|
< Earlier Kibitzing · PAGE 2 OF 2 ·
Later Kibitzing> |
| Sep-23-16 | | NBZ: Tricky: Black must go for stalemate. Qe2 was extremely tempting, because of Kxf4 Qe3+ drawing straight away. But after Kd4! it is hard to find a good knight check. Ne6? of course loses to Qxe6. Qd3+ Kc5 Qe3+ Kc6 and the white king escapes. Somewhat reluctantly I switched to Nd5+ Kd4 Qf2+. At first this seemed bad because it looks like after Kxd5 the king blocks the bishop and breaks the stalemate. But that's only temporary because after Qd4+ Kxd4 reestablishes the stalemate. The king can't run away after Qf2+: if Kc4 Nb6+ picks up the queen. Kd3 Qe3 kc2 qe2+ is a perpetual. Ke5 Qxb2+ Ke4? Nf6+. |
|
| Sep-23-16 | | mqhelisi: i struggled coz i couldnt find a closed ended continuation to 50..Nd5+
51. Kd3 i got spinning in the whirlwind of possible combinations but this line i saw it though |
|
| Sep-23-16 | | YouRang: Friday 50...?
click for larger view
This *has* to be a drawing tactic. I stared at it a good while before I noticed that white's queen and bishop has my (black) king immobilized. So: - The objective must then be stalemate.
- My first move needs to be check, or else white can avoid stalemate. I tried a few checks, but the only promising try is <50...Nd5+>, because 51.Bxd5 maintains the stalemate (I just have to force white to take my queen to escape checks, which shouldn't be hard). I'm guessing <51.Kd4>, hoping to capture the knight with the king, which breaks the stalemate. Again, it took me while to realize that I can let the king have the knight! I just have to keep checking the king such that neither the bishop or queen can break free. So <51...Qxg4+!>
click for larger view
Now, if <52.Kxd5> then <52...Qd4+ 53.Ke6> trying to block the bishop <53...Qe5+ 54.Kf7 Qf6+ 55.Ke8 Qf8+!>
click for larger view
The stalemate is forced.
~~~~
Of course, white shouldn't take the knight, but try to escape with <51.Kd3>. Again I stared at it for a while. Stalemate tactics force you to think opposite of normal chess. My biggest problem is that I have to get rid of both my knight and queen, and black doesn't want them! I'm thinking that in this case, it's more about perpetual check than stalemate. So, assuming that white won't take my knight, I'll have to use it to keep white in check. So, <52...Qf3+>
click for larger viewCan the king escape checks? Not with 53.Kc4? Nb6+ forking K+Q  But without the escape at c4, I'm pretty sure I can prevent the king from getting away. - If 53.Kd4, then ...Qe3+, preventing escape to c5. - If 53.Kc2, then ...Qe2+, forcing the king to the first rank, where I can keep him locked up. - If 53.Kd2, then ...Qe3+, forcing to first rank (or c2), again locking up the king. Great twisty puzzle! |
|
| Sep-23-16 | | diagonalley: hmmmm... clearly stalemate is the only chance of salvation... but seeing how to force it is PDH |
|
| Sep-23-16 | | AlicesKnight: I saw some lines close to <YouRang>'s comments (thanks for this), but was not completely convinced that White could not escape. |
|
| Sep-23-16 | | Boerboel Guy: <WDenayer: I don't understand anything. What happens after 51.Bxd5 ?> Black just checks next to the King to force the stalemate. |
