Sunday, January 27, 2013

The non lazy approach

Because of my latest post mr. Z accused me of lazyness. He sensed that quite right. In stead of taking the lead and working out the themes, I immersed myself in a trial and error approach towards this endgame. In that way, I make the same errors over and over again due to the lack of "density" of problems belonging to the same theme. That leads to a lack of focus due to the fact there are too many themes to handle at the same time. In this post I will try to correct that.

On another note, it is not so easy to learn this endgame since the computer feels obliged to play perfect moves. A perfect move is the move that delays the mate the most. But humans can throw less optimal moves at you which you still have to know how to defuse them. Which can't be exercised against a computer with perfect play. The amount of moves to mate is quite irrelevant. What's important is that the methods you develop are easy to remember. No matter the amount of moves, as long it provides a secure way to the win, the method is good enough. Doing better than good enough is a waste of time and energy.

Before looking into the themes of exceptions, it is good to know the standard method towards the win. Look at the following diagram:

Black to move.
When you reach this "standard" position, the win is fairly straight forward once you know it. The white king is cut off. The pawn is already on the fifth rank, so white cannot give checks from the front since that has become the short side. The black king can escort the pawn towards g2 easy. At that moment you reached the Lucena position, where you have to build the well known bridge.

When I learned rook endings, the Lucena and the Philidor position were presented to me in a way that lead me believe that those positions were equally common in practice. But at CT the ratio is about 100:1. Another curious thing is that putting your rook behind your pawn is seldom a necessity in the standard position.

Ok, let's sum up the themes that prevent me from reaching the standard position:
  • Rook pawns need a totally different approach.
  • Sometimes there is a mate at the edge of the board. Both the attacker and the defender can fall victim.
  • Beware of skewers and pins. King against rook, king against pawn, rook against pawn.
  • The king of the attacker always needs a refuge. Behind his pawn, behind his rook, behind the defender's king, near the defender's rook.
  • Be aware of checks from the long side. From all four directions.
  • Prevent the defenders king and rook working together against the pawn. Cut off the king if you can. Horizontally or vertically.
  • Tactical trick: chase away the defenders king from his defense of his rook.
  • Beware of rook trades by the defender that leaves you with a drawn pawn ending (where you don't have the opposition).
  • Beware of an overloaded king, who has to defend both the rook and the pawn. In that case the defender can take the pawn.
Rook pawns.

Black to move.
The rook pawn requires a totally different approach than the standard position. I haven't worked that out yet. The problem is that your king can't shelter from checks at the leftside of your pawn. It is common to place your rook behind your pawn here. Often you will have to find moves that look counter intuitive. I will work it out later.

Cutting of the king.
White to move.
For the standard position it is necessary to  push the pawn to the fifth rank, in order to prevent checks from the front. In this position it is more important though that the black king is cut off. If the black king reaches the promotion square, it works together with his rook. if the pawn reaches b7, the pawn is attacked twice, and white has no spare tempo left to expel the black king from the promotion square since he has to defend his pawn. Besides that, the white king can't find shelter against checks. So Rd1 wins while b5 draws in this position.

Beware of the longside.

Black to move.
Allthough this position has a lot in common with the standard position, there is one problem though. White can give checks from the longside and black has no shelter. It is important to know when giving checks from the longside works and when it doesn't.

Longside checks not working
Black to move
Black can play c7, since he can shelter from the checks behind his rook and pawn. This is made possible by the position of the black rook. See following diagram.

Shelter is paramount
Black to move.
This position evolved from the previous one. The crucial factor is: can the black king find shelter against the checks? Here he can find shelter at h1.

Longside walk
Black to move.
Here can black save himself from the longside checks by walking towards the white rook. This is made possible by the black rook which defends the pawn.

Longside checks from the front
Black to move.
This would be a standard position if it wasn't for the fact that the pawn hasn't reached the fifth rank yet. This means that white can cause trouble by giving checks from the front. If black is to move then Rd3+ is the only move to win. All other moves draw. If it was white to move, it is a draw anyhow.

Forcing a drawn pawn ending

White to move.
White can force a draw by Re2!
When the rooks are exchanged, all there remains is a pawn ending where black hasn't the opposition, so he can't win.

White to move.
When a king is at the edge of the board, you always have to be aware of mate. The natural move g6 would lead to an unpleasant surprise for white here.

Overloaded king
Black to move.
If black wants to reach a standard position, he would like to play Re2 in order to cut of the white king. Rxf4+ then proves that the king is overloaded.

Pins and skewers
Black to move.
You have to be constanly aware of all possible skewers and pins. King against rook, king against pawn, rook against pawn. In the diagram above it is just irritating, since black can play Rh2. But sometimes it is moere than irritating.

Rook behind your pawn
Black to move.
Most of the time the standard position doesn't need a rook behind the pawn. The diagram above is an exception. The reason is that you need to protect your pawn when the checks from the longside come. The black can then walk towards the checking rook.

Black to move.
There is an important trick you have to know. Imagine in the above position that the black rook is at f7. In that case Kf1 is winning. If white plays Rxe2, then Rd7+ would conquer the white rook.
But in the current position, that trick doesn't work due to 1. ... Kf1 2.Rxe2 Rd5+ Ke4!
The black rook is too close to the white king.

The standard method is a simple way to escort a pawn to promotion. It needs the knowledge of the Lucena. There are quite a few themed problems that prevent you from playing a simple standard position. Most of those are fairly straight forward and need only a good awareness of the potential problems. Common sense does the rest. But there are two cases that are not so straight forward:
  • Rook pawns, which need a standard method on there own.
  • Checks from the longside, which need a close study of how to protect your king against them.
In all these positions the attacker can prevent the defending king from nestling on the promotion square. So far the only chances for the attacker in the case of a defending king on the promotion square lie in knowledge of the Philidor position and preventing the defender to be able to play for a third rank defense.

What is missing in CT is exercises for the defender to maintain the draw.


  1. "What is missing in CT is exercises for the defender to maintain the draw."

    You may "create" some. You can continue play even if you "lost" the position. Just change the position hide the move-display and play

  2. That's a good idea.
    How well do you score in KRpKR?

  3. My Endgame-rating at ct-theory is 1944 ( #12 if active ). But this rating is without watch. I can achieve 2000+ without big problems, its just geting too boring-slow.
    But i am one of the fastest in the top 20. Only ~6% Duplicates.
    I think the imediate feedback of the theory mode did help a lot.
    I dont play OTB that often but i am strong in "late endgames" compared to my opponents ( and i am weak in positional play, especially in "space", openings ... )
    I was only solving problems and when i had a few times "nasty ones" then i was looking into some endgamebooks.