Go Wizardry

All About the Many Aspects of Go
We have millions of friends around the world... and they all play go!

Basic Techniques of the Endgame

For Those Aiming to Become Shodan


Takagawa Shūkaku 9 dan

“Takagawa’s Fuseki” [the series by the same author that appeared the previous year in Kidō] attracted both good and bad reviews, but regardless of that, this time “the endgame” will be the subject to be studied for the year.

The request of the editor was that complex calculations be avoided to the utmost, so let’s study in conformance to that desire. At the same time, my intention is explain things as simply as possible to make this a course for lower and middle kyū players. For that reason, dan level players are best advised to ignore this material.

A Small Difference is a Big Difference

In go, no matter what aspect of the game is considered it may be true, but in the endgame even the very slightest things require close attention. In order to do that, one does not have to read out dozens of moves ahead. When speaking of the number of moves, if one can focus one’s attention one to three moves ahead, a world that one has not been able to see up to then will arise. By playing a subtle forcing move, a couple of points of profit can be suddenly realized, while it is not rare for a minor tesuji that one had not known about to cause a loss of three or even four points. If two or three points of profit or loss are assembled here and there into one large collection, all at once a difference of ten or fifteen points may be opened up.

This time, we will conduct a small investigation into a number of those kinds of “trivial differences” and look at them closely.

For amateur players, the endgame is a difficult challenge. If those around the reader are that way and none are well-versed in it, should the reader become adept at endgame play, the reader is sure to be respected by all as a unique player.


Problem A: Black to Play

White is alive. In what way should Black finish up play here? There does not seem to be much to this matter, but since there is a difference between playing at A or B, it will not do to be sloppy about it.


Diagram 1 A 1 Point Half Play Loss

The impulse to put two stones into atari can psychologically impel one to impulsively play Black 1. Since it makes no sense to let White’s whole group die, White 2 is essential. Well then, when this happens, it becomes clear that Black 1 is a failure. In regards to White’s territory, the move of White A remains as a possible 1 point extra for White. The fact of this extra 1 point possibility makes it a half play loss for Black. If, at the end of the game, White is allowed to play at A and a 1 point loss is incurred, one would like to cry.


Diagram 2 Correct Solution

In response to Black 1, White 2 is essential. The fact that Black 1 is superior to A is obvious to anyone. It is a trivial matter, no?


Problem B: Black to Play

When fixing the shape in the end, is the wedging-in move of A correct, or is it best to push in at B? Perhaps it may be declared that either one would be okay, is that not so? However, it cannot be said that A and B are the same.


Diagram 1 If Pushing In

If Black pushes in at 1, White replies at 2. This finishes everything. It does not seem like there is much of anything going on here.


Diagram 2 A Difference in Kō Threats

However, the way with the wedging-in move of Black 1 is superior. If White defends with 2, one might think that it is the same as in the previous diagram, without any difference, but there are many kō threats here, starting with A, that are effective. One must not scoff that these are merely kō threats. One never knows how big such a minor difference like this might become. When one analyzes one move in comparison to another like this, and realizes that seemingly trivial matters are actually important, one will become perceptibly stronger.


Problem C: Black to Play

With the shape just this way, if White were to hane at A and then connect, then White would do so in sente [after Black B, White C and Black D]. Now it is Black’s move, so what is the best way to play?


Diagram 1 Mediocre Play

Even though it is undesirable to let White hane and connect, if one does not perceive any value in playing with particular care, one might end up just making the hane of Black 1 and connecting at 3. This is not usually good.


Diagram 2 Gote Becomes Sente [Gote no Sente — a technical go term]
White plays 2 elsewhere

Black should stoically descend at 1. This is the Correct Solution.

Supposing that White blocks at 7, Black can take great satisfaction in eliminating the possibility of White’s hane and connection. Consequently, White ends up playing elsewhere, but then Black 3 is tesuji. The sequence through Black 7 makes a big dent in White’s territory, and Black is left with the one stone capture at A.

At the start, had Black ended up exchanging A for White 3, the technique [suji] of Black 1 would have been rendered impossible.


Problem D

Here, White hanes at 1 and connects at 3. There are now two ways for Black to connects, at A or B. Usually, if there is no potential for trouble [bad aji], a solid connection at A is best so that the possibility of kō threats is eliminated, but in this case which way is best?


Diagram 1 Getting Forced

Let’s see what happens if Black connects at 1. In that case, Black incurs the immediate forcing moves of the hane and connection of White 2 and 4. That is the reason why Black is dissatisfied with the move of Black 1 here.


Diagram 2 Effective

Here, the hanging connection of Black 1 is the Correct Solution. By doing so, White 2 and 4 do not become sente. That is because Black 1 works effectively to make it unnecessary to connect at the cutting point of A. Therefore, since White 2 and 4 end in gote, it is reasonable to view the eventual disposition of the play to be White 4 and Black’s descending at 2. In that case, Black’s territory is 13 points. In Diagram 1, Black’s territory was 11 points.

According to the way of connecting, a 2 point difference is produced. In a game where the winning margin is 2 points, choosing Diagram 1 leads to a 1 point loss.


Diagram 3 The Opposite Hane and Connection

When it comes to the way with White 1 and 3, here Black 4 is the Correct Solution. The fact that the move of White A is rendered ineffective is the same as in Diagram 2. When one is a beginner, after White plays the hane of 1 and connects at 3, the conception of Black 4 is something that would never come to mind.


Diagram 4 A 2 Point Loss

Should Black thoughtlessly connect with 1, White plays 2 and 4, and the position ends up reverting to Diagram 1.


Diagram 5 If One Line Wider…

The Black territory is one line wider here, and with this shape, how is the move of Black 1 to be evaluated? This is not possible. Through White 10, Black ends up being destroyed.

