Dang! I just took 4. Rg6 on faith. But how can you say the initial position is a win? I strongly suspect 1... h5 was a mistake. Black shouldn't be looking for improvements. "Progress" is only for the strong side.
As it was shown in the video 1...Rb7 2.Kg3 and White has the winning plan Kg4-h5 etc. If Black doesn't play 1...Rb7 then White will play Rf7 taking control of the 7th rank himself.
Yes, that plan wins. (I assume "...Rg7" is a typo for "...Rh7) So after 1...h5 2.hxg6 Rb7 3.Kg3 Ke5 we reach the critical position of 3:10 4.Rf8! (4.Rf3! also wins, but I have no idea why this is superior to say 4.Rf1) 4...Rh7 5.Kg4 Rxh6 6.Kg5! Ra6 and now either Matt's idea 7.Re8+! or the cute 7.Rf5+! win.
OK, here's why 4.Rf3! also wins and 4.Rf1? doesn't. (4.Rf3! is not as straightforward as 4.Rf8!, but still interesting.) Let's start with 4.Rf1? (Nunn Convention) 4...Rh7 5.Kg4 Rxh6 6.Kg5 Rh8! (frontal defense) 7.Re1+ Kd6! 7.h5 Rg8+! 8.Kf6 Rf8! 9.Kg7 Rf2! 10.h6 Rg2+! 11.Kh7 Kd7! (to control the 8th rank) and the Black king is one file too close....
After 4.Rf1?, Black was able to switch from frontal to rear defense. But 4.Rf3! makes a Lucena-like bridge across the rear defense possible. So: 4.Rf3! Rh7 5.Kg4! (Nunn Convention) 5...Rxh6 6.Kg5! Rh8 (frontal defense) 7.Re3+! Kd6 7.h5 Rg8+ 8.Kf6 Rf8 39.Kg7 Rf1 10.h6 Rg1+ 11.Kf6 Rf1 12.Kg5! Rg1+ 13.Kh4! breaks the rear defense. Black can switch back to the frontal with 13...Rg8 but 14.Kh5 is one of several winning moves. Amazingly tough. Google "shredder endgame tablebase" to check me.
I am using database (not brains, as you honorable people are). I would have guessed that this position is a dead draw. However, I think it's won! 1...h5!? (Black's only chance for improvement) 2.gxh6! is definitely +- Now 2...Rb7 3.Kg3 Ke5 is all best per the database. White throws away the win with 4.Rg6? and in fact has *two* winning moves!
I spent a few days looking at this and I'm 90%+ convinced it's a draw. In fact, I lost a night's sleep. BTW, 4:42 is a win for white, and a simpler way after Rf7+ Ke6 is | Kg6 Rg8+ | Rg7 Rh8 | h7 etc. White can even win by pushing h7 immediately. (h7 Kf4 | Rb7! is cute)
Thank you for bringing chess back into my life. I used to be pretty good a thousand years ago. ;)
Your comments regarding double h pawns versus a single h pawn are very instructive. Thanks for that. However, at 05:21 black should not play -Ke7? I am convinced that he rather should play -Kf5 which is a draw after a) Rg7 Kf6 and b) Ra6 Rb7. In "Rook endings" the position no. 152, which is the same as at 03:03, is stated as a win for white. If the play for white can not be improved, then this assumption, to me, must be wrong. Thus, the starting position is a draw.
The line in the book does not mention your ...Kf5 at all, only ...Ke7. So this is brand new territory. Thanks for your original post.
Might there be some winning chances for White after ..Kf5 Rg7 Kf6 Rg1 !? Or maybe just ...Kf5 Rg1 !? It seems like Black's K is going to be cut off at a distance of two files unless he plays ...Ke7-f8 as Kopaev did.
If white plays the rook to g1, the black king just stays at f5 or f6 and after Rf1+ black plays Ke6. As white can´t trade rooks he has to move his rook from the f-file and then the black king returns to f5. Further, I am rather convinced that Rg7, at 05:24, is a mistake. Ra6 is stronger. Now Kg6 is a threat. After -Rf6, Ra7 Ke6 h7 will win for white. With the previous defence in mind, black should play Kf6 instead of Kf8, at 05:55, with a draw. Finally, at 03:14, Rf8 must be better than Rg6.
Tarrasch eh? That is funny. Tarrasch thought that the Vancura position (W: Pa6, Ra8, Kg2; B: Ra1, Kg7) was a win for White... we now know of course it's a draw.
Jessica - Thx for your compliment! Fortunately I have made several other videos so I won't have to give my cell # out just yet :-p
Yes! interesting and instructive!
SuperPeristilo 1 month ago
very nice,thank you
60tat 1 year ago
Dang! I just took 4. Rg6 on faith. But how can you say the initial position is a win? I strongly suspect 1... h5 was a mistake. Black shouldn't be looking for improvements. "Progress" is only for the strong side.
rickideemus 2 years ago
1...h5 is pretty much the best try.
As it was shown in the video 1...Rb7 2.Kg3 and White has the winning plan Kg4-h5 etc. If Black doesn't play 1...Rb7 then White will play Rf7 taking control of the 7th rank himself.
GreenCastleBlock 2 years ago
Maybe Rf8 instead of Rg6? And on ...Rg7, sacrifice a pawn with Kg4-g5 and Re8+, is this what you had in mind Bill?
