A beautiful study explained

by Sagar Shah
5/27/2016 – "One of the prettiest studies I have seen in recent times!" wrote Indian GM Abhijeet Gupta when he posted it on Twitter. Sagar Shah tried his hand at solving the position and gave it to a number of his students. The reaction was the same – everyone who was able to find the solution had a feeling of unadulterated joy. Here’s the study with a detailed explanation, including why King Leonidas is on the thumbnail!

ChessBase 14 Download ChessBase 14 Download

Everyone uses ChessBase, from the World Champion to the amateur next door. Start your personal success story with ChessBase 14 and enjoy your chess even more!


Along with the ChessBase 14 program you can access the Live Database of 8 million games, and receive three months of free ChesssBase Account Premium membership and all of our online apps! Have a look today!

More...

I was casually scrolling through my twitter feed when one of the tweets by a top Indian grandmaster caught my attention:

And below this was the following position:

White to play and win
This study is composed by A. Belyavsky and it won the First Prize in Korolkov Memorial Composing Tourney, 2008.

Believe us when we say that 2600+ grandmasters are not easily impressed by studies. And if they are, then you should definitely spend time trying to figure out the answer. Because if Abhijeet Gupta, who is rated 2654, says it is one of the prettiest studies he has seen in recent time, it has to be something special! Don't go further if you haven't tried solving. Without putting in the effort you won't be able to appreciate the beauty of the solution. 

The Solution

How do you go about solving a study? The first part is to overcome the mental block that the position should be difficult and the answer very pretty. You need to think as if this is a position that has arisen in your tournament game and you need to find the best way for White to proceed.

Material: 
Always begin with checking the material imbalance. Because without knowing who is materially better you can never be able to decide whether a particular sacrifice actually works. White is an exchange and a pawn, effectively three points down.

Giving the move to your opponent:
Most of the people now start with looking for moves for White. It could be perfectly reasonable to start with this, but instead I would recommend an intermediate step. Give the move to your opponent and check what would he do if it were Black to play. You can immediately see that Black would win the game either with 1...Qxh8# or 1...Rxd1. Two things can be gained from this exercise: you do not overlook your opponent's simple tactics and you know that this is a position where something drastic has to be done. 

Candidate moves:
It's now time to think about moves for yourself. A good rule of thumb is to always begin with checks and captures. This is extremely useful because checks and captures are usually forcing in nature and help you to understand what exactly is going on. The natural candidate moves are 1.Nxd8 and 1.Rxd8.
Before we start calculating these moves it would make sense to see if there are some checks in the position. To a trained mind, the first check that looks the most logical (even though you are giving up the queen!) is 1.Qf5+. The main reason why one would even consider this move is because 1.Qf5+ Kxf5 is met with 2.Ne3+. The fact that the knight is coming into the game with a tempo is good enough for us to at least bring this queen sacrifice into our list of candidate moves.

1.Nxd8, 1.Rxd8 and 1.Qf5+ are the three short listed candidate moves

We can quickly get rid of 1.Nxd8 because of 1...Rxd1. The queen is attacked on b1 and a mate is threatened on h1. There is no good reply that White has. He is just lost. 1.Rxd8 looks much more interesting. 1...Rxd1 now fails to 2.Qxd1! Rxd1 3.Rxd1 when White is an entire rook up. So 1.Rxd8 has to be either met with 1...Rxd8 or 1...Nxd8. 1...Rxd8 loses spectacularly to.... (I hope you have spotted it!)

2.Qf5!! of course! And the queen sacrifice works like magic as after 2...Kxf5 3.Ne3+ Kf6 4.Ng4+ Kf5 5.Ne7# This surely requires a diagram.

Pretty, isn't it!

If you are thinking that I am trying to just show you inferior variations and wasting your time, you are mistaken. It is 100% true that 1...Rxd8, in response to 1.Rxd8, was a bad move, but it helped us to find this beautiful idea that the black king can be checkmated with the two knights. This shows us how important it is to deflect the defenders of the e7 square, so that Ne7+ is possible at some point.

