Friday, June 13, 2014

Rook Endings 1: Rook versus Pawn

Rook endings are among the most common endings and playing them well is one of the most important and subtle endgame skills. To improve my chess (and hopefully yours as well) I am beginning a series of blog postings on rook endings. We are going to begin with the most basic rook ending: rook versus a lone pawn. These seemingly simple endings can actually be quite tricky and are mishandled more often then you might expect. Since the beginning of 2012 through mid-May 2014, 523 games in Chessbase's mega database reached a rook versus pawn ending that was played out for four or move moves. 47 or 9.2% of those 523 games did not reach the result expected by the tablebases at the moment the rook versus pawn ending began. To be sure, many of the poorly handled endings were by weak players who made silly mistakes, but not all of them. Even grandmasters erred. Clearly, this endgame is not trivial. Its key ideas need to be mastered. They are the following:

Attacking ideas:

1. Cutting the enemy king off at his fourth rank.
2. Gaining or losing tempi with rook moves.
3. Avoiding stalemate.
4. Getting the king in the game.
5. Moving the king to the side of the pawn opposite the enemy king.

Defending ideas:

1. Shouldering away the enemy king.
2. Promoting to a knight.

Let's look at them one by one, and then we will examine a few cases in which more than one idea is used.

First Attacking Idea: Cutting off the enemy king at his fourth rank.

Fig 1. Black to Play and Win!

Here black's idea is simple. He plays 1...Rb4, prevent white's king from entering the fourth rank. If white tries to promote his pawn by pushing it, black will win it. Thus, 1...Rb4 2.g6 Rb6 3. g7 Rg6 4. g8=Q Rxg8 wins for black. If white shuffles his king back and forth instead, then black's king will advance and take the pawn

Second Attacking Idea: Gaining or Losing Tempi with Rook Moves.

Fig 2a. White to Play and Win!

Here, if white plays 1. Rh8 he can't get his king in the game fast enough to prevent black from promoting and drawing, but if he starts with 1.Rg8+ Kf4 2.Rh8 Kg5 he gains two tempi and will consequently win the game. Best play would be 3.Kb7 h5 4.Kc6 h4 5.Kd5 Kg4 6.Ke4 Kg3 7.Ke3 h3 8.Rg8+ Kh2 9.Kf2 Kh1 at which point we reach the third attacking idea (fig 3a below). For another example see Wenjun Ju (2528) - Nino Maisuradze (2284) 0.5-0.5, in which Ju threw away half a point by missing this idea.

In fig 2b, white wants to lose a tempo so as to gain the opposition and move his king close to black's pawn.

Fig 2b. White to Move and Win!

White plays 1.Rb1. After 1...Kc4 2.Ka6 b3 3.Ka5 Kc3 4.Ka4 b2 5.Ka3 black's pawn is doomed.

Third Attacking Idea: Avoiding Stalemate.

Fig 3a. White to Play and Win!

Here white must be sure to avoid stalemate. There are two ways to do this. One is to move the rook to the a, b, c, d, or e file, threatening mate if black plays h2. The other is to play 1.Rg1+. Either way the king is forced to h2 at which point the rook moves to the third rank, and after 2...Kh1, 3.Rxh3#

A key position in which avoiding stalemate is the issue is the following:
Fig 3b. White to Move and Win!

Here white plays 1.Kb3 (avoiding stalemate by blocking the Rook's attack on b1) Kb1 2.Ka3+ Ka1 3.Rh8 Kb1 4.Rh1+ Kc2 5.Kxa2. This move was missed twice in recent play. For the examples see here and here. In both cases the attacking side thought it necessary to move the rook over so that it can attack from the side. In this case, if that is done then black plays ...Kb2 and when white checks from the side, he plays ...Ka3 securing the draw as per fig 10 below.

Knight pawns can also pose stalemate problems as in the following example, which appeared in a game in 2012:
Fig 3c. White to Play and Win!

Here white played 62.Rc2?, which drew after 62...Ka1 63.Rxc2 stalemate. Instead, white needed to lose a tempo so as to get his rook on the other side of the pawn. Thus, 62.Rc8 (or Rc7, Rc5, Rc4) Ka1 63.Ra8+ Kb1 64.Ra2 Kc1 65.Rxb2 and white wins easily.

Fourth Attacking Idea: Getting the King in the Game.

