf2 Nxg3 35.Qxg3 Rh1+ 36.Kg2 Rg1+-+) 34...Qxg3+ 35.Kh1 Qxh4+ when Black should have good winning chances with 3 pawns for the exchange and an exposed White king 34.Rxe7? 34.d6 cuts across Black's plans. On 34...Rc8 35.d7 looks much stronger than taking the knight (35.gxh4 Qxd6+ and 3 pawns plus White's exposed king gives Black compensation for the piece) 35...Rd8 36.Rxe7 and Black has lots of problems 34...Rc2+ 35.Kh3 Ng2? Black should bring his queen into play with 35...Qd6 since the knight is immune 36.Kxh4? Rh2+ 37.Bh3 Qf6+ 38.Kg4 h5# 36.Re8+ Kg7 37.Rxb8? 37.Qe7 should win for White 37...Qb2 (37...Qb5 38.Qf8+ Kf6 39.Qh8+ Kg5 40.Re5+ f5 41.Rxf5+ gxf5 42.Qg7#) 38.Qf8+ Kf6 39.Qh8+ Kg5 40.Re5+ f5 41.Qd8+ Kh6 42.Qf8+ Kg5 43.Qe7+ Kh6 44.Be2 a key noncheck cutting off the defense of the knight which is covering h4 44..Ne1 45.Qh4+ Kg7 46.Re7+ Kf8 47.d6 with a mating attack 37...Nxe1 38.d6 f5? The game should have finished with a draw by perpetual check with 38...Nd3 39.d7 Nf2+ 40.Kg2 (40.Kh4? Rc4) 40...Nxg4+ 41.Kf1 (41.Kf3? f5) 41...Nh2+ 42.Ke1 Nf3+ This is a slight variation on the usual perpetual with rook and knight as the rook stands on c2 instead of d2, but White can't take advantage of this since 43.Kd1? drops the d7 pawn to 43...Rd2+ 39.Bxf5? 39.d7 fxg4+ 40.Kxg4+- and the d-pawn will cost Black his rook. Now the White king gets caught in a mating net 39...gxf5 40.d7 Nf3 41.d8Q?! White can put up much more resistance with 41.g4 f4 (The engines point out 41...Ng5+ 42.Kh4 Ne4 (here 42...Nf7 43.d8Q Nxd8 44.Rxd8 is only a draw) as winning, but Black has to calculate that very accurately) 42.g5 Nxg5+ 43.Kg4 Nf7 44.d8Q Nxd8! 45.Rxd8 Rxa2 46.Rd7+ Kf6 47.Kxf4

I thought this endgame with two rook pawns would be a draw as White seems well set up to use Vancura's method. However, Black can take advantage of the fact that his pawns are not advanced to free his rook from its position in front of the a-pawn 47...Kg6 48.Kg4 Ra6 (48...h5+? 49.Kh3 would be a draw as Black will no longer be able to organize a check on the h-file) 49.Rc7 h6 50.Rd7 Ra4+ 51.Kh3 Ra5 52.Rc7 Rh5+ 53.Kg4 Rg5+! 54.Kh4 a5

The rook is free and the White king is cut off. Black just needs to walk his king over to the queenside. For example, 55.Rc6+ Kf5 56.Rxh6 a4 57.Ra6 Rg4+! 58.Kh3 Rc4-+ 59.Kg2 Ke4! 60.Kf2 Kd3! 61.Ke1 Kc2 and the White king is too far away to defend. 41...Rh2# 0-1

Lessons from this ending: 1. Try to use all your pieces in the attack. Black's attack could have been well coordinated against g3 with the queen and knight attacking the weak point at g3 while the rook defelected a defender with 33...Rc1. Likewise in the White attack in the line with 37.Qe7, the White queen and rook harrass the Black king, but the bishop plays an important role cutting off the defense of Ng2 which is guarding a key square. 2. Don't overlooked non forcing moves. The quiet move 44.Be2 would have been a key in the king hunt. 3. Material gain isn't everything. The White d-pawn is worth much more than the Black knight in the line beginning 34.d6, which is why keeping it with 35.d7 is better than taking the knight. 4. Always play for maximum resistance. White is losing at move 41, but he could have set Black much more problems with 41.g4 rather than allowing mate in one. 5. Create plans in the endgame then look for variations to support them. In the double rook pawn ending, Black has a problem in that his rook is in front of the a-pawn. If he can't defend the a-pawn from the side the win is fairly simple. Looking for a sequence to free the rook would lead to the Ra5-h5+ idea, which would be spoiled by the hasty 448...h5?