However, after 1.Rxd8 Black has the better move: 1...Nxd8! And after 2.Ne3 Nxc6! 3. Qf5+ Ke7 4.Nxd5 Rxd5 we reach the following position:


It is clear that White is better in this final position, but winning this would be extremely difficult. It is not at all easy to create a passed pawn and how White should proceed is unclear. So taking with the rook on d8 should only be played by White if he doesn't find something better.

We need to come back to the initial position and check our final candidate move 1.Qf5+. When I gave this position to Amruta (my wife) her first reaction was the move 1.Qf5+ and this would also be the reaction of many players because it looks like an answer to a study! However, no matter how hard you bang your head you will not be able to mate the black king as after 1...Kxf5 2.Ne3+ Kf6 3.Ng4+ Kf5 you have nothing more than a perpetual with 4.Ne3+. The e7 square is securely controlled. 

The queen sacrifice doesn't work and the maximum that White has is a perpetual check

We have exhausted all our candidate moves and we have still not found the answer. This brings us to the most important point that I would like to make in this article.

Persistance and not giving up:
Just like in life, you can succeed on the chess board only if you don't give up. I gave this position to many players and saw that after checking the three obvious moves mentioned above and not getting the answer, they pleaded with me to tell them the solution. With this approach you will never be able to experience the unadulterated joy of cracking a difficult problem. The tougher the challenge, the sweeter is the victory. So don't give up! Let's have a look at the initial position with a fresh mind:

We have to either play a forcing move or stop Black's threat of Qxh8. The possible moves could be 1.Rh7 but it fails to 1...Qf8+. Also 1.Qh7 has absolutely no threat and 1...Rxd1 looks strong. At this point when you are dying of frustration you might find the correct move: 1.Rg8!!

You have taken out the threat of Qxh8 and are now planning 2.Rg6+! fxg6 3.Qxg6#. This variation is very important and narrows down the opportunities that Black has at his disposal. He must take the rook. But what exactly is happening after 1...Qxg8. Well let's have a look at a skeleton position:

White to play. Can you spot the mate in three? Of course you can!

Now let's go back to the position where Black has just taken the rook with his queen.


Suddenly it all becomes clear! 1.Rg8 was a decoy sacrifice. We wanted the queen to be on g8. And why exactly? Because, White now sacrifices his queen with 2.Qf5+!! Kxf5 and as if all this sacrificing was not enough, here goes another piece into the fire with 3.Ne7+!!

The knight has to be taken because 3...Kf6 is met with 4.Nxg8+! (that was the point of 1.Rg8!!). After 3...Rxe7 4.Ne3+ a very weird kind of a position arises on the board.

The long knight on e3, single handedly fights the entire black army!

But look how a black piece falls on every move! 4...Kf6 5.Nxd5+ Kf5 6.Nxe7+ Kf6 7.Nxg8+

Two rooks and a queen gobbled up in three moves, this knight on g8 means business and he is not yet finished!

7...Kf5 8.Ne7+ Kf6 9.Nd5+ Kf5 10.g4#

Abhijeet Gupta was right! This was indeed a beautiful study!

We asked Abhijeet, who solved this study in three minutes, as to which was the point in the study that he loved the most? "One knight dominating the entire black army! That was just so beautiful! He was like a Spartan! Single handedly he defeated the entire black empire!" And that is the reason why we have King Leonidas on the thumbnail of this article!