In this case, had White first haned at A and after Black B, connected at C, then the hane and connection with the marked White stones could not be made in sente. It would not be effective, so attention by White is required.

When hane and connection on two sides is possible, one must keep in mind techniques for defending on both sides simultaneously. A 2 point difference can immediately be produced.


Problem E: Black to Play

If things are left as they are, White will hane at A. When point values are reduced around the board, this becomes big.

Well then, how should Black play here? In this kind of position, according to the way that one plays, an unexpected difference can be produced.


Diagram 1 Mediocre

If one thoughtlessly looks over the situation, one might end up making the hane of Black 1 and connection of 3. This might be characterized as an unsatisfactory endgame play that overlooks the cutting point at A.


Diagram 2 Descending
White plays 2 elsewhere

With this shape, Black must descend at 1. Should White defend by blocking at 5, the hane of White 1 has been eliminated, which may be said to be the profit in Black’s way of playing.

If White plays 2 elsewhere, Black has the endgame play of 3 available. Should White try to intercept with a move at 5, Black cuts at A, and White is destroyed. In the end, Play proceeds to White 6, and although Black ends up with gote in the same way as in Diagram 1, in comparison a 2 point difference in White’s territory is produced.


Problem F: Black to Play

This shape often arises. How should Black play? In this place where nothing special is going on, there is scope for displaying technical skill.


Diagram 1 Crude Play

When one is a beginner, one makes the pincer attachment of Black 1. The intention is surely to reduce White’s territory in the sequence through Black 7. But this is not good.


Diagram 2 A Simple Hane

Black simply makes the hane at 1. Next, the shape is fixed with Black 3 and White 4, making White’s territory here 7 points. In Diagram 1, the territory is 6 points so the way in Diagram 1 seems more profitable, but in Diagram 1, Black ends in gote, while in Diagram 2, Black ends in sente. For a tiny 1 point of profit, one cannot take gote. In Diagram 1, Black 1 is a 1 point gote move. In a position where play has come to a standstill (meaning that there are no other places to play), then playing as in Diagram 1 is fine, but in typical cases this should not be considered at all. To all beginners out there: Please graduate from the stage of playing as in Diagram 1.


Problem G: Black to Play

When there is no potential [aji] for making the placement of A, how should endgame play be directed against White’s territory? This problem is connected with Problem F, and it is not particularly difficult.


Diagram 1 A Beginner’s Mistake

Here, too, it seems that many people would make the pincer attachment of Black 1. Those of middle kyū strength and above would not play this way, but while one is a beginner it appears that this method is attractive. In the same way as is the case in Problem F, of course this way of playing is impossible.


Diagram 2 Commonplace Play is Fine

The usual hane of Black 1 is good.

After the sequence through White 6, Black gets sente. Please note the point that in Diagram 1, Black ends in gote.

It is not just a matter of profit and loss in the local context, but that maintaining sente to turn elsewhere is vital in terms of whole board awareness. In addition, should White play 2 at 6, Black responds with a jumping attachment of 4, saddling White with a greater point loss.


Problem H: White to Play

Here is a variation of the shape in Problem G.

When the shape is such that the marked White stone is added, Black hanes at 1. This time, please consider the position from White’s standpoint. Perhaps this is a little difficult for middle kyū players, not to mention beginners, but here White has an important tesuji.


Diagram 1 Commonplace

The commonplace way to play is with White 1 to 5. This is an ordinary way of playing, but it is insufficient. Since this does not make effective use of the marked White stone, it is mistaken from the standpoint of endgame play.


Diagram 2 Tesuji

White 1 is the Correct Solution. At this point, it is often seen that both sides will refrain from further play here, leaving the situation as it is while turning to make endgame plays elsewhere.

When asking why White 1 is good, there is a rationale behind it.


Diagram 3 After That

Following Diagram 2, if Black were to play, it would be with the moves from Black 1 to 5. Please note that White’s marked stone works effectively, so that White can take sente. Whether Black plays 1 at 4 or extends at 3, the result is the same.


Diagram 4 Profitable for White

If White were to play, it would be to block at 1.

Whether Diagram 3 or Diagram 4 is played is a 50-50 proposition, but from White’s perspective, even when Diagram 3 is the result, White’s territory has only been reduced 1 point more than in Diagram 1, proving the value of the tesuji of White 1 in Diagram 2.

A jumble of ideas have been presented this time, but if it has been conveyed that it is important to focus one’s attention even in positions where nothing special seems to be happening, that is sufficient.

In future articles, artistry in taking sente, endgame play in connection with life and death considerations, conditions affecting the monkey jump, large size endgame plays at the beginning of the endgame, etc., will be studied. Starting with the fifth installment, systematic analysis of various endgame tesuji will be detailed.

According to a certain amateur player, those specializing in the fuseki are flashy, those who like to fight are barbaric, those who concentrate on tesuji are namby-pamby stylists, those who like life and death problems are pedantic, and those who are skilled at endgame are monomaniacal, or some such critique. In go, it is quite fine to be monomaniacal, with eyes wide open, seeking out profit and collecting it.

Tagged as: , , ,

Leave a Reply


book cover

Go on the Go Collection: Volume I

Three booklets have been assembled into the collection here.

Buy this Book at Amazon

Go For Everyone

Go For Everyone

A New Method for Learning to Play the Game of Go

Buy this book

Book Cover

Journey to the West

This is a semi-autobiographical novel that depicts a unique American success story; a rags to riches tale of a man escaping his humble origins to make millions of dollars, but then he throws it all away due to the ancient character flaw of hubris.

Buy this Book at Amazon