This may not sound too sophisticated--after all the king and rook together are needed to win--but my database analysis shows that it is often forgotten. The common error seems to be one of thinking that putting the rook behind the pawn should be the priority. Although the rook often belongs there, the more urgent need in most rook versus pawn endings is to move the king into position.

The following position appeared in the 2013 game Ngoc Truong Son Nguyen (2625) - Dmitry Andreikin (2713) 0.5-0.5.

Fig 4. Black to Play and Win!

Andreikin threw away the win with 71...Rc1? when 71...Kc5 wins easily. (In his defense, it should be said that this was a blitz game.) For 2013 game in which a GM missed this simple idea see Zviad Izoria (2585) - Denis Shmeliov (2431) 0.5-0.5.

Fifth Attacking Idea: Moving the King to the Side of the Pawn Opposite the Enemy King.
Fig 5. White to Play and Win!

Here white wins with 1.Kf6, the idea being that he needs to approach the pawn in order to take it in coordination with the rook. He can't go through the black king, so he has to go around it. Without playing out the moves, we can see that white will get his king to g2 in 5 tempi and black will also need 5 tempi to get his king to e2 and pawn to f1, promoting. So, since white goes first he wins.

This attacking idea is related to our first defending idea, which focuses on keeping the attacking king at bay.

First Defending Idea: Shouldering Away the Enemy King.

This is common idea in pure pawn endings, but it is no less important when trying to draw in a pawn versus rook ending. Here is an interesting example:

Fig 6. Black to Play and Draw!

Here the counterintuitive 1...Kg5! is the saving move. Otherwise, white's king will be able to get onto the g-file and create mating threats with the rook that will cause black to lose. The analysis is a bit more complicated than the prior examples. You can find it here.
Second Defending Idea: Promoting to a Knight.
Fig 7. Black to Play and Draw!

Here if 1...c1=Q? is played, 2. Rh1# wins. Instead, 1...c1=N+ draws as long as black remembers the first principle of drawing a lone rook versus lone knight position: keep your knight as close to your king as possible. In two of the 47 games mentioned above the error by the defending side came in the rook versus knight ending. Those games are here and here.
Combining Ideas.
The following position combines the idea of cutting off the defending king and preventing the shouldering out of one's own king.
Fig 8. White to Play and Win!

Here white must play 1.Rg5 to win. This prevents black from playing Ke5, which would shoulder out the white king, and requires black to take the long path (d6-c6-b5) around his pawn. Note that black can't play 1...c4 as that would be cutting of his king at the fourth rank.
In the following, black draws by a combination of shouldering and promotion to a knight.
Fig 9. Black to Play and Draw

Black must play 1...Kc3 to draw as this allows him to shoulder out the white king: 2.Ke5 Kd3 3.Rh8 c3 4.Rh3+ Kd2 5.Kd4 c2 6.Rh2+ Kd1 7.Kd3. Here, we reach the position in fig 7 above, and 7...c1=N+ draws.
The following position shows the importance of concrete calculation, which is always needed to determine whether an idea should be used. Playing one of the ideas discussed above loses while playing the other draws.
Fig 10. Black to Play and Draw!
Here black might be tempted to play 1...Kb1?, thinking that if white threatens checkmate with 2.Kb3 then he saves himself with 2...a1=N+. But with a rook pawn this no longer works because after 3. Kc3 black will lose his knight. So, instead black must play 1...Ka3, which shoulders out the white king. White can't approach the pawn to take it and the best white can do is repeat moves by playing 2.Rh1 or Rh3+ either of which will be followed up by 2...Kb2.
Another example of how general ideas are not enough is the following:
Fig 11. Black to Play and Draw!

In this game played in 2012 black had the right idea, but at the wrong time. He played 49...Kf4?, with the idea of shouldering off the white king as in fig 9, but this loses due to 50. Kb6 Ke4 51.Kc5 f4 52.Kc4 Ke3 53.Kc3 f3 54.Re6+ when black resigned. Instead, 49...f4 50.Kb6 Kf3 51. Kc5 Ke3 and black has achieved the shouldering operation, but is a tempo ahead over the game and consequently can draw. 
Well, there you have all of the basic ideas you need to play almost all rook versus lone pawn endings well. There are a few complicated and special cases still left to consider. They will be covered in the next blog posting.


  1. This is really great stuff. Written well. Most of this I know and all of this I have seen, but is still a great refresher of how complicated this apparently simple ending can become.

  2. Kassy, Glad you find it interesting. Stay tuned for more as I work my way through increasingly complicated rook endings.