Corresponding Squares

Sort:
Avatar of KevinOSh

I started reading about Corresponding Squares in pawn endgames but I am confused by it.

I tried the lesson https://www.chess.com/lessons/pawn-endings-beginner-to-expert/corresponding-squares but it just feels like very strange guessing game where every king move except the least logical one is wrong.

Does anyone know of a good user friendly explanation of this topic, which doesn't assume the reader is already a chess master?

Avatar of tygxc

https://en.wikipedia.org/wiki/Corresponding_squares 

Avatar of Ziryab

Mark Dvoretsky is clear: "Corresponding squares are squares of reciprocal zugzwang. ... The most commonly seen cases of corresponding squares are: the opposition, mined squares, and triangulation." From Dvoretsky's Endgame Manual, 5th ed. (2020), 15.

The difficulty comes when one is attempting to calculate a board full of mined squares, such as in the famous study by Charles Locock published in British Chess Magazine in 1892. See https://www.chess.com/forum/view/endgames/how-do-you-identify-corresponding-squares

The solution to a position George Walker thought was drawn (see his A New Treatise on Chess, 1841) is simpler. See http://chessskill.blogspot.com/2022/06/corresponding-squares.html

 

Avatar of KevinOSh

 

Oh, so I do already have some experience of corresponding squares in form of opposition. That felt unnatural to me at first but after some practice the moves became easy.

It is the more complex examples, such as that study by Nikolay Grigoriev, 1924, which is completely bewildering to me. The a1 square is the corresponding square of b5, and the c1 square corresponds with the d4 square, but how is anyone supposed to know this?

@Ziryab thank you, the poster of https://www.chess.com/forum/view/endgames/how-do-you-identify-corresponding-squares had pretty much the same query and puzzlement as I do with this.

I think probably what I need is to practice some examples which are the next level up from opposition, and gradually this might make more sense to me. But this is just a hunch.

Avatar of Ziryab

Calculating every pair of corresponding squares is time consuming and seems difficult. I have it on my list of things to study more. I've stalled in my reading of many endgame books when they get to the point of breaking down a complex study.

I suspect understanding distant diagonal opposition will help.

Avatar of bigD521
MovesThatMesmerize wrote:

If they are basing this lesson on white to move, then why are they saying this, "when the White king is on f4, the Black king must be on h5 to defend the g4-pawn"

 

There is no "must" in a losing position. If it is white to move, white can win. 

Well it may be that they are just pointing that out and white must be aware of this. Understanding this, white must plot its course so this does not happen, Then white wins.                 It is whites move as stated in bold letters.

In your example black can draw with no errors.   For what it is worth,, the correct response for black on move 6 is ..... Kg6 Your move failed to take into consideration the key square d6 and it's corresponding squares, e2 and c5. 6........ Kh6 allows white to capture b7  in 5 moves.

@KevinOSh Peanut galley comment,, I would suggest starting with the wiki link.  Then do a search and you find a number of videos/tutorials of which I would select shorter versions because they should be more basic. Progress upwards in complexity from there.  In videos some speak fast, don't forget you can change the speed via the little gear symbol.

Avatar of llama36

I did a blog on it once.

It's not hard if you can understand one pair of squares and then work backwards from there.

For example in the position below, for your move pick up white's king and put it anywhere on white's side of the board.

-

-

The only way to draw is Ke2.

Therefore you found one pair of squares i.e. e2 and e4.

The trick now is to understand in the position below (this time black to move) that when black "touches" e4 (in the sense that it stands next to the e4 square) white's king is forced to "touch" e2.

-

-

Now there's only 1 more trick to understand. When black's king lands on d4, white's king must land on d2... which means when black touches BOTH e4 and d4, white must touch BOTH e2 and d2. In other words, in the image below, if the black king touches red and green, white must also touch red and green.

New puzzle. For white's move, put white's king anywhere on the 1st rank. Which squares draw?

The only squares on the 1st rank that draw are d1 and e1 because they touch both red and green (just like black's king).

-

And now you know enough to build any complex crazy array of corresponding squares. Sometimes there are 3 colors, or 4, etc, but you start very simply, by finding a pair of squares, then work outwards from there.

Avatar of llama36

I found an old image from when I did a blog.

In the image below it's easy to see why this move wins for white... it wins because it's touching red and green, and black's king cannot respond by also touching red and green. After black moves, white will win by landing on the same color (or in the case black plays something like Kh7, white can walk the king to e6 and promote).

 

Avatar of llama36
Ziryab wrote:

Mark Dvoretsky is clear: "Corresponding squares are squares of reciprocal zugzwang. ... The most commonly seen cases of corresponding squares are: the opposition, mined squares, and triangulation." From Dvoretsky's Endgame Manual, 5th ed. (2020), 15.

The difficulty comes when one is attempting to calculate a board full of mined squares, such as in the famous study by Charles Locock published in British Chess Magazine in 1892. See https://www.chess.com/forum/view/endgames/how-do-you-identify-corresponding-squares

The solution to a position George Walker thought was drawn (see his A New Treatise on Chess, 1841) is simpler. See http://chessskill.blogspot.com/2022/06/corresponding-squares.html

 

I see you also used colors.

IMO books made the topic extremely difficult to understand... but maybe that's because I was a beginner at the time. I later figured this out this color method on my own (maybe 15 years later heh) and then it made sense to me.

Probably first, a player should memorize some patterns from basic king and pawn endgames... because if you can't identify one pair of squares then you definitely can't understand more complex positions.

Avatar of bigD521

@llma36  Thank you for your post.

Avatar of llama36
MovesThatMesmerize wrote:

This is how these sites confuse and obfuscate things, then they charge you money based on your laziness so you can watch them connect the dots. I question the merits of these paid features.

 

First, it is black to move, not white. I wish they would tell you that upfront so you aren't looking for a white move in the beginning. Now, we got that out of the way, we can now look at the position in question, FROM THE RIGHT PERSPECTIVE!!! 

 

 

With black to move, the king can't be too far away from the opposing pawns or else they will queen with the help of the king. So, this is why Kg8 needs to be played. Similarly, 6...Kg6 needs to be played, NOT Kh6 as seen in this line below.

 

 

 

I had that position in my blog.

Your error was black's move 6. In my diagram, white's king landed on a "2" so black can draw with 6...Kg6 (which is also a 2). (oh, you mention this).

Notice how "4" on black's side touches 9, 1, 3, and 5... just like "4" on white's side touches 9, 1, 3, 5. That's how you build diagrams like this.

But you start small. Notice d4 and e3 are marked 1 and 2. What is the only square that touches 1 and 2? well, the squares d3 and g7 touch 1 and 2. So you mark them with 3... and so on...

 

Avatar of llama36
bigD521 wrote:

@llma36  Thank you for your post.

No problem happy.png

I think I probably didn't understand this topic at first because books started way too complex, like the number diagram I show.

If a person starts simply (like the colors) then I think it's a lot more accessible.

Avatar of llama36

LOL

Chess.com uses that position for the first (and only?) lesson on corresponding squares?

What kind of moron would make that lesson 1... good lord.

I came up with the exercises in my post #9 in under 1 minute... if the person making the lessons actually understands the topic it doesn't take much effort to make it accessible...

Avatar of KevinOSh

I found a Naroditsky video that explains the ideas pretty clearly with examples