{"id":5945,"date":"2019-10-07T08:50:14","date_gmt":"2019-10-07T16:50:14","guid":{"rendered":"http:\/\/www.danamackenzie.com\/blog\/?p=5945"},"modified":"2019-10-07T09:09:43","modified_gmt":"2019-10-07T17:09:43","slug":"endgame-workshop","status":"publish","type":"post","link":"https:\/\/danamackenzie.com\/blog\/?p=5945","title":{"rendered":"Endgame Workshop"},"content":{"rendered":"\n

Yesterday\u2019s chess party at Mike Splane\u2019s house turned into an endgame workshop, because both Mike Arne and Paulo Santanna came prepared with some beautiful positions.<\/p>\n\n\n\n

Ironically, both of them went to Spain this summer but they didn\u2019t play in the same events. Mike Arne played in the Benasque tournament, and showed us this deceptively simple position.<\/p>\n\n\n\n

\"\"
Black to move.<\/em><\/figcaption><\/figure>\n\n\n\n

FEN: 8\/8\/p4p2\/2k5\/6K1\/p1P4P\/3N2P1\/8 b – – 0 1<\/p>\n\n\n\n

Mike was playing Black, and it\u2019s his move here. As Mike says, it\u2019s obvious that Black\u2019s next two moves are \u2026 a2 and \u2026 a5 in some order. The first move threatens to promote the pawn, but it\u2019s not a real threat yet because of the defense Nb3+. The move \u2026 a5 is needed to put some teeth into Black\u2019s threat; it forces White to play Nb3 right away, lest his knight lose the b3 square.<\/p>\n\n\n\n

Mike didn\u2019t think that the move order made any difference, so he played 1. \u2026 a2??<\/strong> Astoundingly, this turns a won position into a lost position! But not a single person at the party saw White\u2019s winning plan. And in fact, Mike\u2019s opponent didn\u2019t find the win either. He played 2. h4??,<\/strong> and after 2. \u2026 a5<\/strong> Mike was once again on the winning track. The game finished 3. Nb3+ Kc4 4. Na1 <\/strong>(4. Nxa5+? is hopeless because after 4. \u2026 Kxc3 White\u2019s knight cannot stop the remaining a-pawn) 4. \u2026 Kxc3 5. h5 Kb2 6. h6 Kxa1 7. h7 Kb1 8. h8Q a1Q<\/strong>. This has the makings of a very long endgame, but Black should be winning. The f6 pawn definitely helps Black escape from White\u2019s checks. The game actually didn\u2019t last too much longer because of a mistake from Mike\u2019s opponent. 9. Qb8+ Kc2 10. Qc8+ Qc3 11. Qf5+ Kb7 12. Kh5? Qe5! <\/strong>Here White resigned.<\/strong> On the trade of queens, 13. Qxe5 fe, we once again get a pawn race where both sides queen, but this time Black queens first, plays \u2026 Qh1+ and \u2026 Qg1+, trades queens, and then promotes for the third (!) time with his remaining pawn.<\/p>\n\n\n\n

Now let\u2019s see where both players went wrong. After Mike\u2019s mistaken first move, White should have played the ingenious 2. Kf3!! The idea is to bring the king one step closer to the queenside. He still seems hopelessly far away, but this sets up a \u201cstalemate defense\u201d. After 2. \u2026 a5 3. Nb3+ Kc4 4. Na1 Kxc3 5. Ke2 Kb2 6. Kd2 Kxa1 7. Kc2, we see the point of White\u2019s idea: although Black has won the knight, he is unable to move the king out of the way of his own pawns. He is reduced to making pawn moves, and meanwhile White will promote the h-pawn. The cutest line is 7. \u2026 f5 8. h4 f4 9. h5 f3 10. h6 (White can simply ignore Black\u2019s pawn!) fg 11. h7 g1Q 12. h8Q+ with mate next move. <\/p>\n\n\n\n

In order to eliminate the \u201cstalemate defense,\u201d Mike should have played 1. \u2026 a5!! right away. Then White has to play 2. Nb3, and after 2. \u2026 Kc4 3. Na1 Kxc3 we transpose to the line actually played in the game, because White\u2019s king is too far away to play the stalemate defense.<\/p>\n\n\n\n

After Mike finished showing us this wonderful endgame, Paulo got up and showed us a similar study by Mark Dvoretsky, which has some of the same motifs of a knight trying to stop a rook pawn.<\/p>\n\n\n\n