[Event "I Prize,Korolkov Mem.Tourney "] [Site "?"] [Date "2008.04.14"] [Round "?"] [White "Belyavsky=A"] [Black "?"] [Result "1-0"] [SetUp "1"] [FEN "3q1k2/p2r1p1R/2p1n2K/N2rp3/8/5PP1/8/1Q1N4 w - - 0 1"] [PlyCount "23"] [EventDate "2008.??.??"] {White to play and win.} 1. Rh8+ Ke7 2. Nxc6+ (2. Rxd8 $2 Nxd8) 2... Kf6 {This is the position that we used as the starting point in this article. The actual composition had two moves before this point.} (2... Kd6 3. Nxd8 Nxd8 4. Ne3) 3. Rg8 $1 {Creating the threat of Rg6+.} (3. Nxd8 Rxd1 $19) (3. Rxd8 Nxd8 (3... Rxd8 $2 4. Qf5+ $1 Kxf5 5. Ne3+ Kf6 6. Ng4+ Kf5 7. Ne7#) 4. Ne3 Nxc6 5. Qf5+ Ke7 6. Nxd5+ Rxd5 $16 {There is no doubt that White is better but it is unclear whether he will be able to win this, because creating a passed pawn will not be easy.}) (3. Qf5+ Kxf5 4. Ne3+ Kf6 5. Ng4+ Kf5 6. Ne3+ $11) 3... Qxg8 (3... Rxd1 4. Rg6+ fxg6 5. Qxg6#) (3... Ng7 4. Rxd8 Rxd8 5. Nxd8 Rxd8 6. Kh7 $18) 4. Qf5+ $1 Kxf5 5. Ne7+ $1 Rxe7 6. Ne3+ Kf6 7. Nxd5+ Kf5 8. Nxe7+ (8. g4+ $2 Qxg4 $19) 8... Kf6 9. Nxg8+ Kf5 10. Ne7+ Kf6 11. Nd5+ Kf5 12. g4# 1-0

Source

Check out the ChessBase India website at www.chessbase.in – a responsive news page
that runs on notebooks, tablets, smartphones, practically everything.
You can also visit and "like" the ChessBase India Facebook page or Twitter account



Sagar Shah is an International Master from India with two GM norms. He is also a chartered accountant and would like to become the first CA+GM of India. He loves to cover chess tournaments, as that helps him understand and improve at the game he loves so much. He is the co-founder of the ChessBase India website.
Discussion and Feedback Join the public discussion or submit your feedback to the editors


Discuss

Rules for reader comments

 
 

Not registered yet? Register

Leonilo Leonilo 6/18/2016 02:25
I saw all the lines explained by Sagar Shah, but I thought I hadn't solved the problem yet.. and I still should use an engine to verify what happens after:
1.Rg8 Qf8+

I didn't see anything to force a mate or something decisive after 2.Rxf8 Nxf8. Of course white has material advantage but not an easy win.
azlan azlan 6/4/2016 02:39
Incredible. Chesthetica evaluates this study with a score of 4.702.

http://metalab.uniten.edu.my/~azlan/Other/chesthetica_eval.jpg

Considering traditional composed studies score, on average, just over 3.0 (see footnote 6 of the following reference), this is actually quite good indeed.

https://www.researchgate.net/publication/255907604_Evaluating_the_Aesthetics_of_Endgame_Studies_A_Computational_Model_of_Human_Aesthetic_Perception

Chesthetica is not programmed with all the constraints necessary to come up with something like this (assuming it even can) but that would be a whole other project dedicated to composing this type of study (assuming anyone is interested in funding such a thing).
Frits Fritschy Frits Fritschy 5/30/2016 04:13
elmerdssngalang,