|
| Sep-23-16 | | offramp: An example of the dumbing-down of chessgames.com. This used to be a Thursday puzzle, now it's a Friday puzzle. |
|
Sep-23-16
 | | Once: Ooh, that's a good puzzle! I more or less got there, but I needed to turn to Fritz to understand all of it. The point is that black needs to throw his knight and queen away whilst still keeping his king in a stalemated position. Meanwhile white needs to find somewhere to park his king where it can't be checked constantly. He also has to watch out for sneaky knight forks. If he can get one queen move in, he would be able to get off the seventh rank, negate the stalemate and win easily. But where can white park his king? The bottom of the board isn't going to work, so we need to think about the top of the board. One plan might be for white to exchange off the knight and then run his king to the eighth rank, say anywhere from a8 to d8. Then he might be able to block the black queen checks with his own queen. But how does the white king get to the eighth rank? If he tries to get there via a6 or c6, black will jump in with Qb6+ and the queen has to be taken. Something like this: 50... Nd5+ 51. Bxd5 Qf2+ 52. Kd3 Qd2+ 53. Kc4 Qe2+ 54. Kc5 Qe3+ 55. Kc6 Qb6+ 
click for larger viewAnd if white tries to run to the kingside, black can force him to e8 and give his queen away on f8, as in YouRang's excellent post. Clever clever stuff. |
|
Sep-23-16
 | | patzer2: <Once: Ooh, that's a good puzzle! I more or less got there, but I needed to turn to Fritz to understand all of it.> Here's my look with Deep Fritz 15:
<50...Nd5+!>
not 50... Qe2+? 51. Kd4 (51. Kxf4?? Qe3+ 52. Kxe3 + =) 51... Qd3+ 52. Kc5  (26.15 @ 19 depth, Deep Fritz 15) <51. Kd4>
51. Kd3 Qf3+ 52. Kd4 (not 52. Kc4?? Nb6+ ) 52... Qf2+ 53. Kd3 Qe3+ 54. Kc2 Qe2+ 55. Kc1 (not 55. Kb1?? Qe1+ 56. Kc2 Nb4#) 55... Qe1+ 56. Bd1 Qe3+ 57. Kb1 (not 57. Kc2? Nb4+ 58. Kb1 Qd3+ 59. Bc2 (59. Kc1 Nxa2#; 59. Ka1 Qxd1#) 59... Qxc2+ 60. Ka1 Qc1#) 57... Qd3+ 58. Kc1 Qe3+ 59. Kb1 Qd3+ 60. Kc1 Qe3+ = 51. Bxd5 Qd2+ 52. Kf3 (52. Ke4 Qe3+ 53. Kxe3 =) 52... Qf2+ 53. Ke4 Qe3+ 54. Kxe3 = <51... Qf2+ 52. Kxd5> not 52. Ke4? Nf6+ or 52. Kc4?
Nb6+ .
52. Ke5 Qe3+ 53. Kxd5 Qd4+ 54. Kxd4 = transposes to the game. <52... Qd4+ 53. Kxd4 = (stamemate)> 53. Kc6 Qb6+ 54. Kd5 Qd4+ 55. Ke6 Qf6+ 56. Kd5 Qd4+ = 1/2-1/2 is a draw by perpetual check. P.S.: White has at least six opportunities to blunder and loses in this maze of variations. One example is the mate-in-four after 51. Kd3 Qf3+ 52. Kd4 (52. Kc4?? Nb6+ ) 52... Qf2+ 53. Kd3 Qe3+ 54. Kc2 Qe2+ 55. Kc1 (55. Kb1?? Qe1+ 56. Kc2 Nb4#) 55... Qe1+ 56. Bd1 Qe3+ 57. Kc2? (instead 57. Kb1 = holds) in the diagram below: 
click for larger viewSolution: 57...Nb4+ 58. Kb1 Qd3+ 59. Bc2 (59. Kc1 Nxa2#; 59. Ka1 Qxd1#) 59... Qxc2+ 60. Ka1 Qc1# |
|
| Sep-23-16 | | WorstPlayerEver: No one wants to win this way except Galfand. |
|
Sep-23-16
 | | Peligroso Patzer: <Prugno: This is the kind of finish that makes me want to agree with Nigel Short on changing the stalemate rule. A spectacular swindle by Gelfand, for sure, but by no means a fair result.>