\"\"
White to play and draw.<\/em><\/figcaption><\/figure>\n\n\n\n

FEN: 8\/4k2N\/7p\/1p6\/8\/8\/K7\/8 w – – 0 1<\/p>\n\n\n\n

Again, just as in Mike Arne\u2019s position, it seems as if there are two possibilities, 1. Kb3 and 1. Ka3, that are absolutely equivalent. The hard part once again is not finding the right move, but telling it apart from the wrong move!<\/p>\n\n\n\n

Most people would play 1. Kb3??, just because it moves the king one square closer to the kingside. Because this is a Dvoretsky endgame study, you might suspect that this is wrong and 1. Ka3<\/strong>! is right instead. But do you see why?<\/p>\n\n\n\n

The reason is that after 1. \u2026 Ke6!<\/strong>, White needs to extricate his knight in order to stop the h-pawn. Generally speaking, White can draw with the knight against the rook pawn if he can stop it before <\/em>it reaches h2. Then the knight can hop around the \u201ccircuit\u201d f1-h2-g4-e3 and can never be driven away.<\/p>\n\n\n\n

The beauty and ingenuity of Dvoretsky\u2019s study is that in this position White\u2019s knight has to travel a much longer circuit, which goes h7-f8-d7-c5-b3 (The key square!! This is why 1. Kb3?? is a blunder.) and then d2-f1-h2. Thus the drawing line is:<\/p>\n\n\n\n

2. Nf8 Kf5<\/strong> (Black does his best to shut the knight out, but this turns out later to be an unfortunate square) 3. Nd7 h5 4. Nc5 h4 5. Nb3!<\/strong> (not Nd3, because the knight has to blockade the pawn on h1, and as we saw in Mike Arne\u2019s game that doesn\u2019t work, unless the king can help out with a stalemate defense) h3 6. Nd2 h2<\/strong>. It seems as if the knight has arrived too late, but Dvoretzky has one more trick: after 7. Nf1 h1Q 8. Ng3+! <\/strong>forks the king and queen. Amazing! Of course 7. .. h1N 8. Kb4 is also a draw.<\/p>\n\n\n\n

Just going over these two endgames makes me feel as if I have gained 10 rating points. Certainly I understand the knight vs. RP scenario much better. It\u2019s interesting to see how the right moves depend on concepts rather than I-do-this-he-does-that style analysis. The concepts are (1) the idea of a knight \u201ccircuit,\u201d (2) the fact that a pawn on R7 is much harder to stop than a pawn on R6 because the knight gets trapped in the corner, (3) the possibility of the \u201cstalemate defense,\u201d and finally (4) the heroic \u201clet him queen but then fork the king and queen at the last minute\u201d defense.<\/p>\n\n\n\n

After this Mike showed us another endgame with a\nsimilar property \u2013 the correct moves depend on schematic thinking.<\/p>\n\n\n\n

\"\"
Black to play and win.<\/em><\/figcaption><\/figure>\n\n\n\n

FEN: 8\/pp2K1k1\/8\/2p1P3\/6B1\/2P4P\/PP3r2\/8 b – – 0 1<\/p>\n\n\n\n

This was a game played by two Bay Area masters: FM Josiah Stearman (White) and IM Elliott Winslow (Black). Winslow played 1. \u2026 Rd2?, a sensible-looking move that tries to keep White\u2019s king stuck in front of the pawn. But the trouble is that this is completely passive, and it doesn\u2019t even work because White can play e6, Ke8, e7, Bd7, and Kd8 in some move order, just as in the game. The game ended up a draw (Mike didn\u2019t show us how).<\/p>\n\n\n\n

If Black wants to play for a win, he has to try 1. \u2026 Rxb2<\/strong>, with the idea of wiping out all of White\u2019s queenside pawns and then sacrificing the rook for the e-pawn. Does this work?<\/p>\n\n\n\n

As Mike pointed out, it all depends on whether the\nthree connected passed pawns can beat White\u2019s bishop all by themselves.<\/em> He discovered a simple rule (actually, three of\nthem):<\/p>\n\n\n\n

  1. Is\nthe RP the \u201cwrong color\u201d? In other words, is the queening square opposite to\nthe bishop? The answer here is yes: White\u2019s bishop cannot cover a1.<\/li>
  2. Is\nthe defending king too far away to help stop the pawns? Here the answer is yes,\nalthough just barely, as we\u2019ll see.<\/li>
  3. Can\nthe pawns get to c4 and b5? If yes, then the pawns win, because they take away\nall the squares from which White\u2019s bishop could stop the a-pawn.<\/li><\/ol>\n\n\n\n