I already explained why the pawn on a7 is necessary: it prevents a second solution with 3.Rxd8 Nxd8 4.Ne3 Nxc6 5.Qf5+ Ke7 6.Nxd5+ Rxd5 7.Kg7 Nd8 8.Qf6+ Ke8 9.Kg8. Without the a7 pawn, white could play the g-pawn to g5 and play Qf5, threatening f4. (The pawn on g5 prevents f6.) Black can't prevent that.
According to generally accepted standards, a second solution in a chess composition greatly diminishes its value.
The pawn on a7, by marching in the direction of a1, gives black just enough time to organize a defence: he can get the rook to the sixth rank without losing the e5 pawn. After that, he can either check the king on g6 and return to e6 or, if the white king moves to the h-file prior to the check, play a king or knight move, as the threats on f7 and f8 are gone. Which saves the composition ... or so it seems.
I understand IM Shah has some doubts about the correctness of the study, maybe because in this variation white can try the immediate 9... a5 10 Qf5!? Ke7 11 f4 f6 12 g4 Ne6 13 g5 fxg5 14 fxe5! and I doubt black can hold this. On 10... a4 11 f4 a3 12 Qc2 Rd4 there is a check on g4 after fxe5, but things are still unclear: 13 Qc5 Ne6 14 Qa7 Nd8 15 Qxa3 exf4 16 Qf8+ Kd7 17 Qc5 Ne6 18 Qa7+ Kd6 19 Kxf7 and as long as he keeps his pawn, white can still press.
We'll need an expert here to point out the forced draw!
elmerdssngalang elmerdssngalang 5/30/2016 12:06
htd, Qb6+ need not be played. the simple Qh7 twarts the mate threat. So the question remains: what's the practical role of the pawn on a7?
sicilian_D sicilian_D 5/30/2016 06:44
lovely!
htd2013 htd2013 5/29/2016 07:14
@billbrock, elmerdssngalang and trill
a7 pawn is required in the setup.
After 1. Rh8+ Ke7 2. Nxc6+ Kf6 3. Rg8 Ng7 4. Rxd8 Rxd8 5. Nxd8 Rxd8
White has to play 6. Kh7 to avoid mate by Rh8. Here alternate move like Qb6+ cannot be played by white due to a7 pawn!
malfa malfa 5/28/2016 08:58
John TVian, in the field of chess studies it is quite normal that the main variation not be the most stubborn from the losing side, but rather the brightest one esthetically. Compositions are different from real games.
As you suggest, 1...Td3 is also losing: the reason is that now 2.Txd8 Cxd8 is refuted bt 3.Cf2 which gains a decisive tempo by attacking the rook on d3.
chessbibliophile chessbibliophile 5/28/2016 07:14
The name of the composer was initially absent. So some one asked about it. Sagar Shah has done the needful by giving the complete solution with the name of the composer and the event. It’s nice of him to respond to reader feedback.
x_ileon@yahoo.co.uk x_ileon@yahoo.co.uk 5/28/2016 06:02
Sagar does a great job of educating us lesser chess mortals with his expos. As to someone complaining about not giving credit to the composer, did you bother reading the caption on the first diagram???
JohnTVian JohnTVian 5/28/2016 04:18
very wonderful indeed. But what if black were to play 1...Rd3? I think it would force white to take the queen on the next move with a much more drawn out game. Losing for black of course...
trill trill 5/28/2016 12:44
"May I inquire what role does the a7 pawn play?"

Does the a7 pawn need a role? Since this is a study and not a problem, the a7 pawn could be there to simulate a real game.
MichaelCiamarra MichaelCiamarra 5/27/2016 10:28
Yes, it is an extraordinary study! On May 12 I originally posted the study on Facebook and it got reposted and recycled many times by others. Curiously, this isn't the complete study as the first two moves of the original study were omitted, for the sake of clarity according to the source I used for my original post. I found the (abridged) position in a wonderful little book called, The Amazing Chess Adventures of Baron Munchausen by Amatzia Avni. Michael Ciamarra
fightingchess fightingchess 5/27/2016 08:25
This study is composed by A. Belyavsky and it won the First Prize in Korolkov Memorial Composing Tourney, 2008.

kudos to A. Belyavsky
malfa malfa 5/27/2016 08:03
A brilliant combination, but rather mechanical: no counterplay, no use of the a7 pawn, except in order to improve Black's chances in a very secondary attempt. It clearly looks as derived from real play, which can easily be expected by Alexander Beliavsky, whose fame as a great player is far superior to the one as a composer. I know this was Korolkov's heritage, but I prefer the classical school by far.
elmerdssngalang elmerdssngalang 5/27/2016 06:42
The solution is practically a forced line, which I believe makes the a7 pawn virtually irrelevant.
gmwdim gmwdim 5/27/2016 04:42
See, this is actual beauty in chess. Not that 3 move mate mumbo-jumbo that other writer is obsessed with.
billbrock billbrock 5/27/2016 04:05
The a7 pawn insures that Q v RN ending doesn't cook study
Frits Fritschy Frits Fritschy 5/27/2016 03:14
For people wanting more of the same (but with an unnatural position: see https://timkr.home.xs4all.nl/chess/dream.html.