[from May-24-15]
SirRuthless and Calar have already commented in support of the fairness of the result in this game, and I endorse their thoughts. With regard to the stalemate rule more generally, if a stalemate were a win, there would certainly be many more decisive results. For one thing, essentially every K + P vs. lone K ending would be a win (the only exception being positions in which the defending king could capture the pawn). In many other cases with more material (especially other pawn endings), if a stalemate were made a decisive result, there would be a routine win in many positions in which under the current rule the position is either drawn or, if winnable, requires good - and sometimes highly refined - technique. I find the technical aspects of many of these endings extremely elegant. My bottom line on this issue is that the proposed change in the stalemate rule would, for the sake of increasing the number of decisive results and serving a faux sense of fairness, substantially undermine the incredible elegance of chess. I like Nigel Short (who, it goes without saying, is a hugely stronger player than I), and I often agree with his sometimes controversial ideas, but on this point I think he’s completely wrong.. |
|
| Sep-23-16 | | Clodhopper: Wow, I finally "got" a Friday challenge. Only because it is so obvious that Black's goal must be to force stalemate. On most Fridays, I stumble at recognizing the motif: mating attack? material gain? creating an unstoppable passed pawn? Here the situation is so stark that there is only one line of investigation..... |
|
Sep-23-16
 | | PawnSac: < A spectacular swindle by Gelfand, for sure, but by no means a fair result.> Dmitri probably went home and cried on his pillow over this one |
|
Sep-23-16
| | PawnSac: < Boerboel Guy: <WDenayer: I don't understand anything. What happens after 51.Bxd5 ?> Black just checks next to the King to force the stalemate. > as long as he doesn't allow his Q to be captured by the white B or Q (thus breaking the stalemate net). If white refuses to capture the Q black can draw with endless checks |
|
Sep-23-16
 | | agb2002: Black has a knight for a bishop and five pawns. White threatens Kxf4 and Qd8+ followed by Qg5+ (Kh7 Qh4+). Black has Qe2+, Qg1+, Qg3+, Qh3+ and Nd5+.
In the case of 50... Qg3+:
A) 51.Kd2 Qd3+ 52.Kc1 (52.Ke1 Qe2#) 52... Ne2#.
B) 51.Ke4 Qd3+ 52.Ke5 (52.Kxf4 Qe3+ 53.Kxe3 stalemate) 52... Qe3+ 53.Kf6 Nh5+ 54.Kf7 + -. -----
In the case of 50... Nd5+:
A) 51.Bxd5 Qd2+ 52.Kf3 (52.Ke4 Qe3+ 53.Kxe3 =) 52... Qf2+ and stalemate in two moves. B) 51.Kd3 Qf3+
B.1) 52.Kc2 Qe2+ and Black seems to achieve perpetual. B.2) 52.Kd2 Qe3+ as B.1.
B.3) 52.Kc4 Nb6+ wins the white queen.
B.4) 52.Kd4 Qe3+ 53.Kxd5 (53.Kc4 Nb6+ as above) 53... Qe4+ 54.Kc4 Qb4+ 55.Kc6 (55.Kd5 Qe4+ repeats moves) 55... Qb5+ 56.Kc7 Qb6+ 57.Kc8 Qb8+ =. C) 51.Kd4 Qxb2+
C.1) 52.Kc5 Qc3+
C.1.a) 53.Kxd5 Qd4+ and perpetual (Qb6+, Qd4+, Qf6+) or stalemate. C.1.b) 53.Kb5 Qb4+ 54.Kc6 (54.Ka6 Qb6#) 54... Qb6+ transposes to C.1.a. C.2) 52.Ke4 Nf6+ wins the queen.
C.3) 52.Qd3 Qc3+ is similar to B.1.