    So simple! Almost no analysis required. So in fact the following line would have won for Black: 1. \u2026 Rxb2 2. e6 Rxa2 3. Ke8 <\/strong>(unfortunately White has to lose a tempo this way, because 3. Kd8 would allow \u2026 Rd2+) 3. \u2026 Ra3 4. e7 Rxc3 5. Kd8 Re3! <\/strong>(\u2026 Rd3+ is unnecessary and would actually help White) 6. e8Q Rxe8 7. Kxe8 c4. <\/strong><\/p>\n\n\n\n

    \"\"
    Position after 7. \u2026 c4. White to move.<\/em><\/figcaption><\/figure>\n\n\n\n

    FEN: 4K3\/pp4k1\/8\/8\/2p3B1\/7P\/8\/8 w – – 0 7<\/p>\n\n\n\n

    Black is aiming for the b5-c4 setup, and White can\u2019t really stop it. For example, if 8. Bd7 c3 9. Ba4 b5! 10. Bc2 a5 and the pawn runs to a4 and a3. Once the pawns are on a3 and c3, they are unstoppable. <\/p>\n\n\n\n

    The trickiest variation in the position above is 8. Kd7<\/strong>. Here 8. \u2026 b5? is wrong, because 9. Kc6 gets the king into the defense in time. Instead Black recognizes that the pawn on b7 is doing good work, keeping White\u2019s queen out of c6, and he plays instead 8. \u2026 a5! <\/strong>And the a-pawn is unstoppable, for example 9. Bd1 b5 10. Kc6 a4 11. Kxb5 a3.<\/strong><\/p>\n\n\n\n

    Although this might seem like a rather special position, it addresses a very fundamental question: When do three pawns beat a bishop? Every chess player (above a certain level) knows the answer for rooks: two passed pawns on the sixth rank beat a rook. Yet I think most of us do not know the answer for bishops. From the above lines, a simple rule is: two passed pawns separated by a file on the sixth rank beat a bishop. (For example, pawns on a3 and c3 in the above lines.) If you have three connected passed pawns, they win if you can answer \u201cyes\u201d to all of questions 1-3, basically because you can then force a position where the pawns reach a3 and c3.<\/p>\n\n\n\n

    Note: I originally posted this entry with an incorrect position for the Dvoretsky study. (I had the king on a3 already!) My apologies to anyone who was confused.<\/p>\n\n\n\n

    Also, a followup to my last few posts: Teimour Radjabov won the World Cup, which is of some interest to U.S. chess fans because it means that Jeffery Xiong lost to the eventual tournament winner. Doesn’t change the fact that he lost, but maybe it’s a “moral victory.”<\/p>\n","protected":false},"excerpt":{"rendered":"

    Yesterday\u2019s chess party at Mike Splane\u2019s house turned into an endgame workshop, because both Mike Arne and Paulo Santanna came prepared with some beautiful positions. Ironically, both of them went to Spain this summer but they didn\u2019t play in the same events. Mike Arne played in the Benasque tournament, and showed us this deceptively simple […]<\/p>\n","protected":false},"author":80,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_et_pb_use_builder":"","_et_pb_old_content":"","_et_gb_content_width":"","footnotes":""},"categories":[37,10,25,16],"tags":[4391,2769,4392,536,2926,4394,4026,1487,3741,1638,4390,2165,768],"class_list":["post-5945","post","type-post","status-publish","format-standard","hentry","category-chess-clubs","category-endings","category-people","category-positions","tag-bishop-versus-pawns","tag-chess-party","tag-ellliott-winslow","tag-fork","tag-josiah-stearman","tag-knight-circuit","tag-knight-versus-pawn","tag-mark-dvoretsky","tag-mike-arne","tag-mike-splane","tag-paulo-santanna","tag-schematic-thinking","tag-stalemate"],"_links":{"self":[{"href":"https:\/\/danamackenzie.com\/blog\/index.php?rest_route=\/wp\/v2\/posts\/5945","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/danamackenzie.com\/blog\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/danamackenzie.com\/blog\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/danamackenzie.com\/blog\/index.php?rest_route=\/wp\/v2\/users\/80"}],"replies":[{"embeddable":true,"href":"https:\/\/danamackenzie.com\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=5945"}],"version-history":[{"count":4,"href":"https:\/\/danamackenzie.com\/blog\/index.php?rest_route=\/wp\/v2\/posts\/5945\/revisions"}],"predecessor-version":[{"id":5956,"href":"https:\/\/danamackenzie.com\/blog\/index.php?rest_route=\/wp\/v2\/posts\/5945\/revisions\/5956"}],"wp:attachment":[{"href":"https:\/\/danamackenzie.com\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=5945"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/danamackenzie.com\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=5945"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/danamackenzie.com\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=5945"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}