Basic Rook Endgames

We lost on board count to old rivals Teignmouth in the Peter Rooke Cup
at the end of January. One of the last games to finish involved a
simple-looking Rook endgame. Just as it was getting crucial, White,
short of time, found a time-saving blunder:

[Event "Exeter vs Teignmouth"]
[Site "?"]
[Date "2013.01.26"]
[Round "3"]
[White "Gorodi"]
[Black "Pope"]
[Result "0-1"]
[SetUp "1"]
[FEN "8/8/8/8/4k1R1/4p3/r7/5K2 b - - 0 1"]
[PlyCount "1"]

1...Kf3 0-1

So what should the result have been? I discussed it briefly with my
friend Bill Ingham. As is often the way, he fancied his side's chances
to win, and I hoped for a draw, but the discussion showed how vague we
all are about some basic theory. To be fair, some apparently small
differences can change the best obtainable result:

[Event "RPvR"]
[Site "?"]
[Date "2013.02.05"]
[Round "?"]
[White "NN"]
[Black "NN"]
[Result "1/2-1/2"]
[SetUp "1"]
[FEN "2r5/8/4k3/8/2P5/1K6/8/3R4 w - - 0 1"]
[PlyCount "0"]

{White to move: draw.} 1/2-1/2

[Event "RPvR"]
[Site "?"]
[Date "2013.02.05"]
[Round "?"]
[White "NN"]
[Black "NN"]
[Result "1-0"]
[SetUp "1"]
[FEN "2r5/4k3/8/8/2P5/1K6/8/3R4 w - - 0 1"]
[PlyCount "0"]

{White to move: win.} 1-0

So, what was the right result for the Teignmouth game?

Before I answer that, I want to decide: regardless of the right result,
what is the right approach?

You probably know the Philidor and Lucena positions. Philidor showed a
draw, with the defending King having control of the Queening square. The
defender keeps out the attacking King. White can make progress only by
pushing the pawn, the Black shows a draw by sending the Rook a long way
away and playing a flurry of checks.

[Event "rook ending: basic draw"]
[Site "rook ending: basic draw (Phil"]
[Date "1995.??.??"]
[Round "?"]
[White "Philidor"]
[Black "Basic Draw"]
[Result "1/2-1/2"]
[SetUp "1"]
[FEN "4k3/R7/1r6/4P3/4K3/8/8/8 w - - 0 1"]
[PlyCount "13"]

{ Black K has control of the queening square and bR has control of the sixth}
1. Kf5 1... Rc6 {
now there is no shelter on the board from the checks for the White King.} 2. e6
Rc1 3. Kf6 Rf1+ 4. Ke5 Re1+ 5. Kd4 Rd1+ 6. Ke4 Re1+ 7. Kd5 1/2-1/2

The 'Lucena' position, shown by Salvio and attributed to Sciapino, is a
win for the attacking side. With the defending King cut off by at least
two files, the defending Rook has to play the flurry of checks from
across the board. The attacker either makes an escape hatch for the
King, or 'builds a bridge' so that the Rook can step in at the right

[Event "rook ending: basic win (Lucena)"]
[Site "?"]
[Date "2008.??.??"]
[Round "?"]
[White "Lucena"]
[Black "Basic win"]
[Result "1-0"]
[SetUp "1"]
[FEN "2k5/8/4K3/4P3/8/8/3R4/5r2 w - - 0 1"]
[PlyCount "25"]
[EventDate "2008.??.??"]