-----
I'd play 50... Nd5+. |
|
Sep-23-16
| | PawnSac: look at the picture of Dimitri in his bio. The expression is priceless... "I can't believe i let him draw that game!" |
|
| Sep-23-16 | | WorstPlayerEver: Gandalf's Twist |
|
Sep-23-16
 | | kevin86: When I saw that the game ended in a draw, I saw the stalemate trap! |
|
| Sep-23-16 | | YouRang: Looking a few moves earlier, Gelfand (black) was facing a totally lost game. However, he came to this position on move 48:
click for larger view
I believe that it is here that Gelfand (black) formulated the stalemate trick, noticing that the white queen and bishop had his king cornered (which may not have been obvious to Jakovenko at the time). For any hopes of stalemate, black needed to unload his pawn, so he played the wiley <48...g4!>. White ate the poisoned pawn, and the game was suddenly drawable. White could have won with <49.Kg3>. And if <49...hxg3>
click for larger view
White must resist taking the knight! If he does: <50.Kxf4? h2>
click for larger view
White has no way to both (1) avoid stalemate and (2) stop the pawn. ~~~~
After <49...hxg3> (second diagram above), white has different options that win, but from a human standpoint, the most logical is <50.Qa4!>, which breaks the stalemate trap while threatening to exchange queens. If <50...Nh5+>, white should still avoid taking the pawn (51.Kxh3 Qf3+, which makes it more difficult, and even dangerous), and instead play <51.Kh4!>
click for larger view
- If 51...Qxa4, then 52.Bxa4 h2 53.Bc6 stops the pawn and wins. - If 51...Qd8+, then 52.Kxh5 is safe. Black can issue some checks, but white can escape. - If 51...Qf8, then 52.Kxh5 is okay despite 52...Qxf5+ (52.Kh4 and the white Q prevents Qf4+, which is required to keep the king from escaping checks). This game is a great example of how tricky it can be to win a won game. |
|
| Sep-23-16 | | YouRang: I do wonder when Jakovenko realized he was in a stalemate trap. The fact that he ended the game by simply taking the queen (53.Kxd4) makes me wonder if he didn't realize it until it was over. |
|
| Sep-23-16 | | TheFocus: Wonderful stalemate trick!! |
|
| Sep-23-16 | | RandomVisitor: After the improvement 37...Rxd6 black might have drawing chances:
click for larger viewKomodo-10.1-64bit:
+0.35/35 38.Qxh5 Nf7 39.Re1 Nf6 40.Qc5 Rd2 41.Qc3 Rd3 42.Qc4 Rd4 43.Qe6 Rd6 44.Qf5 Rd2 45.b4 Qd4 46.a3 Nd6 47.Qc5 Qxc5 48.bxc5 Nde8 49.a4 Rc2 50.c6 Kf7 51.Ra1 Nc7 52.a5 Na6 53.g4 Rc5 54.f4 g6 55.Kf2 Nd5 56.Bxd5+ Rxd5 57.h4 Rc5 58.Rb1 Rxa5 59.Rb7+ Ke6 60.Rg7 |
|
| Sep-23-16 | | devere: It is easy to see that stalemate is Black's only hope, but not that easy to see that 50...Nd5+! is the only move to force it. Congratulations to Boris Gelfand for seeing this over the board. |
|
| Sep-24-16 | | YouRang: I see that this game brought up the debate regarding the proposal for the rule change to let stalemate be counted as a win. IMO, this game demonstrates why the stalemate rule is correct as it is. What a shame it would be to deprive chess of this wonderful dimension of chess strategy that comes about because one side can earn a draw by forcing the other side to fail to achieve checkmate. A draw is the only natural way to score the game.
It's like (American) football. There's a big difference between getting the ball in the endzone and not getting it into the endzone -- even though the actual difference could be an inch. |
|
Mar-12-17
 | | Dionysius1: Agreed <YouRang>. And anyway, if stalemates become wins for the attackers, won't the same be true for "fortresses"? Professionals in particular should have no reason to grumble about stalemates being draws - isn't it just one of their many skills to be able to construct draws if under attack, or prevent them if they're attacking? |
|
 |
|
< Earlier Kibitzing · PAGE 2 OF 2 ·
Later Kibitzing> |
Kibitzing