I can understand why Sagar Shah truncated the original position, but to honour the composer, it should be mentioned that the white rook was on h7, the black king on f8, the c6 knight still on a5 and there was an additional black pawn at c6.
The complete solution (thanks to Van der Heijden's endgame study database) was (sorry, Dutch version: Queen = D, Rook = T, Knight = P):

1.Th8+ Ke7 2.Pxc6+

[2.Txd8? Pxd8; 2.Db4+? c5 3.Dh4+ Kd6 4.Pb7+ Kc7 5.Pxd8 Txd8 6.Txd8 (6.Th7 Txd1 7.Txf7+ T1d7 8.Kg6 Txf7 9.Kxf7 Pd4 10.De4 Kd6) 6...Txd8 7.Df6 Txd1 8.Dxf7+ Td7 9.Df6 Td5]

2...Kf6

[2...Kd6 3.Pxd8 Pxd8 4.Pe3]

3.Tg8!

[3.Pxd8? Txd1; 3.Txd8? Pxd8 (3...Txd1 4.Dxd1) 4.Pe3 Pxc6 5.Df5+ Ke7 6.Pxd5+ Txd5 7.Kg7 Pd8 8.Df6+ Ke8 9.Kg8 (9.Da6 Tc5 10.Dxa7 Tc6 11.Da4 Ke7 12.Db4+ Td6 13.g4 Pe6+ 14.Kh6 (14.Kg8 Pf8 15.Dc5 Pd7) 14...Pd4+ 15.Kh5 Pxf3 16.Db7+ Kf8 17.Dxf3 Tg6) 9...a5 10.g4 a4 11.Da6 (11.g5 a3 12.Da6 a2 13.Dxa2 Td6) 11...Td4 12.Db5+ Ke7 13.Dxe5+ Pe6 14.g5 Tf4]

3...Dxg8 4.Df5+ Kxf5 5.Pe7+ Txe7 6.Pe3+ Kf6 7.Pxd5+ Kf5 8.Pxe7+ Kf6 9.Pxg8+ Kf5 10.Pe7+ Kf6 11.Pd5+ Kf5 12.g4#

In the original study (as far as I could find, see http://www.arves.org/arves/index.php/en/awards/27-awards, the Korolkov-100 MT 2008 pgn file), the pawn on a7 was missing, and this made possible 3.Txd8 Pxd8 4.Pe3 Pxc6 5.Df5+ Ke7 6.Pxd5+ Txd5 7.Kg7 Pd8 8.Df6+ Ke8 9.Kg8 Tc5 10.g4 e4 (10...Kd7 11.Kf8 Td5 12.g5 Kc8 13.Ke8+-) 11.g5 Tc6 12.De5+ Te6 13.Db5+ Ke7 14.f4 Kd6 15.f5 Te7 16.Db6+ Kd7 17.Dd4+ Kc8 18.Kf8+-. I suppose Belyavsky later improved it by adding the pawn.
Nightplayer Nightplayer 5/27/2016 02:12
A very instructive study! and a good analysis
Mjguru Mjguru 5/27/2016 02:01
How can one not be romantic about chess!
elmerdssngalang elmerdssngalang 5/27/2016 02:00
May I inquire what role does the a7 pawn play?
Setne007 Setne007 5/27/2016 01:11
Great study and great article. So glad I persisted and found the solution (took me ten minutes though :-).
fartpants fartpants 5/27/2016 12:52
Why not give credit to the composer?
karpovich karpovich 5/27/2016 11:55
aside from the study and its beauty the article is classic
psamant psamant 5/27/2016 11:49
Forget three minutes, I didn't find it at all! The best thing in this puzzle was how the knight moves forward to eliminate the queen and then retraces the same steps backwards to finally deliver the mate!
faserbuendel faserbuendel 5/27/2016 10:51
Wow, very beautiful, indeed. There's no way I would've found it in just 3 minutes.
1