While studying the Rook vs Pawns endgame I noticed that in Averbakh's textbook there is a large number of winning and drawing zones in the chapter on rook vs one pawn. The squares of the rook and pawn as well as the square for the king of the pawn side are fixed. The king of the rook party is variable, and so it is his zone that is drawn (as in the three images above).

The corresponding diagrams are well suited to studying this endgame without a chessboard. And they provide great use cases for the Winning Zone tool, too. Within the tool each position can be treated individually, or, since the tool can handle more than one varying piece, many or even all positions can be treated at once. And since we're in tablebase territory, the calculations take as long as a finger snap!

Of course, that's all great. But why does this page exist? Well, some of Averbakh's results don't match those of the Chess Suite tool. So what to do with the corrections to Averbakh's book? Where does systematic basic endgame theory take place nowadays?

Above I spoke on Averbakh's textbook. To be precise, I meant Juri Awerbach, Turmendspiele Band 1, Sportverlag (Berlin, 1986). However, since I assume that my readers hardly have this German edition, I will show pictures of the Russian and English editions below. The Russian edition is Авербах, Шахматные окончания, ладейные (1984). The English edition is Yuri Averbakh, Comprehensive Chess Endings - Volume 5: Rook Endings (1987). Only the German edition is divided into two volumes (Band 1 and 2).

We start with Diagram 40 (luckily the numbers are the same in all editions.). White to move.

On the left you can see the diagram from the Russian edition, in the middle that from the English version and on the right the diagram of the tool. The zones apply to the white king, so his placement is irrelevant. And look: all zones are different! (The German edition equals the Russian.)

Averbakh only gave variants for a few cases. One of them for the king on d3 (Russian and German edition): 1. Rh2+ Kb3 2. Rh8 a2 etc. But here Black's second move is a mistake. Instead, 2.. Kb2 draws. It seems that Kenneth P. Neat, the translator of the English edition, spotted the mistake, omitted the variation from the text, moved the king to c5 and added the square d3 to the drawing zone. If he had also checked the d2-square, he would most likely have obtained the same result as the tool!

[Addendum (January 19, 2023): Michael D. Sharpe (Toronto, Canada) kindly informed me that this example is corrected in the electronic version of Comprehensive Chess Endings (Convekta, 2003). He also sent Averbakh's foreword, from which one can learn that this version has been greatly revised and improved.]

Next is Diagram 42 and again White has the move.

Again the picture on the left is from the Russian edition, the one in the middle from the English and the one on the right was generated by the tool. The pictures from the Russian and English edition are equal this time (and the German edition shows the same zone, too), but the tool claims that a6 and b6 don't belong to the drawing zone. Hence White wins, when the white king is on one of these squares. In both cases 1.Rb1+ is the unique winning move and this is worth analyzing, especially compared to the position where the white king is on c6. Enjoy!

The final diagram is no. 53 and this time it is three times Black: Black has the rook, Black to move and the zone is for the Black king.

Please excuse the bad picture on the left, but with the other two and the order you are used to now, you can see for sure what is meant. (Once again the German edition shows the same zone as the Russian and the English editions.)

The Winning zone tool adds three more squares to the drawing zone: f4, f6 and e8! The most obvious is probably e8, because 1.. Rb6+ 2. Kc5 Kd7 draws. In fact, 1.. Rc5 and 1.. Rd5+ draw, too!

If the Black king is on f6, it is as also quite easy: 1.. Rb1, 1.. Rb2 or 1.. Rb3 all draw, because 2. c8Q Rd1(2/3)+ and then a check on the c-file wins the queen for the rook.

Finally, if the Black king is on f4, then 1.. Rb6+ has to be played. Now the c-file and the seventh rank are a no-no. So, the game continues 2. Kd5 Rb5+ 2. Kd4 Rb1! and the same motif as above applies.

(You might be wondering for a moment, why doesn't f5 belong to the drawing zone?! Since a queen on c8 gives check.)

Addendum January 8th, 2023

I've decided that from now on I'll check every winning or drawing zone I come across in books and collect the improvements here.

There are a number of zones in the chapter K+P vs K+P in Averbakh's book on pawn endings. They are all correct with one exception.

 [Addendum (January 19, 2023): Another valued contribution from Michael D. Sharpe: The source is Deutsche Schachblätter 6/1948, pp. 86-87 and there are eight examples of win/draw zones attached to an obituary for Leick.

At that time, Leick was writing a book about pawn endgames and for this purpose he designed such summary tasks. This is his example VI, and his solution is: White wins if he can capture the h6-pawn in four moves or if he can move to c2 on the first move. So he had the correct zone!]

Averbakh's book on Knight vs Bishop has one chapter (N+P v B) that is full of zones. They are mostly attributed to Averbakh, 1958 and they are mostly correct. Only one series of diagrams contains some errors. It seems the grandmaster had a bad day. The numbers of the diagrams are 83, 84 and 85 in the German edition from 1989 (the numbers in the Russian edition of 1958 are different).

The chapter Q vs 2 connected pawns in Averbakh's book on Queen endings contains 16 winning zones for the white king. Two of them (both without any attribution) are wrong. The number of the diagrams are 20 and 22 in the German edition from 1990.

Strictly speaking, diagrams 25 and 30 are also wrong. They differ by one square on which the white king would be in check. These are the square a1 and c1, resp. in the following diagrams. The very last one is attributed to N. Grigorjew (published posthumously in 1940).

In the chapter Q+P vs Q of the same book there are two zones (one for the black king and the other one for the white king), both attributed to Averbakh 1962 and both seem to be incorrect. The diagram numbers are 101 and 113.

The next example is from the chapter Q vs R+P. The zone in question is attributed to W. Chenkin, 1962 and the diagram number is 328, where the square h4 is missing. (There are more suspicious zones in this chapter, but there I don't even understand the conditions.)