10/2/19 - BCE-488, Vidmar-Bogoljubow, Nottingham 1936

This week begins a series of BCE positions from the games of Efim Bogoljubow. Bogoljubow was Alekhine's challenger in his first two title defenses, but Alekhine won handily both times. Of course, most people would have preferred a rematch between Alekhine and Capablanca, but in those days the organization of the world championship was not an organized system like it is today.

Today's game comes from the historic tournament in Nottingham 1936. This tournament marked the first meeting between Alekhine and Capablanca since their 1927 match. Capablanca won their individual encounter en route to tying for first with Botvinnik. Alas, a rematch between the two great champions never took place and they only faced each other in one other tournament at AVRO 1938.

As for Bogoljubow, he was still one of the top players in the world at this time, but finished in a tie for 10th with only 5.5 out of 14. His opponent in this game, Milan Vidmar was just above him in the crosstable with 6. Their positions could have been reversed had Bogoljubow managed to hold the endgame after 31.Rg2

Things don't look great for Black, as the doubled pawns make him effectively two pawns down. However, Bogoljubow found a try that almost gained him a draw. 31...Rf3!? 32.Bxf3 exf3 33.Rf2? a terribly passive square for the rook. White shouldn't have much trouble converting the point after 33.Rd2! Kh5 34.Rd4 Bc6 35.Kc1 e5 36.Rd6 33...Kh5! 34.Kc2 White could try to avoid a bishop check on e4 with 34.Kc1!? but then Black has a different path to the draw 34...Kxh4! (34...Be4? allows White to activate his rook 35.Rd2! and Black can't capture the h-pawn 35...Kxh4 36.Rd4!) 35.e4 Bxb3! 36.Rxf3 (36.axb3 Kg3!) 36...Bxa2!= 34...Kxh4? 34...Be4+! 35.Kd2 Kxh4!= 35.Kd2? The starting position of BCE-488. Instead, White had a chance to win with the clearance sacrifice 35.e4! Bxe4+ (The bishop sacrifice doesn't work here since the White king is more actively placed 35...Bxb3+ 36.axb3 Kg3 37.Rf1 Kg2 38.Rb1 f2 39.Kd3 f1Q+ 40.Rxf1 Kxf1 41.Kd4+- 36.Kd2 Kg3 (36...Bd5 37.Ke3) 37.Ke3 and White can activate his rook and/or get his queenside pawn majority moving. 35...Kg3 36.Ke1 Be4 The easier draw is Fine's main line with 36...b5 which was pointed out by Alekhine in the tournament book. 37.a4 Bc6 38.Rb2 e5 39.Rd2 Be8? Black squanders two tempi giving White time activate his rook. As shown in the correction link, Black can stop White from freeing his rook by continuing to challenge the f1-a6 diagonal begining with 39...b6 planning Bc6-b7-a6. 40.Kf1 Bc6 meekly returning, but going the other direction didn't help either 40...Bh5 41.Rd6 Bg4 42.Rg6; 40...Bg6 41.Rd7 Bf5 42.Rg7+; or 40...Bf7 41.Rd7 Bxb3 42.Rg7+! Kh3 43.a5 which is similar to the game 41.Rd6! the rook is now active and the rest is fairly easy 41...Be4 42.Rf6! keeping f2 under control 42...Bd3+ 43.Ke1 Bc2 44.a5 Bxb3 45.Rg6+ Kh4 46.Kf2 now the f-pawn is firmly blockaded and the rook is free to mop things up. 46...e4 47.Rd6 Bc4 48.Rd4 Bb5 49.Rxe4+ Kh3 50.Re7 Bc6 51.Rg7 Kh4 52.Rg3 Kh5 53.Rxf3 Kg5 54.Rf4 Bh1 55.Ke2 1-0