{ White has control of the sixth/seventh with the K and the bK is cut off} 1.
Ke7 1... Kc7 2. Rc2+ Kb6 3. e6 Rd1 4. Ke8 Rg1 5. e7 5... Rf1 {} 6. Rc4 {
the critical manoeuvre: 'BUILDING A BRIDGE'} (6. Rh2 {although the 'bridge'
manoeuvre is always emphasised in the books, there is also a win by getting
the king out on the short side, e.g.} 6... Kc7 7. Rh7 Rf2 8. Rf7 Rh2 9. Rf8 Rh7
(9... Re2 10. Kf7) 10. Rf1 Rh8+ 11. Kf7 Rh7+ 12. Kf8 Rh8+ 13. Kg7 Re8 14. Kf7 {
This is important because the bridge-building manoeuvre may not be available,
and vice versa.}) 6... Rf2 7. Kd7 Rd2+ 8. Ke6 Re2+ 9. Kd6 Re3 10. Rb4+ Ka7 (
10... Ka5 11. Rb8) 11. Rb5 Rd3+ 12. Ke6 Re3+ 13. Re5 1-0

Please note: sending the Rook a long way away. If the defending Rook is
too close to the attacking King, the King will be able to chase the Rook
away all by itself.

This is the idea of checking distance, and will help us understand less
clear-cut positions.

We all know that the Rook belongs behind the pawn, but that's not true
for every position. For example, in this one, the defending Rook has
checking distance only in front of the pawn.

[Event "?"]
[Site "?"]
[Date "1923.??.??"]
[Round "?"]
[White "Cheron, Andre"]
[Black "?"]
[Result "1/2-1/2"]
[SetUp "1"]
[FEN "2r5/8/4k3/8/2P5/2K5/8/3R4 w - - 0 1"]
[PlyCount "12"]
[Source "ChessCafe/CB"]
[SourceDate "2003.10.29"]

{This is a typical case of an easy draw due to a frontal attack.} 1. Kb4 (1.
Rd4 Ke5) 1... Rb8+ 2. Ka5 Rc8 (2... Ra8+ 3. Kb6) 3. Kb5 Rb8+ 4. Ka6 Rc8 5. Rd4
Ke5 6. Rh4 Kd6 1/2-1/2

How does this help us understand the Teignmouth position? Well, the
defending King doesn't have access to the Queening square, but there may
be checking distance from the side.

[Event "Rook ending: checking distance"]
[Site "?"]
[Date "????.??.??"]
[Round "?"]
[White "?"]
[Black "?"]
[Result "1/2-1/2"]
[SetUp "1"]
[FEN "4K3/4P1k1/8/8/8/8/r7/5R2 b - - 0 1"]
[PlyCount "11"]

{} 1... Ra8+ 2. Kd7 Ra7+ 3. Kd6 Ra6+ 4. Kd5 Ra5+ 5. Kc6 Ra6+ 6. Kb7 Re6

Eagle-eyed readers will have spotted that I cheated: I swapped around
the Rooks, giving the defender one more file's-worth of checking
distance. In a position like the actual game, with the defending Rook on
the b-file, it's not quite enough:

[Event "Rook ending: checking distance not OK"]
[Site "?"]
[Date "????.??.??"]
[Round "?"]
[White "NN"]
[Black "NN"]
[Result "1-0"]
[SetUp "1"]
[FEN "4K3/4P1k1/8/8/8/8/1r6/5R2 b - - 0 1"]
[PlyCount "8"]

{} 1... Rb8+ 2. Kd7 Rb7+ 3. Kd6 Rb8 (3... Rb6+ 4. Kc7 Re6 5. Kd7) 4. Kc7 Ra8
5. Ra1 $1 {Obtaining a Queen and Rook ending: q.v.} 1-0

Now, when Bill and I were playing around with this position, I managed
to persuade him that it might be a draw, because he couldn't escape the
flurry of checks, because he didn't approach the Rook with his King. I
didn't know the theoretical result, but I did know the best thing to

It's very well worth your while to play out these positions against a
friend or a computer, so that if and when it turns up in one of your
games, you won't hesitate to adopt the right approach.

A very useful online resource is this implementation of the Nalimov
tablebases: which gives
the theoretical best obtainable result from any given simple position.

Chess Quotes

"No fool can play chess, and only fools do."
"You may knock your opponent down with the chessboard, but that does not prove that you are the better player."