GreenCastleBlock 2 years ago
Yes, that plan wins. (I assume "...Rg7" is a typo for "...Rh7) So after 1...h5 2.hxg6 Rb7 3.Kg3 Ke5 we reach the critical position of 3:10 4.Rf8! (4.Rf3! also wins, but I have no idea why this is superior to say 4.Rf1) 4...Rh7 5.Kg4 Rxh6 6.Kg5! Ra6 and now either Matt's idea 7.Re8+! or the cute 7.Rf5+! win.
billbrock1958 2 years ago
OK, here's why 4.Rf3! also wins and 4.Rf1? doesn't. (4.Rf3! is not as straightforward as 4.Rf8!, but still interesting.) Let's start with 4.Rf1? (Nunn Convention) 4...Rh7 5.Kg4 Rxh6 6.Kg5 Rh8! (frontal defense) 7.Re1+ Kd6! 7.h5 Rg8+! 8.Kf6 Rf8! 9.Kg7 Rf2! 10.h6 Rg2+! 11.Kh7 Kd7! (to control the 8th rank) and the Black king is one file too close....
billbrock1958 2 years ago
After 4.Rf1?, Black was able to switch from frontal to rear defense. But 4.Rf3! makes a Lucena-like bridge across the rear defense possible. So: 4.Rf3! Rh7 5.Kg4! (Nunn Convention) 5...Rxh6 6.Kg5! Rh8 (frontal defense) 7.Re3+! Kd6 7.h5 Rg8+ 8.Kf6 Rf8 39.Kg7 Rf1 10.h6 Rg1+ 11.Kf6 Rf1 12.Kg5! Rg1+ 13.Kh4! breaks the rear defense. Black can switch back to the frontal with 13...Rg8 but 14.Kh5 is one of several winning moves. Amazingly tough. Google "shredder endgame tablebase" to check me.
billbrock1958 2 years ago
I am using database (not brains, as you honorable people are). I would have guessed that this position is a dead draw. However, I think it's won! 1...h5!? (Black's only chance for improvement) 2.gxh6! is definitely +- Now 2...Rb7 3.Kg3 Ke5 is all best per the database. White throws away the win with 4.Rg6? and in fact has *two* winning moves!
billbrock1958 2 years ago
3:02 is a tablebase loss (2.gxh6 en passant wins by force).
billbrock1958 2 years ago
I spent a few days looking at this and I'm 90%+ convinced it's a draw. In fact, I lost a night's sleep. BTW, 4:42 is a win for white, and a simpler way after Rf7+ Ke6 is | Kg6 Rg8+ | Rg7 Rh8 | h7 etc. White can even win by pushing h7 immediately. (h7 Kf4 | Rb7! is cute)
Thank you for bringing chess back into my life. I used to be pretty good a thousand years ago. ;)
rickideemus 2 years ago
Your comments regarding double h pawns versus a single h pawn are very instructive. Thanks for that. However, at 05:21 black should not play -Ke7? I am convinced that he rather should play -Kf5 which is a draw after a) Rg7 Kf6 and b) Ra6 Rb7. In "Rook endings" the position no. 152, which is the same as at 03:03, is stated as a win for white. If the play for white can not be improved, then this assumption, to me, must be wrong. Thus, the starting position is a draw.
anylen61 2 years ago
The line in the book does not mention your ...Kf5 at all, only ...Ke7. So this is brand new territory. Thanks for your original post.
Might there be some winning chances for White after ..Kf5 Rg7 Kf6 Rg1 !? Or maybe just ...Kf5 Rg1 !? It seems like Black's K is going to be cut off at a distance of two files unless he plays ...Ke7-f8 as Kopaev did.
GreenCastleBlock 2 years ago
If white plays the rook to g1, the black king just stays at f5 or f6 and after Rf1+ black plays Ke6. As white can´t trade rooks he has to move his rook from the f-file and then the black king returns to f5. Further, I am rather convinced that Rg7, at 05:24, is a mistake. Ra6 is stronger. Now Kg6 is a threat. After -Rf6, Ra7 Ke6 h7 will win for white. With the previous defence in mind, black should play Kf6 instead of Kf8, at 05:55, with a draw. Finally, at 03:14, Rf8 must be better than Rg6.
anylen61 2 years ago
At 3:14: Rf8? Rh7 picks off the pawn, no?
rickideemus 2 years ago
...Kf5 is indeed a tablebase draw.
billbrock1958 2 years ago
Your narrow conclusion is correct (...Kf5 draws) but your broad conclusion appears to be incorrect: I believe the initial position is won for White!
billbrock1958 2 years ago
a couple of times you mentioned that the position was not mentioned in the book. might i ask what 'the book' is?
Also, great vid as per usual.
satan130 2 years ago
Thanks. Gregory Levenfish & Vassily Smyslov "Rook Endings"
GreenCastleBlock 2 years ago
Interesting. It's amazing how complicated a position with only 4 pieces and 3 pawns can be.
flamingmonkey923 2 years ago
Great vid Castle!
jrobichess 2 years ago
excellent analysis!
Sorcerer88 2 years ago 2
Very true. Love his analysis, always instructive and to the point.
ex0duzz 2 years ago
great video
rolfch1p 2 years ago
Tarrasch said "All Rook Endgames are Drawn"-
But he didn't mean that literally.
Matt- I love the sound of your voice.
jessicafischerqueen 2 years ago 3
Tarrasch eh? That is funny. Tarrasch thought that the Vancura position (W: Pa6, Ra8, Kg2; B: Ra1, Kg7) was a win for White... we now know of course it's a draw.
Jessica - Thx for your compliment! Fortunately I have made several other videos so I won't have to give my cell # out just yet :-p
GreenCastleBlock 2 years ago
not always... but always drawn?!? who said that?
fezzjr 2 years ago
Not sure, but it must have been a famous person or people wouldn't still be repeating it.
GreenCastleBlock 2 years ago
very instructive and entertaining endgame.... rook endgames always are =D
fezzjr 2 years ago
'Always'? Someone once said all rook endgames are drawn :-|
GreenCastleBlock 2 years ago